mirrors/llvm-project
0
0
Fork 0
mirror of https://github.com/llvm/llvm-project.git synced 2025-04-16 17:46:40 +00:00
Code Issues Projects Releases Wiki Activity
llvm-project/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

12094 lines
300 KiB
Python
Raw Normal View History

Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
# This file is licensed under the Apache License v2.0 with LLVM Exceptions.
# See https://llvm.org/LICENSE.txt for license information.
# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
# Description:
# The MLIR "Multi-Level Intermediate Representation" Compiler Infrastructure
Fix bazel build for https://reviews.llvm.org/D144552 Differential Revision: https://reviews.llvm.org/D151346
2023-05-24 09:35:53 -07:00
load("@bazel_skylib//rules:expand_template.bzl", "expand_template")
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
load("@bazel_skylib//rules:write_file.bzl", "write_file")
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
load(":tblgen.bzl", "gentbl_cc_library", "td_library")
load(":linalggen.bzl", "genlinalg")
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
load(
":build_defs.bzl",
"cc_headers_only",
"if_cuda_available",
"mlir_c_api_cc_library",
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
package(
default_visibility = ["//visibility:public"],
[Bazel] Enable layering_check for MLIR build This feature checks that headers included by a file are provided by a header exported by one of the direct dependencies of the build rule in which it is contained. It ensures that appropriate layering (a goal of the LLVM project) is preserved. So far, I'm only adding this to MLIR because we've had it turned on internally since the beginning, so MLIR is already layering clean. It would be nice to also enable it for LLVM, but that requires some additional cleanup. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D113952
2021-11-15 15:31:08 -08:00
features = ["layering_check"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
[bazel] NFC: Move licenses declaration from package to function. The `licences` attribute is deprecated, see https://docs.bazel.build/versions/4.0.0/be/common-definitions.html#common-attributes.
2022-09-05 10:43:48 +02:00
licenses(["notice"])
[MLIR] NFC: add back exports_files(["run_lit.sh"]).
2022-09-08 07:46:10 +02:00
exports_files([
"LICENSE.TXT",
"run_lit.sh",
[NFC][bazel] Add export_files for utils/textmate/mlir.json
2023-01-17 17:16:36 -08:00
"utils/textmate/mlir.json",
[MLIR] NFC: add back exports_files(["run_lit.sh"]).
2022-09-08 07:46:10 +02:00
])
[MLIR] NFC: add back `exports_files(["LICENSE.txt"])`.
2022-09-07 21:57:02 +02:00
Fix bazel build for https://reviews.llvm.org/D144552 Differential Revision: https://reviews.llvm.org/D151346
2023-05-24 09:35:53 -07:00
expand_template(
name = "mlir_config_h_gen",
out = "include/mlir/Config/mlir-config.h",
substitutions = {
[MLIR] Reformat the Bazel build This patch normalizes formatting of the the root BUILD.bazel file by: 1) adjusting indentation a little bit, 2) alphabetically ordering dependencies. These small deviations were introduced by some yesterday's patches: * https://reviews.llvm.org/D151104 * https://reviews.llvm.org/D151346 * https://reviews.llvm.org/rG16fe2b37365c00b0c6d0ed22c2e6521f2d5de01a * https://reviews.llvm.org/rG4d1cd1d8caab13d6b76ce6fc4ff76a01a7931c34 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D151499
2023-05-26 14:16:47 -07:00
"#cmakedefine01 MLIR_ENABLE_EXPENSIVE_PATTERN_API_CHECKS": "#define MLIR_ENABLE_EXPENSIVE_PATTERN_API_CHECKS 0",
[mlir][bazel] Follow-up fix for ce954e1cda5c9b55325903d51285cd742152a0c3.
2023-05-31 08:19:26 +00:00
"#cmakedefine MLIR_GREEDY_REWRITE_RANDOMIZER_SEED ${MLIR_GREEDY_REWRITE_RANDOMIZER_SEED}": "/* #undef MLIR_GREEDY_REWRITE_RANDOMIZER_SEED */",
[MLIR] Reformat the Bazel build This patch normalizes formatting of the the root BUILD.bazel file by: 1) adjusting indentation a little bit, 2) alphabetically ordering dependencies. These small deviations were introduced by some yesterday's patches: * https://reviews.llvm.org/D151104 * https://reviews.llvm.org/D151346 * https://reviews.llvm.org/rG16fe2b37365c00b0c6d0ed22c2e6521f2d5de01a * https://reviews.llvm.org/rG4d1cd1d8caab13d6b76ce6fc4ff76a01a7931c34 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D151499
2023-05-26 14:16:47 -07:00
},
Fix bazel build for https://reviews.llvm.org/D144552 Differential Revision: https://reviews.llvm.org/D151346
2023-05-24 09:35:53 -07:00
template = "include/mlir/Config/mlir-config.h.cmake",
)
cc_library(
name = "config",
hdrs = [
"include/mlir/Config/mlir-config.h",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
filegroup(
name = "c_headers",
srcs = glob(["include/mlir-c/**/*"]), # <== i.e. match the entire tree
)
[Bazel] Uniformly export all MLIR td files CMake would have no restrictions on this and the custom list is a pain to maintain. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D106003
2021-07-14 12:09:41 -07:00
exports_files(glob(["include/**/*.td"]))
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
[
gentbl_cc_library(
name = name + "IncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/IR/" + name + ".h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/IR/" + name + ".cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/" + name + ".td",
deps = [":OpBaseTdFiles"],
)
for name in [
"OpAsmInterface",
"RegionKindInterface",
"SymbolInterfaces",
]
]
gentbl_cc_library(
name = "TensorEncodingIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-attr-interface-decls"],
"include/mlir/IR/TensorEncInterfaces.h.inc",
),
(
["-gen-attr-interface-defs"],
"include/mlir/IR/TensorEncInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/TensorEncoding.td",
deps = [":TensorOpsTdFiles"],
)
td_library(
name = "BuiltinDialectTdFiles",
srcs = [
[mlir] Refactor ElementsAttr into an AttrInterface This revision refactors ElementsAttr into an Attribute Interface. This enables a common interface with which to interact with element attributes, without needing to modify the builtin dialect. It also removes a majority (if not all?) of the need for the current OpaqueElementsAttr, which was originally intended as a way to opaquely represent data that was not representable by the other builtin constructs. The new ElementsAttr interface not only allows for users to natively represent their data in the way that best suits them, it also allows for efficient opaque access and iteration of the underlying data. Attributes using the ElementsAttr interface can directly expose support for interacting with the held elements using any C++ data type they claim to support. For example, DenseIntOrFpElementsAttr supports iteration using various native C++ integer/float data types, as well as APInt/APFloat, and more. ElementsAttr instances that refer to DenseIntOrFpElementsAttr can use all of these data types for iteration: ```c++ DenseIntOrFpElementsAttr intElementsAttr = ...; ElementsAttr attr = intElementsAttr; for (uint64_t value : attr.getValues<uint64_t>()) ...; for (APInt value : attr.getValues<APInt>()) ...; for (IntegerAttr value : attr.getValues<IntegerAttr>()) ...; ``` ElementsAttr also supports failable range/iterator access, allowing for selective code paths depending on data type support: ```c++ ElementsAttr attr = ...; if (auto range = attr.tryGetValues<uint64_t>()) { for (uint64_t value : *range) ...; } ``` Differential Revision: https://reviews.llvm.org/D109190
2021-09-21 01:40:45 +00:00
"include/mlir/IR/BuiltinAttributeInterfaces.td",
"include/mlir/IR/BuiltinAttributes.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/IR/BuiltinDialect.td",
"include/mlir/IR/BuiltinLocationAttributes.td",
"include/mlir/IR/BuiltinOps.td",
"include/mlir/IR/BuiltinTypeInterfaces.td",
"include/mlir/IR/BuiltinTypes.td",
],
includes = ["include"],
deps = [
[mlir] Dialect type/attr bytecode read/write generator. Tool to help generate dialect bytecode Attribute & Type reader/writing. Show usage by flipping builtin dialect. It helps reduce boilerplate when writing dialect bytecode attribute and type readers/writers. It is not an attempt at a generic spec mechanism but rather practically focussing on boilerplate reduction while also considering that it need not be the only in memory format and make it relatively easy to change. There should be some cleanup in follow up as we expand to more dialects. Differential Revision: https://reviews.llvm.org/D144820
2023-04-24 11:53:58 -07:00
":BytecodeTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":CallInterfacesTdFiles",
":CastInterfacesTdFiles",
":DataLayoutInterfacesTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
[mlir][Bazel] Adjust BUILD files for a9d003ef855ff7ed1bf4f8229ee9944b55936e6f
2023-06-05 09:57:01 +02:00
td_library(
name = "BuiltinDialectBytecodeTdFiles",
srcs = ["include/mlir/IR/BuiltinDialectBytecode.td"],
includes = ["include"],
deps = [
":BytecodeTdFiles",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "BuiltinDialectIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-dialect-decls"],
"include/mlir/IR/BuiltinDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
"include/mlir/IR/BuiltinDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/BuiltinDialect.td",
deps = [":BuiltinDialectTdFiles"],
)
[mlir] Dialect type/attr bytecode read/write generator. Tool to help generate dialect bytecode Attribute & Type reader/writing. Show usage by flipping builtin dialect. It helps reduce boilerplate when writing dialect bytecode attribute and type readers/writers. It is not an attempt at a generic spec mechanism but rather practically focussing on boilerplate reduction while also considering that it need not be the only in memory format and make it relatively easy to change. There should be some cleanup in follow up as we expand to more dialects. Differential Revision: https://reviews.llvm.org/D144820
2023-04-24 11:53:58 -07:00
gentbl_cc_library(
name = "BuiltinDialectBytecodeGen",
strip_include_prefix = "include",
tbl_outs = [
(
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
[
"-gen-bytecode",
"-bytecode-dialect=Builtin",
],
[mlir] Dialect type/attr bytecode read/write generator. Tool to help generate dialect bytecode Attribute & Type reader/writing. Show usage by flipping builtin dialect. It helps reduce boilerplate when writing dialect bytecode attribute and type readers/writers. It is not an attempt at a generic spec mechanism but rather practically focussing on boilerplate reduction while also considering that it need not be the only in memory format and make it relatively easy to change. There should be some cleanup in follow up as we expand to more dialects. Differential Revision: https://reviews.llvm.org/D144820
2023-04-24 11:53:58 -07:00
"include/mlir/IR/BuiltinDialectBytecode.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/BuiltinDialectBytecode.td",
deps = [":BuiltinDialectTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "BuiltinAttributesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["--gen-attrdef-decls"],
"include/mlir/IR/BuiltinAttributes.h.inc",
),
(
["--gen-attrdef-defs"],
"include/mlir/IR/BuiltinAttributes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/BuiltinAttributes.td",
deps = [":BuiltinDialectTdFiles"],
)
[mlir] Refactor ElementsAttr into an AttrInterface This revision refactors ElementsAttr into an Attribute Interface. This enables a common interface with which to interact with element attributes, without needing to modify the builtin dialect. It also removes a majority (if not all?) of the need for the current OpaqueElementsAttr, which was originally intended as a way to opaquely represent data that was not representable by the other builtin constructs. The new ElementsAttr interface not only allows for users to natively represent their data in the way that best suits them, it also allows for efficient opaque access and iteration of the underlying data. Attributes using the ElementsAttr interface can directly expose support for interacting with the held elements using any C++ data type they claim to support. For example, DenseIntOrFpElementsAttr supports iteration using various native C++ integer/float data types, as well as APInt/APFloat, and more. ElementsAttr instances that refer to DenseIntOrFpElementsAttr can use all of these data types for iteration: ```c++ DenseIntOrFpElementsAttr intElementsAttr = ...; ElementsAttr attr = intElementsAttr; for (uint64_t value : attr.getValues<uint64_t>()) ...; for (APInt value : attr.getValues<APInt>()) ...; for (IntegerAttr value : attr.getValues<IntegerAttr>()) ...; ``` ElementsAttr also supports failable range/iterator access, allowing for selective code paths depending on data type support: ```c++ ElementsAttr attr = ...; if (auto range = attr.tryGetValues<uint64_t>()) { for (uint64_t value : *range) ...; } ``` Differential Revision: https://reviews.llvm.org/D109190
2021-09-21 01:40:45 +00:00
gentbl_cc_library(
name = "BuiltinAttributeInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["--gen-attr-interface-decls"],
"include/mlir/IR/BuiltinAttributeInterfaces.h.inc",
),
(
["--gen-attr-interface-defs"],
"include/mlir/IR/BuiltinAttributeInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/BuiltinAttributeInterfaces.td",
deps = [":BuiltinDialectTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "BuiltinLocationAttributesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["--gen-attrdef-decls"],
"include/mlir/IR/BuiltinLocationAttributes.h.inc",
),
(
["--gen-attrdef-defs"],
"include/mlir/IR/BuiltinLocationAttributes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/BuiltinLocationAttributes.td",
deps = [":BuiltinDialectTdFiles"],
)
gentbl_cc_library(
name = "BuiltinOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/IR/BuiltinOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/IR/BuiltinOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/BuiltinOps.td",
[bazel] Port 7ceffae18c43
2022-01-19 14:14:36 +01:00
deps = [
":BuiltinDialectTdFiles",
":FunctionInterfacesTdFiles",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
gentbl_cc_library(
name = "BuiltinTypesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["--gen-typedef-decls"],
"include/mlir/IR/BuiltinTypes.h.inc",
),
(
["--gen-typedef-defs"],
"include/mlir/IR/BuiltinTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/BuiltinTypes.td",
deps = [":BuiltinDialectTdFiles"],
)
gentbl_cc_library(
name = "BuiltinTypeInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["--gen-type-interface-decls"],
"include/mlir/IR/BuiltinTypeInterfaces.h.inc",
),
(
["--gen-type-interface-defs"],
"include/mlir/IR/BuiltinTypeInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/BuiltinTypeInterfaces.td",
[bazel] Port 7ceffae18c43
2022-01-19 14:14:36 +01:00
deps = [
":BuiltinDialectTdFiles",
],
)
td_library(
name = "FunctionInterfacesTdFiles",
srcs = ["include/mlir/IR/FunctionInterfaces.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
gentbl_cc_library(
name = "FunctionInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/IR/FunctionOpInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/IR/FunctionOpInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/IR/FunctionInterfaces.td",
deps = [
":OpBaseTdFiles",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
cc_library(
name = "IR",
srcs = glob([
"lib/IR/*.cpp",
"lib/IR/*.h",
[bazel] Merge BytecodeOpInterface target into IR. Reviewed By: akuegel Differential Revision: https://reviews.llvm.org/D152133
2023-06-05 11:45:19 +02:00
"lib/Bytecode/Reader/*.h",
"lib/Bytecode/Writer/*.h",
"lib/Bytecode/*.h",
]) + [
"lib/Bytecode/BytecodeOpInterface.cpp",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/IR/*.h",
[bazel] Merge BytecodeOpInterface target into IR. Reviewed By: akuegel Differential Revision: https://reviews.llvm.org/D152133
2023-06-05 11:45:19 +02:00
"include/mlir/Bytecode/*.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
]) + [
"include/mlir/Interfaces/CallInterfaces.h",
"include/mlir/Interfaces/DataLayoutInterfaces.h",
"include/mlir/Interfaces/FoldInterfaces.h",
[mlir][bazel] reformat BUILD.bazel with buildifier Reviewed By: jreiffers Differential Revision: https://reviews.llvm.org/D147244
2023-03-30 10:10:57 -07:00
"include/mlir/Interfaces/SideEffectInterfaces.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
[mlir] Refactor ElementsAttr into an AttrInterface This revision refactors ElementsAttr into an Attribute Interface. This enables a common interface with which to interact with element attributes, without needing to modify the builtin dialect. It also removes a majority (if not all?) of the need for the current OpaqueElementsAttr, which was originally intended as a way to opaquely represent data that was not representable by the other builtin constructs. The new ElementsAttr interface not only allows for users to natively represent their data in the way that best suits them, it also allows for efficient opaque access and iteration of the underlying data. Attributes using the ElementsAttr interface can directly expose support for interacting with the held elements using any C++ data type they claim to support. For example, DenseIntOrFpElementsAttr supports iteration using various native C++ integer/float data types, as well as APInt/APFloat, and more. ElementsAttr instances that refer to DenseIntOrFpElementsAttr can use all of these data types for iteration: ```c++ DenseIntOrFpElementsAttr intElementsAttr = ...; ElementsAttr attr = intElementsAttr; for (uint64_t value : attr.getValues<uint64_t>()) ...; for (APInt value : attr.getValues<APInt>()) ...; for (IntegerAttr value : attr.getValues<IntegerAttr>()) ...; ``` ElementsAttr also supports failable range/iterator access, allowing for selective code paths depending on data type support: ```c++ ElementsAttr attr = ...; if (auto range = attr.tryGetValues<uint64_t>()) { for (uint64_t value : *range) ...; } ``` Differential Revision: https://reviews.llvm.org/D109190
2021-09-21 01:40:45 +00:00
":BuiltinAttributeInterfacesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":BuiltinAttributesIncGen",
[mlir] Dialect type/attr bytecode read/write generator. Tool to help generate dialect bytecode Attribute & Type reader/writing. Show usage by flipping builtin dialect. It helps reduce boilerplate when writing dialect bytecode attribute and type readers/writers. It is not an attempt at a generic spec mechanism but rather practically focussing on boilerplate reduction while also considering that it need not be the only in memory format and make it relatively easy to change. There should be some cleanup in follow up as we expand to more dialects. Differential Revision: https://reviews.llvm.org/D144820
2023-04-24 11:53:58 -07:00
":BuiltinDialectBytecodeGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":BuiltinDialectIncGen",
":BuiltinLocationAttributesIncGen",
":BuiltinOpsIncGen",
":BuiltinTypeInterfacesIncGen",
":BuiltinTypesIncGen",
[bazel] Merge BytecodeOpInterface target into IR. Reviewed By: akuegel Differential Revision: https://reviews.llvm.org/D152133
2023-06-05 11:45:19 +02:00
":BytecodeOpInterfaceIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":CallOpInterfacesIncGen",
":DataLayoutInterfacesIncGen",
[bazel] Port 7ceffae18c43
2022-01-19 14:14:36 +01:00
":FunctionInterfacesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":InferTypeOpInterfaceIncGen",
":OpAsmInterfaceIncGen",
":RegionKindInterfaceIncGen",
":SideEffectInterfacesIncGen",
":Support",
":SymbolInterfacesIncGen",
":TensorEncodingIncGen",
"//llvm:Support",
],
)
cc_library(
name = "Pass",
srcs = glob([
"lib/Pass/*.cpp",
"lib/Pass/*.h",
]),
hdrs = glob([
"include/mlir/Pass/*.h",
]),
includes = ["include"],
deps = [
":IR",
[Fix/Build] Fixing missing dependency for bazel.
2022-06-29 20:43:40 +00:00
":Parser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
"//llvm:Support",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
name = "CAPIIR",
srcs = [
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"lib/CAPI/Dialect/Func.cpp",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"lib/CAPI/IR/AffineExpr.cpp",
"lib/CAPI/IR/AffineMap.cpp",
"lib/CAPI/IR/BuiltinAttributes.cpp",
"lib/CAPI/IR/BuiltinTypes.cpp",
"lib/CAPI/IR/Diagnostics.cpp",
"lib/CAPI/IR/DialectHandle.cpp",
"lib/CAPI/IR/IR.cpp",
"lib/CAPI/IR/IntegerSet.cpp",
"lib/CAPI/IR/Pass.cpp",
"lib/CAPI/IR/Support.cpp",
],
hdrs = [
"include/mlir-c/AffineExpr.h",
"include/mlir-c/AffineMap.h",
"include/mlir-c/BuiltinAttributes.h",
"include/mlir-c/BuiltinTypes.h",
"include/mlir-c/Diagnostics.h",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir-c/Dialect/Func.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir-c/ExecutionEngine.h",
"include/mlir-c/IR.h",
"include/mlir-c/IntegerSet.h",
[mlir] support interfaces in Python bindings Introduce the initial support for operation interfaces in C API and Python bindings. Interfaces are a key component of MLIR's extensibility and should be available in bindings to make use of full potential of MLIR. This initial implementation exposes InferTypeOpInterface all the way to the Python bindings since it can be later used to simplify the operation construction methods by inferring their return types instead of requiring the user to do so. The general infrastructure for binding interfaces is defined and InferTypeOpInterface can be used as an example for binding other interfaces. Reviewed By: gysit Differential Revision: https://reviews.llvm.org/D111656
2021-10-14 17:18:28 +02:00
"include/mlir-c/Interfaces.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir-c/Pass.h",
[mlir] Overhaul C/Python registration APIs to properly scope registration/loading activities. Since the very first commits, the Python and C MLIR APIs have had mis-placed registration/load functionality for dialects, extensions, etc. This was done pragmatically in order to get bootstrapped and then just grew in. Downstreams largely bypass and do their own thing by providing various APIs to register things they need. Meanwhile, the C++ APIs have stabilized around this and it would make sense to follow suit. The thing we have observed in canonical usage by downstreams is that each downstream tends to have native entry points that configure its installation to its preferences with one-stop APIs. This patch leans in to this approach with `RegisterEverything.h` and `mlir._mlir_libs._mlirRegisterEverything` being the one-stop entry points for the "upstream packages". The `_mlir_libs.__init__.py` now allows customization of the environment and Context by adding "initialization modules" to the `_mlir_libs` package. If present, `_mlirRegisterEverything` is treated as such a module. Others can be added by downstreams by adding a `_site_initialize_{i}.py` module, where '{i}' is a number starting with zero. The number will be incremented and corresponding module loaded until one is not found. Initialization modules can: * Perform load time customization to the global environment (i.e. registering passes, hooks, etc). * Define a `register_dialects(registry: DialectRegistry)` function that can extend the `DialectRegistry` that will be used to bootstrap the `Context`. * Define a `context_init_hook(context: Context)` function that will be added to a list of callbacks which will be invoked after dialect registration during `Context` initialization. Note that the `MLIRPythonExtension.RegisterEverything` is not included by default when building a downstream (its corresponding behavior was prior). For downstreams which need the default MLIR initialization to take place, they must add this back in to their Python CMake build just like they add their own components (i.e. to `add_mlir_python_common_capi_library` and `add_mlir_python_modules`). It is perfectly valid to not do this, in which case, only the things explicitly depended on and initialized by downstreams will be built/packaged. If the downstream has not been set up for this, it is recommended to simply add this back for the time being and pay the build time/package size cost. CMake changes: * `MLIRCAPIRegistration` -> `MLIRCAPIRegisterEverything` (renamed to signify what it does and force an evaluation: a number of places were incidentally linking this very expensive target) * `MLIRPythonSoure.Passes` removed (without replacement: just drop) * `MLIRPythonExtension.AllPassesRegistration` removed (without replacement: just drop) * `MLIRPythonExtension.Conversions` removed (without replacement: just drop) * `MLIRPythonExtension.Transforms` removed (without replacement: just drop) Header changes: * `mlir-c/Registration.h` is deleted. Dialect registration functionality is now in `IR.h`. Registration of upstream features are in `mlir-c/RegisterEverything.h`. When updating MLIR and a couple of downstreams, I found that proper usage was commingled so required making a choice vs just blind S&R. Python APIs removed: * mlir.transforms and mlir.conversions (previously only had an __init__.py which indirectly triggered `mlirRegisterTransformsPasses()` and `mlirRegisterConversionPasses()` respectively). Downstream impact: Remove these imports if present (they now happen as part of default initialization). * mlir._mlir_libs._all_passes_registration, mlir._mlir_libs._mlirTransforms, mlir._mlir_libs._mlirConversions. Downstream impact: None expected (these were internally used). C-APIs changed: * mlirRegisterAllDialects(MlirContext) now takes an MlirDialectRegistry instead. It also used to trigger loading of all dialects, which was already marked with a TODO to remove -- it no longer does, and for direct use, dialects must be explicitly loaded. Downstream impact: Direct C-API users must ensure that needed dialects are loaded or call `mlirContextLoadAllAvailableDialects(MlirContext)` to emulate the prior behavior. Also see the `ir.c` test case (e.g. ` mlirContextGetOrLoadDialect(ctx, mlirStringRefCreateFromCString("func"));`). * mlirDialectHandle* APIs were moved from Registration.h (which now is restricted to just global/upstream registration) to IR.h, arguably where it should have been. Downstream impact: include correct header (likely already doing so). C-APIs added: * mlirContextLoadAllAvailableDialects(MlirContext): Corresponds to C++ API with the same purpose. Python APIs added: * mlir.ir.DialectRegistry: Mapping for an MlirDialectRegistry. * mlir.ir.Context.append_dialect_registry(MlirDialectRegistry) * mlir.ir.Context.load_all_available_dialects() * mlir._mlir_libs._mlirAllRegistration: New native extension that exposes a `register_dialects(MlirDialectRegistry)` entry point and performs all upstream pass/conversion/transforms registration on init. In this first step, we eagerly load this as part of the __init__.py and use it to monkey patch the Context to emulate prior behavior. * Type caster and capsule support for MlirDialectRegistry This should make it possible to build downstream Python dialects that only depend on a subset of MLIR. See: https://github.com/llvm/llvm-project/issues/56037 Here is an example PR, minimally adapting IREE to these changes: https://github.com/iree-org/iree/pull/9638/files In this situation, IREE is opting to not link everything, since it is already configuring the Context to its liking. For projects that would just like to not think about it and pull in everything, add `MLIRPythonExtension.RegisterEverything` to the list of Python sources getting built, and the old behavior will continue. Reviewed By: mehdi_amini, ftynse Differential Revision: https://reviews.llvm.org/D128593
2022-07-16 16:09:03 -07:00
"include/mlir-c/RegisterEverything.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir-c/Support.h",
"include/mlir/CAPI/AffineExpr.h",
"include/mlir/CAPI/AffineMap.h",
"include/mlir/CAPI/Diagnostics.h",
"include/mlir/CAPI/IR.h",
"include/mlir/CAPI/IntegerSet.h",
[bazel] fix bazel Reviewed By: wrengr Differential Revision: https://reviews.llvm.org/D150405
2023-05-11 22:31:09 +00:00
"include/mlir/CAPI/Interfaces.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/CAPI/Pass.h",
"include/mlir/CAPI/Registration.h",
"include/mlir/CAPI/Support.h",
"include/mlir/CAPI/Utils.h",
"include/mlir/CAPI/Wrap.h",
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
header_deps = [
[bazel] Fix layering check after 0610e2f `BytecodeWriter.h` is now included in `IR.h`, and the corresponding dependency should be moved to `header_deps`. This change fixes Google's internal build, which seems to have a stricter layering check than bazel's, but it's still the right thing to do. Differential Revision: https://reviews.llvm.org/D149576
2023-05-01 12:04:32 +02:00
":BytecodeWriter",
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
":IR",
":Pass",
":Support",
"//llvm:Support",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
":AsmParser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":InferTypeOpInterface",
":Parser",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
[mlir] support interfaces in Python bindings Introduce the initial support for operation interfaces in C API and Python bindings. Interfaces are a key component of MLIR's extensibility and should be available in bindings to make use of full potential of MLIR. This initial implementation exposes InferTypeOpInterface all the way to the Python bindings since it can be later used to simplify the operation construction methods by inferring their return types instead of requiring the user to do so. The general infrastructure for binding interfaces is defined and InferTypeOpInterface can be used as an example for binding other interfaces. Reviewed By: gysit Differential Revision: https://reviews.llvm.org/D111656
2021-10-14 17:18:28 +02:00
name = "CAPIInterfaces",
srcs = [
"lib/CAPI/Interfaces/Interfaces.cpp",
],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
],
[mlir] support interfaces in Python bindings Introduce the initial support for operation interfaces in C API and Python bindings. Interfaces are a key component of MLIR's extensibility and should be available in bindings to make use of full potential of MLIR. This initial implementation exposes InferTypeOpInterface all the way to the Python bindings since it can be later used to simplify the operation construction methods by inferring their return types instead of requiring the user to do so. The general infrastructure for binding interfaces is defined and InferTypeOpInterface can be used as an example for binding other interfaces. Reviewed By: gysit Differential Revision: https://reviews.llvm.org/D111656
2021-10-14 17:18:28 +02:00
includes = ["include"],
deps = [
":IR",
":InferTypeOpInterface",
"//llvm:Support",
],
)
[MLIR][CAPI] Add C API dialect registration methods for Arith, Math, MemRef and Vector dialects Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D155450
2023-07-17 14:32:31 +00:00
mlir_c_api_cc_library(
name = "CAPIArith",
srcs = ["lib/CAPI/Dialect/Arith.cpp"],
hdrs = ["include/mlir-c/Dialect/Arith.h"],
capi_deps = [
":CAPIIR",
],
includes = ["include"],
deps = [
":ArithDialect",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
name = "CAPIAsync",
srcs = [
"lib/CAPI/Dialect/Async.cpp",
"lib/CAPI/Dialect/AsyncPasses.cpp",
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
hdrs = ["include/mlir-c/Dialect/Async.h"],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
header_deps = [
":AsyncPassIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AsyncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AsyncTransforms",
":Pass",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
name = "CAPILinalg",
srcs = [
"lib/CAPI/Dialect/Linalg.cpp",
"lib/CAPI/Dialect/LinalgPasses.cpp",
],
hdrs = [
"include/mlir-c/Dialect/Linalg.h",
],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
header_deps = [
":LinalgPassIncGen",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LinalgTransforms",
":Pass",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
[Bazel] Fix build for 35d4593e6b555f852088211f5561c0e360c98c91
2021-07-05 22:39:10 +02:00
name = "CAPILLVM",
srcs = [
"lib/CAPI/Dialect/LLVM.cpp",
],
hdrs = [
"include/mlir-c/Dialect/LLVM.h",
],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
],
[Bazel] Fix build for 35d4593e6b555f852088211f5561c0e360c98c91
2021-07-05 22:39:10 +02:00
includes = ["include"],
deps = [
":LLVMDialect",
],
)
[MLIR][CAPI] Add C API dialect registration methods for Arith, Math, MemRef and Vector dialects Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D155450
2023-07-17 14:32:31 +00:00
mlir_c_api_cc_library(
name = "CAPIMath",
srcs = ["lib/CAPI/Dialect/Math.cpp"],
hdrs = ["include/mlir-c/Dialect/Math.h"],
capi_deps = [
":CAPIIR",
],
includes = ["include"],
deps = [
":MathDialect",
],
)
mlir_c_api_cc_library(
name = "CAPIMemRef",
srcs = ["lib/CAPI/Dialect/MemRef.cpp"],
hdrs = ["include/mlir-c/Dialect/MemRef.h"],
capi_deps = [
":CAPIIR",
],
includes = ["include"],
deps = [
":MemRefDialect",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
name = "CAPIGPU",
srcs = [
"lib/CAPI/Dialect/GPU.cpp",
"lib/CAPI/Dialect/GPUPasses.cpp",
],
hdrs = [
"include/mlir-c/Dialect/GPU.h",
],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
header_deps = [
":GPUPassIncGen",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":GPUDialect",
":GPUTransforms",
":Pass",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
name = "CAPISparseTensor",
srcs = [
"lib/CAPI/Dialect/SparseTensor.cpp",
"lib/CAPI/Dialect/SparseTensorPasses.cpp",
],
hdrs = [
"include/mlir-c/Dialect/SparseTensor.h",
],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
header_deps = [
":SparseTensorPassIncGen",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":Pass",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":SparseTensorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SparseTensorTransforms",
":Support",
],
)
[mlir] Introduce C API for the Quantization dialect types Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D116546
2022-01-03 19:01:07 +01:00
mlir_c_api_cc_library(
name = "CAPIQuant",
srcs = [
"lib/CAPI/Dialect/Quant.cpp",
],
hdrs = [
"include/mlir-c/Dialect/Quant.h",
],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
[mlir] Introduce C API for the Quantization dialect types Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D116546
2022-01-03 19:01:07 +01:00
],
includes = ["include"],
deps = [
":QuantOps",
],
)
[mlir] Introduce C API for PDL dialect types This change introduces C API helper functions to work with PDL types. Modification closely follow the format of the https://reviews.llvm.org/D116546. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D117221
2022-01-13 11:33:42 +01:00
mlir_c_api_cc_library(
name = "CAPIPDL",
srcs = [
"lib/CAPI/Dialect/PDL.cpp",
],
hdrs = [
"include/mlir-c/Dialect/PDL.h",
],
header_deps = [
":CAPIIRHeaders",
],
includes = ["include"],
deps = [
":CAPIIR",
":PDLDialect",
":PDLOpsIncGen",
":PDLTypesIncGen",
],
)
[mlir] add OperationType to the Transform dialect Add a new OperationType handle type to the Transform dialect. This transform type is parameterized by the name of the payload operation it can point to. It is intended as a constraint on transformations that are only applicable to a specific kind of payload operations. If a transformation is applicable to a small set of operation classes, it can be wrapped into a transform op by using a disjunctive constraint, such as `Type<Or<[Transform_ConcreteOperation<"foo">.predicate, Transform_ConcreteOperation<"bar">.predicate]>>` for its operand without modifying this type. Broader sets of accepted operations should be modeled as specific types. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D135586
2022-10-10 14:33:59 +00:00
mlir_c_api_cc_library(
name = "CAPITransformDialect",
srcs = [
"lib/CAPI/Dialect/Transform.cpp",
],
hdrs = [
"include/mlir-c/Dialect/Transform.h",
],
capi_deps = [
":CAPIIR",
],
includes = ["include"],
deps = [
":TransformDialect",
],
)
[mlir][mlprogram] Add CAPI project for MLProgram Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D135291
2022-10-05 10:55:25 -07:00
mlir_c_api_cc_library(
name = "CAPIMLProgram",
srcs = [
"lib/CAPI/Dialect/MLProgram.cpp",
],
hdrs = [
"include/mlir-c/Dialect/MLProgram.h",
],
header_deps = [
":CAPIIRHeaders",
],
includes = ["include"],
deps = [
":CAPIIR",
":MLProgramDialect",
":MLProgramOpsIncGen",
":MLProgramTypesIncGen",
],
)
[MLIR][CAPI] Add C API dialect registration methods for Arith, Math, MemRef and Vector dialects Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D155450
2023-07-17 14:32:31 +00:00
mlir_c_api_cc_library(
name = "CAPIVector",
srcs = ["lib/CAPI/Dialect/Vector.cpp"],
hdrs = ["include/mlir-c/Dialect/Vector.h"],
capi_deps = [
":CAPIIR",
],
includes = ["include"],
deps = [
":VectorDialect",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
name = "CAPIConversion",
srcs = ["lib/CAPI/Conversion/Passes.cpp"],
hdrs = ["include/mlir-c/Conversion.h"],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
header_deps = [
":ConversionPassIncGen",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":ConversionPasses",
":Pass",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
name = "CAPIDebug",
srcs = ["lib/CAPI/Debug/Debug.cpp"],
hdrs = ["include/mlir-c/Debug.h"],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":Support",
"//llvm:Support",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
name = "CAPIExecutionEngine",
srcs = ["lib/CAPI/ExecutionEngine/ExecutionEngine.cpp"],
hdrs = [
"include/mlir-c/ExecutionEngine.h",
"include/mlir/CAPI/ExecutionEngine.h",
],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
header_deps = [
":ExecutionEngine",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir][Bazel] Adjust BUILD files to 0e9523efda8a4ad95ecb1d5b5e65e10bcc3711f5
2023-03-21 15:00:10 +01:00
":BuiltinToLLVMIRTranslation",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ExecutionEngineUtils",
":LLVMToLLVMIRTranslation",
fix build issue on bazel Needed to fix: https://github.com/llvm/llvm-project/commit/53a5c3ab4dbe434753e0e4e3505efa2e5692e752 https://github.com/llvm/llvm-project/commit/db7cc0348c9a0aa44b23948949afef488d5083dd Reviewed By: Peiming, anlunx Differential Revision: https://reviews.llvm.org/D152202
2023-06-05 15:13:32 -07:00
":OpenMPToLLVMIRTranslation",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:OrcJIT",
"//llvm:Support",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
name = "CAPITransforms",
srcs = ["lib/CAPI/Transforms/Passes.cpp"],
hdrs = ["include/mlir-c/Transforms.h"],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
header_deps = [
":TransformsPassIncGen",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":Pass",
":Transforms",
],
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
mlir_c_api_cc_library(
[mlir] Overhaul C/Python registration APIs to properly scope registration/loading activities. Since the very first commits, the Python and C MLIR APIs have had mis-placed registration/load functionality for dialects, extensions, etc. This was done pragmatically in order to get bootstrapped and then just grew in. Downstreams largely bypass and do their own thing by providing various APIs to register things they need. Meanwhile, the C++ APIs have stabilized around this and it would make sense to follow suit. The thing we have observed in canonical usage by downstreams is that each downstream tends to have native entry points that configure its installation to its preferences with one-stop APIs. This patch leans in to this approach with `RegisterEverything.h` and `mlir._mlir_libs._mlirRegisterEverything` being the one-stop entry points for the "upstream packages". The `_mlir_libs.__init__.py` now allows customization of the environment and Context by adding "initialization modules" to the `_mlir_libs` package. If present, `_mlirRegisterEverything` is treated as such a module. Others can be added by downstreams by adding a `_site_initialize_{i}.py` module, where '{i}' is a number starting with zero. The number will be incremented and corresponding module loaded until one is not found. Initialization modules can: * Perform load time customization to the global environment (i.e. registering passes, hooks, etc). * Define a `register_dialects(registry: DialectRegistry)` function that can extend the `DialectRegistry` that will be used to bootstrap the `Context`. * Define a `context_init_hook(context: Context)` function that will be added to a list of callbacks which will be invoked after dialect registration during `Context` initialization. Note that the `MLIRPythonExtension.RegisterEverything` is not included by default when building a downstream (its corresponding behavior was prior). For downstreams which need the default MLIR initialization to take place, they must add this back in to their Python CMake build just like they add their own components (i.e. to `add_mlir_python_common_capi_library` and `add_mlir_python_modules`). It is perfectly valid to not do this, in which case, only the things explicitly depended on and initialized by downstreams will be built/packaged. If the downstream has not been set up for this, it is recommended to simply add this back for the time being and pay the build time/package size cost. CMake changes: * `MLIRCAPIRegistration` -> `MLIRCAPIRegisterEverything` (renamed to signify what it does and force an evaluation: a number of places were incidentally linking this very expensive target) * `MLIRPythonSoure.Passes` removed (without replacement: just drop) * `MLIRPythonExtension.AllPassesRegistration` removed (without replacement: just drop) * `MLIRPythonExtension.Conversions` removed (without replacement: just drop) * `MLIRPythonExtension.Transforms` removed (without replacement: just drop) Header changes: * `mlir-c/Registration.h` is deleted. Dialect registration functionality is now in `IR.h`. Registration of upstream features are in `mlir-c/RegisterEverything.h`. When updating MLIR and a couple of downstreams, I found that proper usage was commingled so required making a choice vs just blind S&R. Python APIs removed: * mlir.transforms and mlir.conversions (previously only had an __init__.py which indirectly triggered `mlirRegisterTransformsPasses()` and `mlirRegisterConversionPasses()` respectively). Downstream impact: Remove these imports if present (they now happen as part of default initialization). * mlir._mlir_libs._all_passes_registration, mlir._mlir_libs._mlirTransforms, mlir._mlir_libs._mlirConversions. Downstream impact: None expected (these were internally used). C-APIs changed: * mlirRegisterAllDialects(MlirContext) now takes an MlirDialectRegistry instead. It also used to trigger loading of all dialects, which was already marked with a TODO to remove -- it no longer does, and for direct use, dialects must be explicitly loaded. Downstream impact: Direct C-API users must ensure that needed dialects are loaded or call `mlirContextLoadAllAvailableDialects(MlirContext)` to emulate the prior behavior. Also see the `ir.c` test case (e.g. ` mlirContextGetOrLoadDialect(ctx, mlirStringRefCreateFromCString("func"));`). * mlirDialectHandle* APIs were moved from Registration.h (which now is restricted to just global/upstream registration) to IR.h, arguably where it should have been. Downstream impact: include correct header (likely already doing so). C-APIs added: * mlirContextLoadAllAvailableDialects(MlirContext): Corresponds to C++ API with the same purpose. Python APIs added: * mlir.ir.DialectRegistry: Mapping for an MlirDialectRegistry. * mlir.ir.Context.append_dialect_registry(MlirDialectRegistry) * mlir.ir.Context.load_all_available_dialects() * mlir._mlir_libs._mlirAllRegistration: New native extension that exposes a `register_dialects(MlirDialectRegistry)` entry point and performs all upstream pass/conversion/transforms registration on init. In this first step, we eagerly load this as part of the __init__.py and use it to monkey patch the Context to emulate prior behavior. * Type caster and capsule support for MlirDialectRegistry This should make it possible to build downstream Python dialects that only depend on a subset of MLIR. See: https://github.com/llvm/llvm-project/issues/56037 Here is an example PR, minimally adapting IREE to these changes: https://github.com/iree-org/iree/pull/9638/files In this situation, IREE is opting to not link everything, since it is already configuring the Context to its liking. For projects that would just like to not think about it and pull in everything, add `MLIRPythonExtension.RegisterEverything` to the list of Python sources getting built, and the old behavior will continue. Reviewed By: mehdi_amini, ftynse Differential Revision: https://reviews.llvm.org/D128593
2022-07-16 16:09:03 -07:00
name = "CAPIRegisterEverything",
srcs = ["lib/CAPI/RegisterEverything/RegisterEverything.cpp"],
hdrs = ["include/mlir-c/RegisterEverything.h"],
[bazel] Separate capi_deps from deps in mlir_c_api_cc_library. This is important because *Objects targets need to only depend on other *Objects targets, not on the unsuffixed CAPI rules. Depending on how it is linked in the current setup, it can cause duplicate symbols. Differential Revision: https://reviews.llvm.org/D117176
2022-01-13 01:35:29 +00:00
capi_deps = [
":CAPIIR",
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[Bazel][mlir] Port 120cd5aafca2ccc8b04d208598e817ffc6d0b869
2023-06-17 02:57:46 +00:00
":AllExtensions",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AllPassesAndDialects",
[mlir][Bazel] Adjust BUILD files to 0e9523efda8a4ad95ecb1d5b5e65e10bcc3711f5
2023-03-21 15:00:10 +01:00
":BuiltinToLLVMIRTranslation",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMToLLVMIRTranslation",
],
)
##---------------------------------------------------------------------------##
# Sources of Python bindings.
#----------------------------------------------------------------------------##
exports_files(
glob(["lib/Bindings/Python/**/*.cpp"]),
)
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
MLIR_BINDINGS_PYTHON_HEADERS = [
"lib/Bindings/Python/*.h",
"include/mlir-c/Bindings/Python/*.h",
"include/mlir/Bindings/Python/*.h",
]
cc_library(
name = "MLIRBindingsPythonHeaders",
includes = [
"include",
"lib/Bindings/Python",
],
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
textual_hdrs = glob(MLIR_BINDINGS_PYTHON_HEADERS),
deps = [
":CAPIIRHeaders",
"@pybind11",
"@python_runtime//:headers",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "MLIRBindingsPythonHeadersAndDeps",
includes = [
"include",
"lib/Bindings/Python",
],
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
textual_hdrs = glob(MLIR_BINDINGS_PYTHON_HEADERS),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [
":CAPIIR",
"@pybind11",
"@python_runtime//:headers",
],
)
[mlir] Split out Python bindings for dialects into separate libs Historically, the bindings for the Linalg dialect were included into the "core" bindings library because they depended on the C++ implementation of the "core" bindings. The other dialects followed the pattern. Now that this dependency is gone, split out each dialect into a separate Python extension library. Depends On D116649, D116605 Reviewed By: stellaraccident Differential Revision: https://reviews.llvm.org/D116662
2022-01-05 11:21:21 +01:00
# These flags are needed for pybind11 to work.
PYBIND11_COPTS = [
"-fexceptions",
"-frtti",
]
PYBIND11_FEATURES = [
# Cannot use header_modules (parse_headers feature fails).
"-use_header_modules",
]
Don't define //mlir:MLIRBindingsPythonCore in terms of the NoCAPI and CAPIDeps targets. We noticed that the library structure causes link ordering problems in Google's internal build. However, we don't think the problem is specific to Google's build, it probably can be reproduced anywhere with the right library structure. In general splitting the Python bindings from their dependencies (the C API targets) creates the possibility that the two libraries might end up in the wrong order on the linker command line. We can avoid this problem happening by reverting the structure of the MLIRBindingsPythonCore to represent its dependencies in the usual way, rather than composing an incomplete `MLIRBindingsPythonCoreNoCAPI` target and their CAPI dependencies. It was probably a mistake to rewrite this particular `cc_library()` rule in terms of the two, since nothing guarantees that the two will be correctly ordered by the linker when both are being linked into the same binary, and it was only an incidental "cleanup" done in passing. Otherwise the previous PR (D113565) is fine, since that was about the case where both are being built into two separate shared libraries. It just shouldn't have made this (unrelated) change. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113773
2021-11-12 12:02:18 -08:00
MLIR_PYTHON_BINDINGS_SOURCES = [
"lib/Bindings/Python/IRAffine.cpp",
"lib/Bindings/Python/IRAttributes.cpp",
"lib/Bindings/Python/IRCore.cpp",
"lib/Bindings/Python/IRInterfaces.cpp",
"lib/Bindings/Python/IRModule.cpp",
"lib/Bindings/Python/IRTypes.cpp",
"lib/Bindings/Python/Pass.cpp",
]
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
Don't define //mlir:MLIRBindingsPythonCore in terms of the NoCAPI and CAPIDeps targets. We noticed that the library structure causes link ordering problems in Google's internal build. However, we don't think the problem is specific to Google's build, it probably can be reproduced anywhere with the right library structure. In general splitting the Python bindings from their dependencies (the C API targets) creates the possibility that the two libraries might end up in the wrong order on the linker command line. We can avoid this problem happening by reverting the structure of the MLIRBindingsPythonCore to represent its dependencies in the usual way, rather than composing an incomplete `MLIRBindingsPythonCoreNoCAPI` target and their CAPI dependencies. It was probably a mistake to rewrite this particular `cc_library()` rule in terms of the two, since nothing guarantees that the two will be correctly ordered by the linker when both are being linked into the same binary, and it was only an incidental "cleanup" done in passing. Otherwise the previous PR (D113565) is fine, since that was about the case where both are being built into two separate shared libraries. It just shouldn't have made this (unrelated) change. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113773
2021-11-12 12:02:18 -08:00
name = "MLIRBindingsPythonCore",
srcs = MLIR_PYTHON_BINDINGS_SOURCES,
[mlir] Split out Python bindings for dialects into separate libs Historically, the bindings for the Linalg dialect were included into the "core" bindings library because they depended on the C++ implementation of the "core" bindings. The other dialects followed the pattern. Now that this dependency is gone, split out each dialect into a separate Python extension library. Depends On D116649, D116605 Reviewed By: stellaraccident Differential Revision: https://reviews.llvm.org/D116662
2022-01-05 11:21:21 +01:00
copts = PYBIND11_COPTS,
features = PYBIND11_FEATURES,
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
deps = [
Don't define //mlir:MLIRBindingsPythonCore in terms of the NoCAPI and CAPIDeps targets. We noticed that the library structure causes link ordering problems in Google's internal build. However, we don't think the problem is specific to Google's build, it probably can be reproduced anywhere with the right library structure. In general splitting the Python bindings from their dependencies (the C API targets) creates the possibility that the two libraries might end up in the wrong order on the linker command line. We can avoid this problem happening by reverting the structure of the MLIRBindingsPythonCore to represent its dependencies in the usual way, rather than composing an incomplete `MLIRBindingsPythonCoreNoCAPI` target and their CAPI dependencies. It was probably a mistake to rewrite this particular `cc_library()` rule in terms of the two, since nothing guarantees that the two will be correctly ordered by the linker when both are being linked into the same binary, and it was only an incidental "cleanup" done in passing. Otherwise the previous PR (D113565) is fine, since that was about the case where both are being built into two separate shared libraries. It just shouldn't have made this (unrelated) change. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113773
2021-11-12 12:02:18 -08:00
":CAPIAsync",
":CAPIDebug",
":CAPIGPU",
":CAPIIR",
":CAPIInterfaces",
":MLIRBindingsPythonHeadersAndDeps",
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
"//llvm:Support",
"@pybind11",
"@python_runtime//:headers",
],
)
cc_library(
Don't define //mlir:MLIRBindingsPythonCore in terms of the NoCAPI and CAPIDeps targets. We noticed that the library structure causes link ordering problems in Google's internal build. However, we don't think the problem is specific to Google's build, it probably can be reproduced anywhere with the right library structure. In general splitting the Python bindings from their dependencies (the C API targets) creates the possibility that the two libraries might end up in the wrong order on the linker command line. We can avoid this problem happening by reverting the structure of the MLIRBindingsPythonCore to represent its dependencies in the usual way, rather than composing an incomplete `MLIRBindingsPythonCoreNoCAPI` target and their CAPI dependencies. It was probably a mistake to rewrite this particular `cc_library()` rule in terms of the two, since nothing guarantees that the two will be correctly ordered by the linker when both are being linked into the same binary, and it was only an incidental "cleanup" done in passing. Otherwise the previous PR (D113565) is fine, since that was about the case where both are being built into two separate shared libraries. It just shouldn't have made this (unrelated) change. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113773
2021-11-12 12:02:18 -08:00
name = "MLIRBindingsPythonCoreNoCAPI",
srcs = MLIR_PYTHON_BINDINGS_SOURCES,
[mlir] Split out Python bindings for dialects into separate libs Historically, the bindings for the Linalg dialect were included into the "core" bindings library because they depended on the C++ implementation of the "core" bindings. The other dialects followed the pattern. Now that this dependency is gone, split out each dialect into a separate Python extension library. Depends On D116649, D116605 Reviewed By: stellaraccident Differential Revision: https://reviews.llvm.org/D116662
2022-01-05 11:21:21 +01:00
copts = PYBIND11_COPTS,
features = PYBIND11_FEATURES,
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [
Don't define //mlir:MLIRBindingsPythonCore in terms of the NoCAPI and CAPIDeps targets. We noticed that the library structure causes link ordering problems in Google's internal build. However, we don't think the problem is specific to Google's build, it probably can be reproduced anywhere with the right library structure. In general splitting the Python bindings from their dependencies (the C API targets) creates the possibility that the two libraries might end up in the wrong order on the linker command line. We can avoid this problem happening by reverting the structure of the MLIRBindingsPythonCore to represent its dependencies in the usual way, rather than composing an incomplete `MLIRBindingsPythonCoreNoCAPI` target and their CAPI dependencies. It was probably a mistake to rewrite this particular `cc_library()` rule in terms of the two, since nothing guarantees that the two will be correctly ordered by the linker when both are being linked into the same binary, and it was only an incidental "cleanup" done in passing. Otherwise the previous PR (D113565) is fine, since that was about the case where both are being built into two separate shared libraries. It just shouldn't have made this (unrelated) change. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113773
2021-11-12 12:02:18 -08:00
":CAPIAsyncHeaders",
":CAPIDebugHeaders",
":CAPIGPUHeaders",
":CAPIIRHeaders",
":MLIRBindingsPythonHeaders",
"//llvm:Support",
"@pybind11",
"@python_runtime//:headers",
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
],
)
Don't define //mlir:MLIRBindingsPythonCore in terms of the NoCAPI and CAPIDeps targets. We noticed that the library structure causes link ordering problems in Google's internal build. However, we don't think the problem is specific to Google's build, it probably can be reproduced anywhere with the right library structure. In general splitting the Python bindings from their dependencies (the C API targets) creates the possibility that the two libraries might end up in the wrong order on the linker command line. We can avoid this problem happening by reverting the structure of the MLIRBindingsPythonCore to represent its dependencies in the usual way, rather than composing an incomplete `MLIRBindingsPythonCoreNoCAPI` target and their CAPI dependencies. It was probably a mistake to rewrite this particular `cc_library()` rule in terms of the two, since nothing guarantees that the two will be correctly ordered by the linker when both are being linked into the same binary, and it was only an incidental "cleanup" done in passing. Otherwise the previous PR (D113565) is fine, since that was about the case where both are being built into two separate shared libraries. It just shouldn't have made this (unrelated) change. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113773
2021-11-12 12:02:18 -08:00
# Target that bundles together the CAPI objects needed for
# MLIRBindingsPythonCoreNoCAPI.
Split headers from implementations in MLIR C API Bazel build. This allows clients to build, e.g., the Python bindings against the C API headers, without including the C API implementations. This is useful when distributing software as multiple shared libraries. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D113565
2021-11-11 08:34:18 -08:00
cc_library(
name = "MLIRBindingsPythonCAPIObjects",
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
deps = [
":CAPIAsyncObjects",
":CAPIDebugObjects",
":CAPIGPUObjects",
":CAPIIRObjects",
":CAPIInterfacesObjects",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
# Dynamic library with the MLIR Python extension.
cc_binary(
name = "_mlir.so",
srcs = ["lib/Bindings/Python/MainModule.cpp"],
# These flags are needed for pybind11 to work.
copts = PYBIND11_COPTS,
features = PYBIND11_FEATURES,
linkshared = 1,
linkstatic = 0,
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
deps = [
":MLIRBindingsPythonCore",
":MLIRBindingsPythonHeadersAndDeps",
],
)
[mlir] Split out Python bindings for dialects into separate libs Historically, the bindings for the Linalg dialect were included into the "core" bindings library because they depended on the C++ implementation of the "core" bindings. The other dialects followed the pattern. Now that this dependency is gone, split out each dialect into a separate Python extension library. Depends On D116649, D116605 Reviewed By: stellaraccident Differential Revision: https://reviews.llvm.org/D116662
2022-01-05 11:21:21 +01:00
cc_binary(
name = "_mlirDialectsLinalg.so",
srcs = ["lib/Bindings/Python/DialectLinalg.cpp"],
copts = PYBIND11_COPTS,
features = PYBIND11_FEATURES,
linkshared = 1,
linkstatic = 0,
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
deps = [
":CAPIIR",
":CAPILinalg",
":MLIRBindingsPythonHeadersAndDeps",
],
)
cc_binary(
name = "_mlirDialectsQuant.so",
srcs = ["lib/Bindings/Python/DialectQuant.cpp"],
copts = PYBIND11_COPTS,
features = PYBIND11_FEATURES,
linkshared = 1,
linkstatic = 0,
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
deps = [
":CAPIIR",
":CAPIQuant",
":MLIRBindingsPythonHeadersAndDeps",
],
)
cc_binary(
name = "_mlirDialectsSparseTensor.so",
srcs = ["lib/Bindings/Python/DialectSparseTensor.cpp"],
copts = PYBIND11_COPTS,
features = PYBIND11_FEATURES,
linkshared = 1,
linkstatic = 0,
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
deps = [
":CAPIIR",
[bazel] Fix the python bindings for 7ceffae18c43 Also run buildifier.
2022-01-19 14:20:55 +01:00
":CAPISparseTensor",
[mlir] Split out Python bindings for dialects into separate libs Historically, the bindings for the Linalg dialect were included into the "core" bindings library because they depended on the C++ implementation of the "core" bindings. The other dialects followed the pattern. Now that this dependency is gone, split out each dialect into a separate Python extension library. Depends On D116649, D116605 Reviewed By: stellaraccident Differential Revision: https://reviews.llvm.org/D116662
2022-01-05 11:21:21 +01:00
":MLIRBindingsPythonHeadersAndDeps",
],
)
Update file names and extensions for MLIR Python execution engine changes.
2021-07-29 03:19:14 +00:00
# Dynamic library with the MLIR Conversions Python extension.
cc_binary(
name = "_mlirExecutionEngine.so",
srcs = ["lib/Bindings/Python/ExecutionEngineModule.cpp"],
copts = PYBIND11_COPTS,
features = PYBIND11_FEATURES,
linkshared = 1,
linkstatic = 0,
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
deps = [
":CAPIExecutionEngine",
":MLIRBindingsPythonHeadersAndDeps",
"@pybind11",
"@python_runtime//:headers",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
# Dynamic library with the MLIR Linalg dialect+passes Python extension.
cc_binary(
name = "_mlirLinalgPasses.so",
srcs = ["lib/Bindings/Python/LinalgPasses.cpp"],
copts = PYBIND11_COPTS,
features = PYBIND11_FEATURES,
linkshared = 1,
linkstatic = 0,
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
deps = [
":CAPILinalg",
":MLIRBindingsPythonHeadersAndDeps",
"@pybind11",
"@python_runtime//:headers",
],
)
##---------------------------------------------------------------------------##
td_library(
[mlir] Fix BUILD issues and dependencies. Differential Revision: https://reviews.llvm.org/D123868
2022-04-15 18:59:03 +00:00
name = "AttrTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = [
[mlir][Bazel] Adjust build file to account for new td files.
2022-03-15 09:04:07 +01:00
"include/mlir/IR/AttrTypeBase.td",
[mlir][ods] Added EnumAttr, an AttrDef implementation of enum attributes `EnumAttr` is a pure TableGen implementation of enum attributes using `AttrDef`. This is meant as a drop-in replacement for `StrEnumAttr`, which is soon to be deprecated. `StrEnumAttr` is often used over `IntEnumAttr` because its more readable in MLIR assembly formats. However, storing and manipulating strings is not efficient. Defining `StrEnumAttr` can also be awkward and relies on a lot of special logic in `EnumsGen`, and has some hidden sharp edges. Also, `EnumAttr` stores the enum directly, removing the need to convert to/from integers when calling attribute getters on ops. Reviewed By: mehdi_amini Differential Revision: https://reviews.llvm.org/D115181
2021-12-17 02:44:56 +00:00
"include/mlir/IR/EnumAttr.td",
[mlir] Fix BUILD issues and dependencies. Differential Revision: https://reviews.llvm.org/D123868
2022-04-15 18:59:03 +00:00
],
includes = ["include"],
)
td_library(
name = "OpBaseTdFiles",
srcs = [
"include/mlir/IR/DialectBase.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/IR/OpAsmInterface.td",
"include/mlir/IR/OpBase.td",
[mlir][Bazel] Adjust build file to account for new td files.
2022-03-15 09:04:07 +01:00
"include/mlir/IR/PatternBase.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/IR/RegionKindInterface.td",
"include/mlir/IR/SymbolInterfaces.td",
"include/mlir/IR/TensorEncoding.td",
],
includes = ["include"],
[mlir] Fix BUILD issues and dependencies. Differential Revision: https://reviews.llvm.org/D123868
2022-04-15 18:59:03 +00:00
deps = [
":AttrTdFiles",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
[mlir] Dialect type/attr bytecode read/write generator. Tool to help generate dialect bytecode Attribute & Type reader/writing. Show usage by flipping builtin dialect. It helps reduce boilerplate when writing dialect bytecode attribute and type readers/writers. It is not an attempt at a generic spec mechanism but rather practically focussing on boilerplate reduction while also considering that it need not be the only in memory format and make it relatively easy to change. There should be some cleanup in follow up as we expand to more dialects. Differential Revision: https://reviews.llvm.org/D144820
2023-04-24 11:53:58 -07:00
td_library(
name = "BytecodeTdFiles",
srcs = ["include/mlir/IR/BytecodeBase.td"],
includes = ["include"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "CallInterfacesTdFiles",
srcs = ["include/mlir/Interfaces/CallInterfaces.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
td_library(
name = "CastInterfacesTdFiles",
srcs = ["include/mlir/Interfaces/CastInterfaces.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
td_library(
name = "ControlFlowInterfacesTdFiles",
srcs = ["include/mlir/Interfaces/ControlFlowInterfaces.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
[mlir][bazel] Port for 660f714e26999d266232a1fbb02712bb879bd34e, second attempt.
2023-05-27 08:37:45 +02:00
td_library(
name = "BytecodeOpInterfaceTdFiles",
srcs = ["include/mlir/Bytecode/BytecodeOpInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "CopyOpInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/CopyOpInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
td_library(
name = "DerivedAttributeOpInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/DerivedAttributeOpInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
[mlir][interfaces][NFC] Move DestinationStyleOpInterface to mlir/Interfaces This is the second (and final) step of making "destination style" usable without depending on the Linalg dialect. (The first step was D135129.) This change allows us to provide default bufferization implementations for all destination-style ops. It also allows us to simplify `TilingInterface`. (E.g., `getDestinationOperands` can be removed.) Differential Revision: https://reviews.llvm.org/D136179
2022-10-18 17:23:42 +02:00
td_library(
name = "DestinationStyleOpInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/DestinationStyleOpInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
[bazel] Port 95aff23e2921
2022-06-03 20:43:02 +02:00
td_library(
name = "InferIntRangeInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/InferIntRangeInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "InferTypeOpInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/InferTypeOpInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
td_library(
name = "LoopLikeInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/LoopLikeInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
td_library(
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
name = "MemorySlotInterfacesTdFiles",
srcs = ["include/mlir/Interfaces/MemorySlotInterfaces.td"],
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
[mlir][interfaces] Add ShapedDimOpInterface This interface is implemented by memref.dim and tensor.dim. This change makes it possible to remove a build dependency of the Affine dialect on the Tensor dialect (and maybe also the MemRef dialect in the future). Differential Revision: https://reviews.llvm.org/D133595
2022-10-03 13:53:16 +09:00
td_library(
name = "ShapedOpInterfacesTdFiles",
srcs = ["include/mlir/Interfaces/ShapedOpInterfaces.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
[fix/build] bazel rule for ParallelCombiningOpInterface
2022-07-01 10:33:48 +02:00
td_library(
name = "ParallelCombiningOpInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/ParallelCombiningOpInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
[mlir] Add RuntimeVerifiableOpInterface and transform Static op verification cannot detect cases where an op is valid at compile time but may be invalid at runtime. An example of such an op is `memref::ExpandShapeOp`. Invalid at compile time: `memref.expand_shape %m [[0, 1]] : memref<11xf32> into memref<2x5xf32>` Valid at compile time (because we do not know any better): `memref.expand_shape %m [[0, 1]] : memref<?xf32> into memref<?x5xf32>`. This op may or may not be valid at runtime depending on the runtime shape of `%m`. Invalid runtime ops such as the one above are hard to debug because they can crash the program execution at a seemingly unrelated position or (even worse) compute an invalid result without crashing. This revision adds a new op interface `RuntimeVerifiableOpInterface` that can be implemented by ops that provide additional runtime verification. Such runtime verification can be computationally expensive, so it is only generated on an opt-in basis by running `-generate-runtime-verification`. A simple runtime verifier for `memref::ExpandShapeOp` is provided as an example. Differential Revision: https://reviews.llvm.org/D138576
2022-12-21 10:51:10 +01:00
td_library(
name = "RuntimeVerifiableOpInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/RuntimeVerifiableOpInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "SideEffectInterfacesTdFiles",
srcs = [
"include/mlir/Interfaces/SideEffectInterfaceBase.td",
"include/mlir/Interfaces/SideEffectInterfaces.td",
],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
[mlir] Add an interface to allow operations to specify how they can be tiled. An interface to allow for tiling of operations is introduced. The tiling of the linalg.pad_tensor operation is modified to use this interface. Differential Revision: https://reviews.llvm.org/D108611
2021-08-30 15:55:30 -07:00
td_library(
name = "TilingInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/TilingInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "VectorInterfacesTdFiles",
srcs = ["include/mlir/Interfaces/VectorInterfaces.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
td_library(
name = "ViewLikeInterfaceTdFiles",
srcs = ["include/mlir/Interfaces/ViewLikeInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
td_library(
name = "ReducerTdFiles",
srcs = ["include/mlir/Reducer/Passes.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
##---------------------------------------------------------------------------##
# Affine dialect.
##---------------------------------------------------------------------------##
td_library(
name = "PassBaseTdFiles",
srcs = ["include/mlir/Pass/PassBase.td"],
includes = ["include"],
)
td_library(
name = "RewritePassBaseTdFiles",
srcs = ["include/mlir/Rewrite/PassUtil.td"],
includes = ["include"],
)
td_library(
name = "AffineOpsTdFiles",
srcs = [
"include/mlir/Dialect/Affine/IR/AffineMemoryOpInterfaces.td",
"include/mlir/Dialect/Affine/IR/AffineOps.td",
],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithOpsTdFiles",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LoopLikeInterfaceTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "AffineOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Affine/IR/AffineOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Affine/IR/AffineOps.cpp.inc",
),
(
[mlir][arith] fixing dependencies on memref/arith
2021-12-30 20:39:22 +00:00
[
"-gen-dialect-decls",
"-dialect=affine",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/Dialect/Affine/IR/AffineOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[mlir][arith] fixing dependencies on memref/arith
2021-12-30 20:39:22 +00:00
[
"-gen-dialect-defs",
"-dialect=affine",
],
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
"include/mlir/Dialect/Affine/IR/AffineOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Affine/IR/AffineOps.td",
deps = [":AffineOpsTdFiles"],
)
gentbl_cc_library(
name = "AffineMemoryOpInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Dialect/Affine/IR/AffineMemoryOpInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Dialect/Affine/IR/AffineMemoryOpInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Affine/IR/AffineMemoryOpInterfaces.td",
deps = [":AffineOpsTdFiles"],
)
[mlir] bazel for D137997
2022-11-21 10:20:19 +01:00
td_library(
name = "AffineTransformOpsTdFiles",
srcs = [
"include/mlir/Dialect/Affine/TransformOps/AffineTransformOps.td",
],
includes = ["include"],
deps = [
":TransformDialectTdFiles",
],
)
gentbl_cc_library(
name = "AffineTransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Affine/TransformOps/AffineTransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Affine/TransformOps/AffineTransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Affine/TransformOps/AffineTransformOps.td",
deps = [
":AffineTransformOpsTdFiles",
],
)
cc_library(
name = "AffineTransformOps",
srcs = glob(["lib/Dialect/Affine/TransformOps/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/Affine/TransformOps/*.h"]),
includes = ["include"],
deps = [
[mlir][affine][transform] Simplify affine.min/max ops with given constraints This transform op uses `mlir::simplifyConstrainedMinMaxOp` to simplify `affine.min` and `affine.max` ops based on a given constraints. Differential Revision: https://reviews.llvm.org/D140997
2023-01-13 10:23:51 +01:00
":AffineAnalysis",
[mlir] bazel for D137997
2022-11-21 10:20:19 +01:00
":AffineDialect",
":AffineTransformOpsIncGen",
":AffineTransforms",
":AffineUtils",
":FuncDialect",
":IR",
":TransformDialect",
[mlir][affine][transform] Simplify affine.min/max ops with given constraints This transform op uses `mlir::simplifyConstrainedMinMaxOp` to simplify `affine.min` and `affine.max` ops based on a given constraints. Differential Revision: https://reviews.llvm.org/D140997
2023-01-13 10:23:51 +01:00
":Transforms",
[mlir] bazel for D137997
2022-11-21 10:20:19 +01:00
":VectorDialect",
],
)
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
##---------------------------------------------------------------------------##
# AMDGPU dialect.
##---------------------------------------------------------------------------##
td_library(
name = "AMDGPUTdFiles",
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
srcs = ["include/mlir/Dialect/AMDGPU/IR/AMDGPU.td"],
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
includes = ["include"],
deps = [
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "AMDGPUIncGen",
strip_include_prefix = "include",
tbl_outs = [
bazel fixes for c55b41d5199d2394dd6cdb8f52180d8b81d809d4
2022-09-01 08:19:50 +02:00
(
[
"-gen-attrdef-decls",
"-dialect=amdgpu",
],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
"include/mlir/Dialect/AMDGPU/IR/AMDGPUAttributes.h.inc",
bazel fixes for c55b41d5199d2394dd6cdb8f52180d8b81d809d4
2022-09-01 08:19:50 +02:00
),
(
[
"-gen-attrdef-defs",
"-dialect=amdgpu",
],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
"include/mlir/Dialect/AMDGPU/IR/AMDGPUAttributes.cpp.inc",
bazel fixes for c55b41d5199d2394dd6cdb8f52180d8b81d809d4
2022-09-01 08:19:50 +02:00
),
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
(
[
"-gen-dialect-decls",
"-dialect=amdgpu",
],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
"include/mlir/Dialect/AMDGPU/IR/AMDGPUDialect.h.inc",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
),
(
[
"-gen-dialect-defs",
"-dialect=amdgpu",
],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
"include/mlir/Dialect/AMDGPU/IR/AMDGPUDialect.cpp.inc",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
),
bazel fixes for c55b41d5199d2394dd6cdb8f52180d8b81d809d4
2022-09-01 08:19:50 +02:00
(
["-gen-enum-decls"],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
"include/mlir/Dialect/AMDGPU/IR/AMDGPUEnums.h.inc",
bazel fixes for c55b41d5199d2394dd6cdb8f52180d8b81d809d4
2022-09-01 08:19:50 +02:00
),
(
["-gen-enum-defs"],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
"include/mlir/Dialect/AMDGPU/IR/AMDGPUEnums.cpp.inc",
bazel fixes for c55b41d5199d2394dd6cdb8f52180d8b81d809d4
2022-09-01 08:19:50 +02:00
),
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
(
["-gen-op-decls"],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
"include/mlir/Dialect/AMDGPU/IR/AMDGPU.h.inc",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
),
(
["-gen-op-defs"],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
"include/mlir/Dialect/AMDGPU/IR/AMDGPU.cpp.inc",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
),
(
["-gen-op-doc"],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
"g3doc/Dialects/AMDGPU/IR/AMDGPU.md",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
),
],
tblgen = ":mlir-tblgen",
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
td_file = "include/mlir/Dialect/AMDGPU/IR/AMDGPU.td",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
deps = [":AMDGPUTdFiles"],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "AMDGPUDialect",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
srcs = ["lib/Dialect/AMDGPU/IR/AMDGPUDialect.cpp"],
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
hdrs = ["include/mlir/Dialect/AMDGPU/IR/AMDGPUDialect.h"],
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
includes = ["include"],
deps = [
":AMDGPUIncGen",
[bazel] Port d6abdf46bc
2022-11-21 18:18:25 +01:00
":ArithDialect",
[bazel] Port 499abb243c
2023-02-09 21:35:08 +01:00
":GPUDialect",
[mlir][linalg] Add lowering of named ops on complex numbers This lets linalg.dot and friends lower to a complex muladd using ops from the complex dialect. Differential Revision: https://reviews.llvm.org/D125461
2022-05-12 13:35:27 +02:00
":IR",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
":SideEffectInterfaces",
"//llvm:Core",
"//llvm:Support",
],
)
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
gentbl_cc_library(
name = "AMDGPUPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=AMDGPU",
],
"include/mlir/Dialect/AMDGPU/Transforms/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/AMDGPU/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "AMDGPUTransforms",
srcs = glob(
[
"lib/Dialect/AMDGPU/Transforms/*.cpp",
"lib/Dialect/AMDGPU/Transforms/*.h",
],
),
hdrs = glob(["include/mlir/Dialect/AMDGPU/Transforms/*.h"]),
includes = ["include"],
deps = [
":AMDGPUDialect",
":AMDGPUPassIncGen",
":AMDGPUUtils",
":ArithDialect",
":ControlFlowDialect",
":IR",
":Pass",
":TransformUtils",
":Transforms",
],
)
cc_library(
name = "AMDGPUUtils",
srcs = glob(["lib/Dialect/AMDGPU/Utils/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/AMDGPU/Utils/*.h"]),
includes = ["include"],
deps = [
":AMDGPUDialect",
":Support",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
##---------------------------------------------------------------------------##
# EmitC dialect.
##---------------------------------------------------------------------------##
td_library(
name = "EmitCTdFiles",
srcs = [
"include/mlir/Dialect/EmitC/IR/EmitC.td",
"include/mlir/Dialect/EmitC/IR/EmitCAttributes.td",
"include/mlir/Dialect/EmitC/IR/EmitCBase.td",
"include/mlir/Dialect/EmitC/IR/EmitCTypes.td",
],
includes = ["include"],
deps = [
[mlir][bazel] Fix bazel build files This adds a missing dependency on builtin dialect td files. Differential Revision: https://reviews.llvm.org/D130890
2022-08-01 11:40:39 +02:00
":BuiltinDialectTdFiles",
[bazel] Port 84fe39a45b73
2022-04-28 18:29:43 +02:00
":CastInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "EmitCAttributesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["--gen-attrdef-decls"],
"include/mlir/Dialect/EmitC/IR/EmitCAttributes.h.inc",
),
(
["--gen-attrdef-defs"],
"include/mlir/Dialect/EmitC/IR/EmitCAttributes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/EmitC/IR/EmitCAttributes.td",
deps = [":EmitCTdFiles"],
)
gentbl_cc_library(
name = "EmitCOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-dialect-decls"],
"include/mlir/Dialect/EmitC/IR/EmitCDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
"include/mlir/Dialect/EmitC/IR/EmitCDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-decls"],
"include/mlir/Dialect/EmitC/IR/EmitC.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/EmitC/IR/EmitC.cpp.inc",
),
(
["-gen-typedef-decls"],
"include/mlir/Dialect/EmitC/IR/EmitCTypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/EmitC/IR/EmitCTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/EmitC/IR/EmitC.td",
deps = [":EmitCTdFiles"],
)
[Bazel] Update for Cpp emitter (2f0750dd2e) Update the Bazel build for https://github.com/llvm/llvm-project/commit/2f0750dd2e. Differential Revision: https://reviews.llvm.org/D109171
2021-09-02 10:04:48 -07:00
cc_library(
name = "TargetCpp",
srcs = glob([
"lib/Target/Cpp/*.cpp",
"lib/Target/Cpp/*.h",
]),
hdrs = glob(["include/mlir/Target/Cpp/*.h"]),
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":EmitCDialect",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[Bazel] Update for Cpp emitter (2f0750dd2e) Update the Bazel build for https://github.com/llvm/llvm-project/commit/2f0750dd2e. Differential Revision: https://reviews.llvm.org/D109171
2021-09-02 10:04:48 -07:00
":IR",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":MathDialect",
[Bazel] Update for Cpp emitter (2f0750dd2e) Update the Bazel build for https://github.com/llvm/llvm-project/commit/2f0750dd2e. Differential Revision: https://reviews.llvm.org/D109171
2021-09-02 10:04:48 -07:00
":SCFDialect",
":Support",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":TranslateLib",
[Bazel] Add explicit dependency on llvm:Support to reflect layering Differential Revision: https://reviews.llvm.org/D109173
2021-09-02 10:31:06 -07:00
"//llvm:Support",
[Bazel] Update for Cpp emitter (2f0750dd2e) Update the Bazel build for https://github.com/llvm/llvm-project/commit/2f0750dd2e. Differential Revision: https://reviews.llvm.org/D109171
2021-09-02 10:04:48 -07:00
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
##---------------------------------------------------------------------------##
# Async dialect.
##---------------------------------------------------------------------------##
td_library(
name = "AsyncOpsTdFiles",
srcs = [
"include/mlir/Dialect/Async/IR/AsyncDialect.td",
"include/mlir/Dialect/Async/IR/AsyncOps.td",
"include/mlir/Dialect/Async/IR/AsyncTypes.td",
],
includes = ["include"],
deps = [
Add Async Function to the Async Dialect Add Async Function to the Async Dialect Today `async.execute` operation semantics requires attached region to be executed in a thread managed by the runtime, and always returns an `!async.token` result. We need to model async functions that are not necessarily executed in a runtime-managed threads, but eventually lowered to llvm coroutines. Example: ``` async.func @foo(%arg0: !async.value<f32>) -> !async.token { %0 = async.await %arg0: !async.value<f32> "do_something_with_f32"(%0) return } ``` If `arg0` is available this function will be executed in the caller thread. If it's not available it will be suspended and resumed later later on a thread managed by the async runtime. Currently this is not representable with `async.execute` operations. The longer term goal is to make async dialect more like https://github.com/lewissbaker/cppcoro to be able to represent structured host concurrency in MLIR. (1) Add async.func, async.call, and async.return operations in Async Dialect Reviewed By: ezhulenev, rriddle Differential Revision: https://reviews.llvm.org/D137189
2022-11-02 11:27:26 -07:00
":CallInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ControlFlowInterfacesTdFiles",
Add Async Function to the Async Dialect Add Async Function to the Async Dialect Today `async.execute` operation semantics requires attached region to be executed in a thread managed by the runtime, and always returns an `!async.token` result. We need to model async functions that are not necessarily executed in a runtime-managed threads, but eventually lowered to llvm coroutines. Example: ``` async.func @foo(%arg0: !async.value<f32>) -> !async.token { %0 = async.await %arg0: !async.value<f32> "do_something_with_f32"(%0) return } ``` If `arg0` is available this function will be executed in the caller thread. If it's not available it will be suspended and resumed later later on a thread managed by the async runtime. Currently this is not representable with `async.execute` operations. The longer term goal is to make async dialect more like https://github.com/lewissbaker/cppcoro to be able to represent structured host concurrency in MLIR. (1) Add async.func, async.call, and async.return operations in Async Dialect Reviewed By: ezhulenev, rriddle Differential Revision: https://reviews.llvm.org/D137189
2022-11-02 11:27:26 -07:00
":FunctionInterfacesTdFiles",
[async] Get the number of worker threads from the runtime. Reviewed By: ezhulenev Differential Revision: https://reviews.llvm.org/D117751
2022-01-31 12:00:03 -08:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "AsyncOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Async/IR/AsyncOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Async/IR/AsyncOps.cpp.inc",
),
(
["-gen-dialect-decls"],
"include/mlir/Dialect/Async/IR/AsyncOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
"include/mlir/Dialect/Async/IR/AsyncOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-typedef-decls"],
"include/mlir/Dialect/Async/IR/AsyncOpsTypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/Async/IR/AsyncOpsTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Async/IR/AsyncOps.td",
deps = [":AsyncOpsTdFiles"],
)
gentbl_cc_library(
name = "AsyncPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Async",
],
"include/mlir/Dialect/Async/Passes.h.inc",
),
(
[
"-gen-pass-capi-header",
"--prefix=Async",
],
"include/mlir/Dialect/Async/Passes.capi.h.inc",
),
(
[
"-gen-pass-capi-impl",
"--prefix=Async",
],
"include/mlir/Dialect/Async/Passes.capi.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Async/Passes.td",
deps = [":PassBaseTdFiles"],
)
##---------------------------------------------------------------------------##
# ArmNeon dialect.
##---------------------------------------------------------------------------##
td_library(
name = "ArmNeonTdFiles",
srcs = ["include/mlir/Dialect/ArmNeon/ArmNeon.td"],
includes = ["include"],
deps = [
":LLVMOpsTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "ArmNeonIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=arm_neon",
],
"include/mlir/Dialect/ArmNeon/ArmNeonDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=arm_neon",
],
"include/mlir/Dialect/ArmNeon/ArmNeonDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-decls"],
"include/mlir/Dialect/ArmNeon/ArmNeon.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/ArmNeon/ArmNeon.cpp.inc",
),
(
["-gen-op-doc"],
"g3doc/Dialects/ArmNeon/ArmNeon.md",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/ArmNeon/ArmNeon.td",
deps = [":ArmNeonTdFiles"],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "ArmNeonDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = ["lib/Dialect/ArmNeon/IR/ArmNeonDialect.cpp"],
hdrs = ["include/mlir/Dialect/ArmNeon/ArmNeonDialect.h"],
includes = ["include"],
deps = [
":ArmNeonIncGen",
":IR",
":SideEffectInterfaces",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "ArmNeonConversionIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-llvmir-conversions"],
"include/mlir/Dialect/ArmNeon/ArmNeonConversions.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/ArmNeon/ArmNeon.td",
deps = [":ArmNeonTdFiles"],
)
cc_library(
name = "ArmNeon2dToIntr",
srcs = glob([
"lib/Conversion/ArmNeon2dToIntr/*.cpp",
"lib/Conversion/ArmNeon2dToIntr/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/ArmNeon2dToIntr/*.h",
]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ArmNeonDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":MemRefDialect",
":OpenACCDialect",
":Pass",
":SCFDialect",
":Support",
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
Fix build file for https://github.com/llvm/llvm-project/commit/12648492998bd22d268eb1d4d476c6c3acc6c43d Differential Revision: https://reviews.llvm.org/D151427
2023-05-25 13:36:55 +02:00
##---------------------------------------------------------------------------##
# ArmSME dialect.
##---------------------------------------------------------------------------##
[Bazel] Fix for 7a2fdc685f609730af29e5e969843e9eb71a184c
2023-06-14 23:11:53 +00:00
td_library(
name = "ArmSMETdFiles",
srcs = [
"include/mlir/Dialect/ArmSME/IR/ArmSME.td",
],
includes = ["include"],
deps = [
":ArithOpsTdFiles",
":FuncTdFiles",
":LLVMOpsTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
Fix build file for https://github.com/llvm/llvm-project/commit/12648492998bd22d268eb1d4d476c6c3acc6c43d Differential Revision: https://reviews.llvm.org/D151427
2023-05-25 13:36:55 +02:00
gentbl_cc_library(
name = "ArmSMETransformsPassIncGen",
strip_include_prefix = "include",
tbl_outs = [(
[
"-gen-pass-decls",
"-name=ArmSME",
],
"include/mlir/Dialect/ArmSME/Transforms/Passes.h.inc",
)],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/ArmSME/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
[Bazel] Fix for 7a2fdc685f609730af29e5e969843e9eb71a184c
2023-06-14 23:11:53 +00:00
gentbl_cc_library(
name = "ArmSMEIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/ArmSME/IR/ArmSME.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/ArmSME/IR/ArmSME.cpp.inc",
),
(
["-gen-typedef-decls"],
"include/mlir/Dialect/ArmSME/IR/ArmSMETypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/ArmSME/IR/ArmSMETypes.cpp.inc",
),
(
[
"-gen-dialect-decls",
"-dialect=arm_sme",
],
"include/mlir/Dialect/ArmSME/IR/ArmSMEDialect.h.inc",
),
(
[
"-gen-dialect-defs",
"-dialect=arm_sme",
],
"include/mlir/Dialect/ArmSME/IR/ArmSMEDialect.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/ArmSME/IR/ArmSME.td",
deps = [":ArmSMETdFiles"],
)
gentbl_cc_library(
name = "ArmSMEConversionIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-llvmir-conversions"],
"include/mlir/Dialect/ArmSME/IR/ArmSMEConversions.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/ArmSME/IR/ArmSME.td",
deps = [":ArmSMETdFiles"],
)
cc_library(
name = "ArmSMEDialect",
srcs = ["lib/Dialect/ArmSME/IR/ArmSME.cpp"],
hdrs = ["include/mlir/Dialect/ArmSME/IR/ArmSME.h"],
includes = ["include"],
deps = [
":ArmSMEIncGen",
":IR",
":LLVMDialect",
[mlir][Bazel] Add missing dependencies after 564713c47175d9f61fe8e18e750fb7f7b486d533
2023-07-03 13:16:28 +02:00
":SCFDialect",
[Bazel] Fix for 7a2fdc685f609730af29e5e969843e9eb71a184c
2023-06-14 23:11:53 +00:00
":SideEffectInterfaces",
[mlir][Bazel] Add missing dependency.
2023-07-18 15:51:49 +02:00
":VectorDialect",
[Bazel] Fix for 7a2fdc685f609730af29e5e969843e9eb71a184c
2023-06-14 23:11:53 +00:00
"//llvm:Core",
"//llvm:Support",
],
)
[bazel][mlir] Port ca9a3354d04b15366088d7831b40f891e3d77b95
2023-07-25 11:38:35 +02:00
cc_library(
name = "ArmSMEUtils",
srcs = glob(["lib/Dialect/ArmSME/Utils/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/ArmSME/Utils/*.h"]),
includes = ["include"],
deps = [
":ArmSMEDialect",
":IR",
":Dialect",
],
)
Fix build file for https://github.com/llvm/llvm-project/commit/12648492998bd22d268eb1d4d476c6c3acc6c43d Differential Revision: https://reviews.llvm.org/D151427
2023-05-25 13:36:55 +02:00
cc_library(
name = "ArmSMETransforms",
srcs = glob(["lib/Dialect/ArmSME/Transforms/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/ArmSME/Transforms/*.h"]),
includes = ["include"],
deps = [
[mlir][Bazel] Add missing dependencies after 564713c47175d9f61fe8e18e750fb7f7b486d533
2023-07-03 13:16:28 +02:00
":ArithDialect",
[bazel][mlir] Port 65305aeab99ad8ea09dd85e28a41c657152a08fb
2023-06-16 13:20:32 +02:00
":ArmSMEDialect",
[bazel][mlir] Port ca9a3354d04b15366088d7831b40f891e3d77b95
2023-07-25 11:38:35 +02:00
":ArmSMEUtils",
Fix build file for https://github.com/llvm/llvm-project/commit/12648492998bd22d268eb1d4d476c6c3acc6c43d Differential Revision: https://reviews.llvm.org/D151427
2023-05-25 13:36:55 +02:00
":ArmSMETransformsPassIncGen",
":FuncDialect",
[mlir][Bazel] Add missing dependencies after 564713c47175d9f61fe8e18e750fb7f7b486d533
2023-07-03 13:16:28 +02:00
":IR",
[bazel][mlir] Port 65305aeab99ad8ea09dd85e28a41c657152a08fb
2023-06-16 13:20:32 +02:00
":LLVMCommonConversion",
[mlir][Bazel] Add missing dependencies after 564713c47175d9f61fe8e18e750fb7f7b486d533
2023-07-03 13:16:28 +02:00
":LLVMDialect",
Fix build file for https://github.com/llvm/llvm-project/commit/12648492998bd22d268eb1d4d476c6c3acc6c43d Differential Revision: https://reviews.llvm.org/D151427
2023-05-25 13:36:55 +02:00
":Pass",
[mlir][Bazel] Add missing dependencies after 564713c47175d9f61fe8e18e750fb7f7b486d533
2023-07-03 13:16:28 +02:00
":SCFDialect",
[mlir][Bazel] Add missing dependency.
2023-07-18 16:10:48 +02:00
":Transforms",
[mlir][Bazel] Add missing dependencies after 564713c47175d9f61fe8e18e750fb7f7b486d533
2023-07-03 13:16:28 +02:00
":VectorDialect",
Fix build file for https://github.com/llvm/llvm-project/commit/12648492998bd22d268eb1d4d476c6c3acc6c43d Differential Revision: https://reviews.llvm.org/D151427
2023-05-25 13:36:55 +02:00
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
##---------------------------------------------------------------------------##
# ArmSVE dialect.
##---------------------------------------------------------------------------##
td_library(
name = "ArmSVETdFiles",
srcs = [
"include/mlir/Dialect/ArmSVE/ArmSVE.td",
],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithOpsTdFiles",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMOpsTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "ArmSVEIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/ArmSVE/ArmSVE.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/ArmSVE/ArmSVE.cpp.inc",
),
(
["-gen-typedef-decls"],
"include/mlir/Dialect/ArmSVE/ArmSVETypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/ArmSVE/ArmSVETypes.cpp.inc",
),
(
[
"-gen-dialect-decls",
"-dialect=arm_sve",
],
"include/mlir/Dialect/ArmSVE/ArmSVEDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=arm_sve",
],
"include/mlir/Dialect/ArmSVE/ArmSVEDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/ArmSVE/ArmSVE.td",
deps = [":ArmSVETdFiles"],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "ArmSVEDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = ["lib/Dialect/ArmSVE/IR/ArmSVEDialect.cpp"],
hdrs = ["include/mlir/Dialect/ArmSVE/ArmSVEDialect.h"],
includes = ["include"],
deps = [
":ArmSVEIncGen",
":IR",
":LLVMDialect",
":SideEffectInterfaces",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "ArmSVETransforms",
srcs = glob(["lib/Dialect/ArmSVE/Transforms/*.cpp"]),
hdrs = ["include/mlir/Dialect/ArmSVE/Transforms.h"],
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ArmSVEDialect",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
":TransformUtils",
"//llvm:Core",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "ArmSVEConversionIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-llvmir-conversions"],
"include/mlir/Dialect/ArmSVE/ArmSVEConversions.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/ArmSVE/ArmSVE.td",
deps = [":ArmSVETdFiles"],
)
##---------------------------------------------------------------------------##
# AMX dialect.
##---------------------------------------------------------------------------##
td_library(
name = "AMXTdFiles",
srcs = ["include/mlir/Dialect/AMX/AMX.td"],
includes = ["include"],
deps = [
":LLVMOpsTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "AMXIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=amx",
],
"include/mlir/Dialect/AMX/AMXDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=amx",
],
"include/mlir/Dialect/AMX/AMXDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-decls"],
"include/mlir/Dialect/AMX/AMX.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/AMX/AMX.cpp.inc",
),
(
["-gen-op-doc"],
"g3doc/Dialects/AMX/AMX.md",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/AMX/AMX.td",
deps = [":AMXTdFiles"],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "AMXDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = ["lib/Dialect/AMX/IR/AMXDialect.cpp"],
hdrs = ["include/mlir/Dialect/AMX/AMXDialect.h"],
includes = ["include"],
deps = [
":AMXIncGen",
":IR",
":LLVMDialect",
":SideEffectInterfaces",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "AMXTransforms",
srcs = glob(["lib/Dialect/AMX/Transforms/*.cpp"]),
hdrs = ["include/mlir/Dialect/AMX/Transforms.h"],
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AMXDialect",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
"//llvm:Core",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "AMXConversionIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-llvmir-conversions"],
"include/mlir/Dialect/AMX/AMXConversions.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/AMX/AMX.td",
deps = [":AMXTdFiles"],
)
##---------------------------------------------------------------------------##
# X86Vector dialect.
##---------------------------------------------------------------------------##
td_library(
name = "X86VectorTdFiles",
srcs = ["include/mlir/Dialect/X86Vector/X86Vector.td"],
includes = ["include"],
deps = [
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMOpsTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "X86VectorIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=x86vector",
],
"include/mlir/Dialect/X86Vector/X86VectorDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=x86vector",
],
"include/mlir/Dialect/X86Vector/X86VectorDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-decls"],
"include/mlir/Dialect/X86Vector/X86Vector.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/X86Vector/X86Vector.cpp.inc",
),
(
["-gen-op-doc"],
"g3doc/Dialects/X86Vector/X86Vector.md",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/X86Vector/X86Vector.td",
deps = [":X86VectorTdFiles"],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "X86VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = ["lib/Dialect/X86Vector/IR/X86VectorDialect.cpp"],
hdrs = ["include/mlir/Dialect/X86Vector/X86VectorDialect.h"],
includes = ["include"],
deps = [
":IR",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
":SideEffectInterfaces",
":X86VectorIncGen",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "X86VectorTransforms",
srcs = glob(["lib/Dialect/X86Vector/Transforms/*.cpp"]),
hdrs = ["include/mlir/Dialect/X86Vector/Transforms.h"],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir][vector] Generalize vector.transpose lowering to n-D vectors The existing vector.transpose lowering patterns only triggers if the input vector is 2D. The revision extends the pattern to handle n-D vectors which are effectively 2-D vectors (e.g., vector<1x4x1x8x1). It refactors a common check about 2-D vectors from X86Vector lowering to VectorUtils.h so it can be reused by both sides. Reviewed By: dcaballe Differential Revision: https://reviews.llvm.org/D149908
2023-05-04 14:31:44 -07:00
":VectorUtils",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":X86VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "X86VectorConversionIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-llvmir-conversions"],
"include/mlir/Dialect/X86Vector/X86VectorConversions.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/X86Vector/X86Vector.td",
deps = [":X86VectorTdFiles"],
)
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
##---------------------------------------------------------------------------##
# IRDL dialect.
##---------------------------------------------------------------------------##
td_library(
name = "IRDLTdFiles",
srcs = [
"include/mlir/Dialect/IRDL/IR/IRDL.td",
[mlir][irdl] Fix BUILD files. https://reviews.llvm.org/D153983 introduced a new .td file with attributes and enums. This patch adds the corresponding rules to the BUILD file in order to fix the bazel build. Reviewed By: ingomueller-net Differential Revision: https://reviews.llvm.org/D156355
2023-07-26 18:11:44 +00:00
"include/mlir/Dialect/IRDL/IR/IRDLAttributes.td",
[bazel] Port c8a581c331f27a1ece8e42206831e56b7a222d26
2023-05-17 14:50:01 +02:00
"include/mlir/Dialect/IRDL/IR/IRDLInterfaces.td",
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
"include/mlir/Dialect/IRDL/IR/IRDLOps.td",
"include/mlir/Dialect/IRDL/IR/IRDLTypes.td",
],
includes = ["include"],
deps = [
":InferTypeOpInterfaceTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
],
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
)
gentbl_cc_library(
name = "IRDLIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-dialect-decls"],
"include/mlir/Dialect/IRDL/IR/IRDLDialect.h.inc",
),
(
["-gen-dialect-defs"],
"include/mlir/Dialect/IRDL/IR/IRDLDialect.cpp.inc",
),
(
["-gen-op-decls"],
"include/mlir/Dialect/IRDL/IR/IRDL.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/IRDL/IR/IRDL.cpp.inc",
),
(
["-gen-typedef-decls"],
"include/mlir/Dialect/IRDL/IR/IRDLTypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/IRDL/IR/IRDLTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/IRDL/IR/IRDLOps.td",
deps = [":IRDLTdFiles"],
)
[bazel] Port c8a581c331f27a1ece8e42206831e56b7a222d26
2023-05-17 14:50:01 +02:00
gentbl_cc_library(
name = "IRDLInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Dialect/IRDL/IR/IRDLInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Dialect/IRDL/IR/IRDLInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/IRDL/IR/IRDLInterfaces.td",
deps = [":IRDLTdFiles"],
)
[mlir][irdl] Fix BUILD files. https://reviews.llvm.org/D153983 introduced a new .td file with attributes and enums. This patch adds the corresponding rules to the BUILD file in order to fix the bazel build. Reviewed By: ingomueller-net Differential Revision: https://reviews.llvm.org/D156355
2023-07-26 18:11:44 +00:00
gentbl_cc_library(
name = "IRDLAttributesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/IRDL/IR/IRDLAttributes.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/IRDL/IR/IRDLAttributes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/IRDL/IR/IRDLAttributes.td",
deps = [":IRDLTdFiles"],
)
gentbl_cc_library(
name = "IRDLEnumsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-enum-decls"],
"include/mlir/Dialect/IRDL/IR/IRDLEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/IRDL/IR/IRDLEnums.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/IRDL/IR/IRDLAttributes.td",
deps = [":IRDLTdFiles"],
)
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
gentbl_cc_library(
name = "IRDLOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/IRDL/IR/IRDLOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/IRDL/IR/IRDLOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/IRDL/IR/IRDLOps.td",
deps = [":IRDLTdFiles"],
)
gentbl_cc_library(
name = "IRDLTypesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-typedef-decls"],
"include/mlir/Dialect/IRDL/IR/IRDLTypesGen.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/IRDL/IR/IRDLTypesGen.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/IRDL/IR/IRDLTypes.td",
deps = [":IRDLTdFiles"],
)
cc_library(
name = "IRDLDialect",
[bazel] Port for 8ac8c922fb3f15706f5cb1db2cc655d30b095766
2023-04-23 23:54:29 +02:00
srcs = [
"lib/Dialect/IRDL/IR/IRDL.cpp",
[bazel] Port c8a581c331f27a1ece8e42206831e56b7a222d26
2023-05-17 14:50:01 +02:00
"lib/Dialect/IRDL/IR/IRDLOps.cpp",
[bazel] Port for 8ac8c922fb3f15706f5cb1db2cc655d30b095766
2023-04-23 23:54:29 +02:00
"lib/Dialect/IRDL/IRDLLoading.cpp",
[bazel] Port c8a581c331f27a1ece8e42206831e56b7a222d26
2023-05-17 14:50:01 +02:00
"lib/Dialect/IRDL/IRDLVerifiers.cpp",
[bazel] Port for 8ac8c922fb3f15706f5cb1db2cc655d30b095766
2023-04-23 23:54:29 +02:00
],
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
hdrs = [
"include/mlir/Dialect/IRDL/IR/IRDL.h",
[bazel] Port c8a581c331f27a1ece8e42206831e56b7a222d26
2023-05-17 14:50:01 +02:00
"include/mlir/Dialect/IRDL/IR/IRDLInterfaces.h",
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
"include/mlir/Dialect/IRDL/IR/IRDLTraits.h",
[bazel] Port for 8ac8c922fb3f15706f5cb1db2cc655d30b095766
2023-04-23 23:54:29 +02:00
"include/mlir/Dialect/IRDL/IRDLLoading.h",
[bazel] Port c8a581c331f27a1ece8e42206831e56b7a222d26
2023-05-17 14:50:01 +02:00
"include/mlir/Dialect/IRDL/IRDLVerifiers.h",
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
],
includes = ["include"],
deps = [
":Dialect",
":IR",
[mlir][irdl] Fix BUILD files. https://reviews.llvm.org/D153983 introduced a new .td file with attributes and enums. This patch adds the corresponding rules to the BUILD file in order to fix the bazel build. Reviewed By: ingomueller-net Differential Revision: https://reviews.llvm.org/D156355
2023-07-26 18:11:44 +00:00
":IRDLAttributesIncGen",
":IRDLEnumsIncGen",
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
":IRDLIncGen",
[bazel] Port c8a581c331f27a1ece8e42206831e56b7a222d26
2023-05-17 14:50:01 +02:00
":IRDLInterfacesIncGen",
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
":IRDLOpsIncGen",
":IRDLTypesIncGen",
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
":InferTypeOpInterface",
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
":Support",
"//llvm:Core",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
##---------------------------------------------------------------------------##
# SCF dialect.
##---------------------------------------------------------------------------##
td_library(
name = "SCFTdFiles",
[mlir] Introduce device mapper attribute for `thread_dim_map` and `mapped to dims` `scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [0, 1]} } { thread_dim_mapping = [0, 1]} ``` It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation. The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]} } { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]} ``` Reviewed By: ftynse, nicolasvasilache Differential Revision: https://reviews.llvm.org/D137413
2022-11-10 17:55:49 +01:00
srcs = [
"include/mlir/Dialect/SCF/IR/DeviceMappingInterface.td",
"include/mlir/Dialect/SCF/IR/SCFOps.td",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":ControlFlowInterfacesTdFiles",
[MLIR] Add InferTypeOpInterface to scf.if op Differential Revision: https://reviews.llvm.org/D142049
2023-01-19 13:18:22 -05:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LoopLikeInterfaceTdFiles",
[fix/build] bazel rule for ParallelCombiningOpInterface
2022-07-01 10:33:48 +02:00
":ParallelCombiningOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfacesTdFiles",
Fix bazel build after 59b273a166b9d1165015b5cf10c21641cc01d682. Reviewed By: tpopp Differential Revision: https://reviews.llvm.org/D126765
2022-06-01 12:06:28 +02:00
":ViewLikeInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
gentbl_cc_library(
name = "SCFIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
"include/mlir/Dialect/SCF/IR/SCFOps.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-op-defs"],
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
"include/mlir/Dialect/SCF/IR/SCFOps.cpp.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-dialect-decls"],
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
"include/mlir/Dialect/SCF/IR/SCFOpsDialect.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
"include/mlir/Dialect/SCF/IR/SCFOpsDialect.cpp.inc",
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
td_file = "include/mlir/Dialect/SCF/IR/SCFOps.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [":SCFTdFiles"],
)
[mlir] Introduce device mapper attribute for `thread_dim_map` and `mapped to dims` `scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [0, 1]} } { thread_dim_mapping = [0, 1]} ``` It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation. The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]} } { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]} ``` Reviewed By: ftynse, nicolasvasilache Differential Revision: https://reviews.llvm.org/D137413
2022-11-10 17:55:49 +01:00
gentbl_cc_library(
name = "SCFDeviceMappingInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-attr-interface-decls"],
"include/mlir/Dialect/SCF/IR/DeviceMappingAttrInterface.h.inc",
),
(
["-gen-attr-interface-defs"],
"include/mlir/Dialect/SCF/IR/DeviceMappingAttrInterface.cpp.inc",
),
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/SCF/IR/DeviceMappingAttributes.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/SCF/IR/DeviceMappingAttributes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SCF/IR/DeviceMappingInterface.td",
deps = [":SCFTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "SCFPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=SCF",
],
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
"include/mlir/Dialect/SCF/Transforms/Passes.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
],
tblgen = ":mlir-tblgen",
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
td_file = "include/mlir/Dialect/SCF/Transforms/Passes.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "SCFTransforms",
srcs = glob([
"lib/Dialect/SCF/Transforms/*.cpp",
"lib/Dialect/SCF/Transforms/*.h",
]),
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
hdrs = glob([
"include/mlir/Dialect/SCF/Transforms/*.h",
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
":AffineAnalysis",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationDialect",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":BufferizationTransforms",
[mlir][interfaces] Remove getDestinationOperands from TilingInterface `getDestinationOperands` was almost a duplicate of `DestinationStyleOpInterface::getOutputOperands`. Now that the interface has been moved to mlir/Interfaces, it is no longer needed. Differential Revision: https://reviews.llvm.org/D136240
2022-10-24 09:17:44 +02:00
":DestinationStyleOpInterface",
[mlir][SCF] Peel scf.for loops for even step divison Add ForLoopBoundSpecialization pass, which specializes scf.for loops into a "main loop" where `step` divides the iteration space evenly and into an scf.if that handles the last iteration. This transformation is useful for vectorization and loop tiling. E.g., when vectorizing loads/stores, programs will spend most of their time in the main loop, in which only unmasked loads/stores are used. Only the in the last iteration (scf.if), slower masked loads/stores are used. Subsequent commits will apply this transformation in the SparseDialect and in Linalg's loop tiling. Differential Revision: https://reviews.llvm.org/D105804
2021-08-03 10:21:02 +09:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":MemRefDialect",
":Pass",
":SCFDialect",
":SCFPassIncGen",
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
":SCFUtils",
[mlir] Remove `Transforms/SideEffectUtils.h` and move the methods into `Interface/SideEffectInterfaces.h`. The methods in `SideEffectUtils.h` (and their implementations in `SideEffectUtils.cpp`) seem to have similar intent to methods already existing in `SideEffectInterfaces.h`. Move the decleration (and implementation) from `SideEffectUtils.h` (and `SideEffectUtils.cpp`) into `SideEffectInterfaces.h` (and `SideEffectInterface.cpp`). Also drop the `SideEffectInterface::hasNoEffect` method in favor of `mlir::isMemoryEffectFree` which actually recurses into the operation instead of just relying on the `hasRecursiveMemoryEffectTrait` exclusively. Differential Revision: https://reviews.llvm.org/D137857
2022-11-11 17:10:38 +00:00
":SideEffectInterfaces",
[mlir] Fix bazel build Differential Revision: https://reviews.llvm.org/D106311
2021-07-19 14:00:51 -07:00
":Support",
[mlir][SCF] Canonicalize dim(x) where x is an iter_arg * Add `DimOfIterArgFolder`. * Move existing cross-dialect canonicalization patterns to `LoopCanonicalization.cpp`. * Rename `SCFAffineOpCanonicalization` pass to `SCFForLoopCanonicalization`. * Expand documentaton of scf.for: The type of loop-carried variables may not change with iterations. (Not even the dynamic type.) Differential Revision: https://reviews.llvm.org/D108806
2021-08-30 01:12:14 +00:00
":TensorDialect",
[mlir][TilingInterface] Enable tile and fuse using TilingInterface. This patch implements tile and fuse transformation for ops that implement the tiling interface. To do so, - `TilingInterface` needs a new method that generates a tiled implementation of the operation based on the tile of the result needed. - A pattern is added that replaces a `tensor.extract_slice` whose source is defined by an operation that implements the `TilingInterface` with a tiled implementation that produces the extracted slice in-place (using the method added to `TilingInterface`). - A pattern is added that takes a sequence of operations that implement the `TilingInterface` (for now `LinalgOp`s), tiles the consumer, and greedily fuses its producers iteratively. Differential Revision: https://reviews.llvm.org/D127809
2022-06-21 16:59:26 +00:00
":TensorTransforms",
[mlir][TilingInterface] Add pattern to tile using TilingInterface and implement TilingInterface for Linalg ops. This patch adds support for tiling operations that implement the TilingInterface. - It separates the loop constructs that are used to iterate over tile from the implementation of the tiling itself. For example, the use of destructive updates is more related to use of scf.for for iterating over tiles that are tensors. - To test the transformation, TilingInterface is implemented for LinalgOps. The separation of the looping constructs used from the implementation of tile code generation greatly simplifies the latter. - The implementation of TilingInterface for LinalgOp is kept as an external model for now till this approach can be fully flushed out to replace the existing tiling + fusion approaches in Linalg. Differential Revision: https://reviews.llvm.org/D127133
2022-06-13 19:56:32 +00:00
":TilingInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
"//llvm:Support",
],
)
[mlir] Introduce Transform ops for loops Introduce transform ops for "for" loops, in particular for peeling, software pipelining and unrolling, along with a couple of "IR navigation" ops. These ops are intended to be generalized to different kinds of loops when possible and therefore use the "loop" prefix. They currently live in the SCF dialect as there is no clear place to put transform ops that may span across several dialects, this decision is postponed until the ops actually need to handle non-SCF loops. Additionally refactor some common utilities for transform ops into trait or interface methods, and change the loop pipelining to be a returning pattern. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D127300
2022-06-09 11:10:32 +02:00
td_library(
name = "SCFTransformOpsTdFiles",
srcs = [
"include/mlir/Dialect/SCF/TransformOps/SCFTransformOps.td",
],
includes = ["include"],
deps = [
":PDLDialect",
":TransformDialectTdFiles",
],
)
gentbl_cc_library(
name = "SCFTransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/SCF/TransformOps/SCFTransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/SCF/TransformOps/SCFTransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SCF/TransformOps/SCFTransformOps.td",
deps = [
":SCFTransformOpsTdFiles",
],
)
cc_library(
name = "SCFTransformOps",
srcs = glob(["lib/Dialect/SCF/TransformOps/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/SCF/TransformOps/*.h"]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Fix bazel build for https://reviews.llvm.org/D138980
2022-11-30 10:16:34 -08:00
":AffineUtils",
[mlir] Introduce Transform ops for loops Introduce transform ops for "for" loops, in particular for peeling, software pipelining and unrolling, along with a couple of "IR navigation" ops. These ops are intended to be generalized to different kinds of loops when possible and therefore use the "loop" prefix. They currently live in the SCF dialect as there is no clear place to put transform ops that may span across several dialects, this decision is postponed until the ops actually need to handle non-SCF loops. Additionally refactor some common utilities for transform ops into trait or interface methods, and change the loop pipelining to be a returning pattern. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D127300
2022-06-09 11:10:32 +02:00
":FuncDialect",
":IR",
[mlir][SCF] Add `loop.promote_if_one_iteration` transform op This transform op promotes loops with one iteration. I.e., the loop op is replaced by just the loop body. Differential Revision: https://reviews.llvm.org/D154361
2023-07-04 08:57:55 +02:00
":LoopLikeInterface",
[mlir] Introduce Transform ops for loops Introduce transform ops for "for" loops, in particular for peeling, software pipelining and unrolling, along with a couple of "IR navigation" ops. These ops are intended to be generalized to different kinds of loops when possible and therefore use the "loop" prefix. They currently live in the SCF dialect as there is no clear place to put transform ops that may span across several dialects, this decision is postponed until the ops actually need to handle non-SCF loops. Additionally refactor some common utilities for transform ops into trait or interface methods, and change the loop pipelining to be a returning pattern. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D127300
2022-06-09 11:10:32 +02:00
":SCFDialect",
":SCFTransformOpsIncGen",
":SCFTransforms",
":SCFUtils",
":SideEffectInterfaces",
":TransformDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir] Introduce Transform ops for loops Introduce transform ops for "for" loops, in particular for peeling, software pipelining and unrolling, along with a couple of "IR navigation" ops. These ops are intended to be generalized to different kinds of loops when possible and therefore use the "loop" prefix. They currently live in the SCF dialect as there is no clear place to put transform ops that may span across several dialects, this decision is postponed until the ops actually need to handle non-SCF loops. Additionally refactor some common utilities for transform ops into trait or interface methods, and change the loop pipelining to be a returning pattern. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D127300
2022-06-09 11:10:32 +02:00
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
##---------------------------------------------------------------------------##
# SparseTensor dialect.
##---------------------------------------------------------------------------##
td_library(
name = "SparseTensorTdFiles",
srcs = [
"include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrDefs.td",
"include/mlir/Dialect/SparseTensor/IR/SparseTensorBase.td",
"include/mlir/Dialect/SparseTensor/IR/SparseTensorOps.td",
[mlir][sparse] introduce sparse_tensor::StorageSpecifierType and related operations on it Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D139961
2022-12-13 19:49:43 +00:00
"include/mlir/Dialect/SparseTensor/IR/SparseTensorTypes.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "SparseTensorAttrDefsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["--gen-attrdef-decls"],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrDefs.h.inc",
),
(
["--gen-attrdef-defs"],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrDefs.cpp.inc",
),
[mlir][sparse] introduce sparse_tensor::StorageSpecifierType and related operations on it Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D139961
2022-12-13 19:49:43 +00:00
(
["--gen-enum-decls"],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrEnums.h.inc",
),
(
["--gen-enum-defs"],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrEnums.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrDefs.td",
deps = [":SparseTensorTdFiles"],
)
gentbl_cc_library(
name = "SparseTensorOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=sparse_tensor",
],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=sparse_tensor",
],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-decls"],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorOps.cpp.inc",
),
(
["-gen-op-doc"],
"g3doc/Dialects/SparseTensor/SparseTensor.md",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SparseTensor/IR/SparseTensorOps.td",
deps = [":SparseTensorTdFiles"],
)
[mlir][sparse] introduce sparse_tensor::StorageSpecifierType and related operations on it Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D139961
2022-12-13 19:49:43 +00:00
gentbl_cc_library(
name = "SparseTensorTypesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["--gen-typedef-decls"],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorTypes.h.inc",
),
(
["--gen-typedef-defs"],
"include/mlir/Dialect/SparseTensor/IR/SparseTensorTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SparseTensor/IR/SparseTensorTypes.td",
deps = [":SparseTensorTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "SparseTensorPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=SparseTensor",
],
"include/mlir/Dialect/SparseTensor/Transforms/Passes.h.inc",
),
(
[
"-gen-pass-capi-header",
"--prefix=SparseTensor",
],
"include/mlir/Dialect/SparseTensor/Transforms/Passes.capi.h.inc",
),
(
[
"-gen-pass-capi-impl",
"--prefix=SparseTensor",
],
"include/mlir/Dialect/SparseTensor/Transforms/Passes.capi.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SparseTensor/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
[mlir][sparse] Factoring out SparseTensorEnums library This differential splits the SparseTensorEnums library out from the SparseTensorRuntime library. The actual moving of files will be handled in the next differential. Depends On D135996 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136002
2022-10-17 18:13:05 -07:00
# This library is shared by both SparseTensorDialect and
# SparseTensorRuntime, so it must not depend on any of the MLIR/LLVM
# internals or else mlir_c_runner_utils will inherit that dependency.
cc_library(
name = "SparseTensorEnums",
[mlir][sparse] Moving Enums.h into Dialect/SparseTensor/IR Move the SparseTensorEnums library out of the ExecutionEngine directory and into Dialect/SparseTensor/IR. Depends On D136002 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136005
2022-10-17 18:33:40 -07:00
hdrs = ["include/mlir/Dialect/SparseTensor/IR/Enums.h"],
[mlir][sparse] Factoring out SparseTensorEnums library This differential splits the SparseTensorEnums library out from the SparseTensorRuntime library. The actual moving of files will be handled in the next differential. Depends On D135996 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136002
2022-10-17 18:13:05 -07:00
includes = ["include"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "SparseTensorDialect",
[mlir][sparse] Start migration to new surface syntax for STEA We are in the progress of migrating to a much improved surface syntax for the Sparse Tensor Encoding Attribute (STEA). You can see a preview of this in the StableHLO RFC at https://github.com/openxla/stablehlo/blob/main/rfcs/20230210-sparsity.md //**This design is courtesy Wren Romano.**// This initial revision (1) Introduces the first version of a new parser written by Wren Romano (2) Introduces a simple "migration plan" using NEW_SYNTAX on the STEA, which will allow us to test the new parser with new examples, as well as migrate existing examples over without the need to rewrite them all This first "drop" merely provides the entry points to parse the new syntax. The parser is still under active development. For example, we need to address the "lookahead" issue when parsing the lvl spec (viz. do we see l0 = d0 or a direct d0). Another larger task is to actually implement "affine" parsing (since the MLIR affine parser is not accessible in other parts of the tree). EXAMPLE: Currently, CSR looks like #CSR = #sparse_tensor.encoding<{ lvlTypes = ["dense","compressed"], dimToLvl = affine_map<(i,j) -> (i,j)> }> but you can "force" the new parser with #CSR = #sparse_tensor.encoding<{ NEW_SYNTAX = (d0, d1) -> (l0 = d0 : dense, l1 = d1 : compressed) }> Reviewed By: Peiming Differential Revision: https://reviews.llvm.org/D153997
2023-06-28 12:54:12 -07:00
srcs = [
"lib/Dialect/SparseTensor/IR/Detail/DimLvlMap.cpp",
"lib/Dialect/SparseTensor/IR/Detail/DimLvlMap.h",
"lib/Dialect/SparseTensor/IR/Detail/DimLvlMapParser.cpp",
"lib/Dialect/SparseTensor/IR/Detail/DimLvlMapParser.h",
"lib/Dialect/SparseTensor/IR/Detail/LvlTypeParser.cpp",
"lib/Dialect/SparseTensor/IR/Detail/LvlTypeParser.h",
"lib/Dialect/SparseTensor/IR/Detail/TemplateExtras.h",
"lib/Dialect/SparseTensor/IR/Detail/Var.cpp",
"lib/Dialect/SparseTensor/IR/Detail/Var.h",
"lib/Dialect/SparseTensor/IR/SparseTensorDialect.cpp",
],
[mlir][sparse] Factoring out SparseTensorType class This change adds a new `SparseTensorType` class for making the "dim" vs "lvl" distinction more overt, and for abstracting over the differences between sparse-tensors and dense-tensors. In addition, this change also adds new type aliases `Dimension`, `Level`, and `FieldIndex` to make code more self-documenting. Although the diff is very large, the majority of the changes are mechanical in nature (e.g., changing types to use the new aliases, updating variable names to match, etc). Along the way I also made many variables `const` when they could be; the majority of which required only adding the keyword. A few places had conditional definitions of these variables, requiring actual code changes; however, that was only done when the overall change was extremely local and easy to extract. All these changes are included in the current patch only because it would be too onerous to split them off into a separate patch. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D143800
2023-02-14 18:20:45 -08:00
hdrs = [
"include/mlir/Dialect/SparseTensor/IR/SparseTensor.h",
[mlir][sparse] Make sparse_tensor::StorageLayout publicly available. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D150739
2023-05-16 23:52:45 +00:00
"include/mlir/Dialect/SparseTensor/IR/SparseTensorStorageLayout.h",
[mlir][sparse] Factoring out SparseTensorType class This change adds a new `SparseTensorType` class for making the "dim" vs "lvl" distinction more overt, and for abstracting over the differences between sparse-tensors and dense-tensors. In addition, this change also adds new type aliases `Dimension`, `Level`, and `FieldIndex` to make code more self-documenting. Although the diff is very large, the majority of the changes are mechanical in nature (e.g., changing types to use the new aliases, updating variable names to match, etc). Along the way I also made many variables `const` when they could be; the majority of which required only adding the keyword. A few places had conditional definitions of these variables, requiring actual code changes; however, that was only done when the overall change was extremely local and easy to extract. All these changes are included in the current patch only because it would be too onerous to split them off into a separate patch. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D143800
2023-02-14 18:20:45 -08:00
"include/mlir/Dialect/SparseTensor/IR/SparseTensorType.h",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
[mlir][sparse] add print methods to Merger (for debugging) Reviewed By: gussmith23 Differential Revision: https://reviews.llvm.org/D104939
2021-06-25 12:02:08 -07:00
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][NFC] Use `getConstantIntValue` instead of casting to `ConstantIndexOp` `getConstantIntValue` extracts constant values from all constant-like ops, not just `arith::ConstantIndexOp`. Differential Revision: https://reviews.llvm.org/D154356
2023-07-04 14:03:02 +02:00
":DialectUtils",
[mlir][sparse] add print methods to Merger (for debugging) Reviewed By: gussmith23 Differential Revision: https://reviews.llvm.org/D104939
2021-06-25 12:02:08 -07:00
":IR",
[bazel] Port 92a836da0759
2022-04-28 22:51:27 +02:00
":InferTypeOpInterface",
[mlir][sparse] add print methods to Merger (for debugging) Reviewed By: gussmith23 Differential Revision: https://reviews.llvm.org/D104939
2021-06-25 12:02:08 -07:00
":SparseTensorAttrDefsIncGen",
[mlir][sparse] Factoring out SparseTensorEnums library This differential splits the SparseTensorEnums library out from the SparseTensorRuntime library. The actual moving of files will be handled in the next differential. Depends On D135996 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136002
2022-10-17 18:13:05 -07:00
":SparseTensorEnums",
[mlir][sparse] add print methods to Merger (for debugging) Reviewed By: gussmith23 Differential Revision: https://reviews.llvm.org/D104939
2021-06-25 12:02:08 -07:00
":SparseTensorOpsIncGen",
[mlir][sparse] introduce sparse_tensor::StorageSpecifierType and related operations on it Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D139961
2022-12-13 19:49:43 +00:00
":SparseTensorTypesIncGen",
[mlir][sparse] add print methods to Merger (for debugging) Reviewed By: gussmith23 Differential Revision: https://reviews.llvm.org/D104939
2021-06-25 12:02:08 -07:00
"//llvm:Support",
],
)
cc_library(
name = "SparseTensorUtils",
srcs = glob(["lib/Dialect/SparseTensor/Utils/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/SparseTensor/Utils/*.h"]),
includes = ["include"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][sparse] introduce complex type to sparse tensor support This is the first implementation of complex (f64 and f32) support in the sparse compiler, with complex add/mul as first operations. Note that various features are still TBD, such as other ops, and reading in complex values from file. Also, note that the std::complex<float> had a bit of an ABI issue when passed as single argument. It is still TBD if better solutions are possible. Reviewed By: bixia Differential Revision: https://reviews.llvm.org/D125596
2022-05-13 17:54:32 -07:00
":ComplexDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
[Bazel] Add more mlir dependencies after D124298 The Bazel layering_check feature compiles libraries with `-fmodule-name=X -fmodules-strict-decluse` which require #include to be in deps.
2022-04-23 13:06:54 -07:00
":MathDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":SparseTensorDialect",
[mlir][sparse] Removing the DimLvlType and DimLevelFormat types This removes another massive source of redundancy, and instead has the Merger.{h,cpp} reuse the SparseTensorEnums library. Depends On D136005 Reviewed By: Peiming Differential Revision: https://reviews.llvm.org/D136123
2022-10-17 19:11:20 -07:00
":SparseTensorEnums",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
cc_library(
name = "SparseTensorTransforms",
[mlir][sparse] Factoring out Transforms/CodegenUtils.{cpp,h} This moves a bunch of helper functions from `Transforms/SparseTensorConversion.cpp` into `Transforms/CodegenUtils.{cpp,h}` so that they can be reused by `Transforms/Sparsification.cpp`, etc. See also the dependent D115010 which cleans up some corner cases in this change. Reviewed By: aartbik, rriddle Differential Revision: https://reviews.llvm.org/D115008
2022-01-04 15:06:28 -08:00
srcs = glob([
"lib/Dialect/SparseTensor/Transforms/*.cpp",
"lib/Dialect/SparseTensor/Transforms/*.h",
]),
[mlir][sparse] Factoring magic numbers into a header Addresses https://bugs.llvm.org/show_bug.cgi?id=52303 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D112962
2021-11-05 15:15:39 -07:00
hdrs = [
[mlir][sparse][bufferize] Implement BufferizableOpInterface Only the analysis part of the interface is implemented. The bufferization itself is performed by the SparseTensorConversion pass. Differential Revision: https://reviews.llvm.org/D128138
2022-06-24 13:43:30 +02:00
"include/mlir/Dialect/SparseTensor/Transforms/BufferizableOpInterfaceImpl.h",
[mlir][sparse] Factoring magic numbers into a header Addresses https://bugs.llvm.org/show_bug.cgi?id=52303 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D112962
2021-11-05 15:15:39 -07:00
"include/mlir/Dialect/SparseTensor/Transforms/Passes.h",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][sparse] support dynamic sparse tensor slices. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D141532
2023-01-10 22:35:49 +00:00
":ArithUtils",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationDialect",
[mlir][sparse][bufferize] Implement BufferizableOpInterface Only the analysis part of the interface is implemented. The bufferization itself is performed by the SparseTensorConversion pass. Differential Revision: https://reviews.llvm.org/D128138
2022-06-24 13:43:30 +02:00
":BufferizationTransforms",
[mlir][sparse] add support for complex zero/one building Reviewed By: bixia Differential Revision: https://reviews.llvm.org/D126039
2022-05-19 19:33:33 -07:00
":ComplexDialect",
[mlir][Bazel] Port 11069cbcb47845074d526490fff8daff8afda11d
2022-09-30 08:33:10 +02:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
":FuncTransforms",
[mlir][sparse][gpu] a first prototype sparse GPU code generator This implements a proof-of-concept GPU code generator to the sparse compiler pipeline, currently only capable of generating CUDA threads for outermost parallel loops. The objective, obviously, is to grow this concept to a full blown GPU code generator, capable of the right combinaton of code generation as well as exploiting idiomatic kernels or vector specific libraries (think cuSparse). Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D147483
2023-04-03 15:46:23 -07:00
":GPUDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[bazel] fix bazel file. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D140589
2022-12-22 23:03:23 +00:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LinalgTransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":LinalgUtils",
[mlir][sparse] introduce vectorization pass for sparse loops This brings back previous SIMD functionality, but in a separate pass. The idea is to improve this new pass incrementally, going beyond for-loops to while-loops for co-iteration as welll (masking), while introducing new abstractions to make the lowering more progressive. The separation of sparsification and vectorization is a very good first step on this journey. Also brings back ArmSVE support Still to be fine-tuned: + use of "index" in SIMD loop (viz. a[i] = i) + check that all ops really have SIMD support + check all forms of reductions + chain reduction SIMD values Reviewed By: dcaballe Differential Revision: https://reviews.llvm.org/D138236
2022-11-18 12:18:00 -08:00
":MathDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MemRefDialect",
":Pass",
":SCFDialect",
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
":SCFTransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":SparseTensorDialect",
[mlir][sparse] Factoring out SparseTensorEnums library This differential splits the SparseTensorEnums library out from the SparseTensorRuntime library. The actual moving of files will be handled in the next differential. Depends On D135996 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136002
2022-10-17 18:13:05 -07:00
":SparseTensorEnums",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SparseTensorPassIncGen",
Update Bazel build for 744146f60b Updates Bazel build files to match https://github.com/llvm/llvm-project/commit/744146f60b Differential Revision: https://reviews.llvm.org/D104885
2021-06-24 16:33:13 -07:00
":SparseTensorUtils",
[mlir][sparse] migrate sparse rewriting to sparse transformations pass The rules in the linalg file were very specific to sparse tensors so will find a better home under sparse tensor dialect than linalg dialect. Also moved some rewriting from sparsification into this new "pre-rewriting" file. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D129910
2022-07-15 16:41:02 -07:00
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TensorDialect",
":Transforms",
[mlir][sparse] introduce vectorization pass for sparse loops This brings back previous SIMD functionality, but in a separate pass. The idea is to improve this new pass incrementally, going beyond for-loops to while-loops for co-iteration as welll (masking), while introducing new abstractions to make the lowering more progressive. The separation of sparsification and vectorization is a very good first step on this journey. Also brings back ArmSVE support Still to be fine-tuned: + use of "index" in SIMD loop (viz. a[i] = i) + check that all ops really have SIMD support + check all forms of reductions + chain reduction SIMD values Reviewed By: dcaballe Differential Revision: https://reviews.llvm.org/D138236
2022-11-18 12:18:00 -08:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
[mlir][sparse] Adding standard pipeline for tests. Addresses https://bugs.llvm.org/show_bug.cgi?id=52409 aka https://github.com/llvm/llvm-project/issues/51751 Reviewed By: aartbik, mehdi_amini Differential Revision: https://reviews.llvm.org/D117919
2022-01-26 16:44:32 -08:00
cc_library(
name = "SparseTensorPipelines",
srcs = glob(["lib/Dialect/SparseTensor/Pipelines/*.cpp"]),
hdrs = ["include/mlir/Dialect/SparseTensor/Pipelines/Passes.h"],
includes = ["include"],
[mlir][sparse][gpu] sparse GPU code generator pipeline setup Reviewed By: Peiming Differential Revision: https://reviews.llvm.org/D147571
2023-04-05 12:57:23 -07:00
local_defines = if_cuda_available(["MLIR_GPU_TO_CUBIN_PASS_ENABLE"]),
[mlir][sparse] Adding standard pipeline for tests. Addresses https://bugs.llvm.org/show_bug.cgi?id=52409 aka https://github.com/llvm/llvm-project/issues/51751 Reviewed By: aartbik, mehdi_amini Differential Revision: https://reviews.llvm.org/D117919
2022-01-26 16:44:32 -08:00
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithTransforms",
[mlir][sparse] Adding standard pipeline for tests. Addresses https://bugs.llvm.org/show_bug.cgi?id=52409 aka https://github.com/llvm/llvm-project/issues/51751 Reviewed By: aartbik, mehdi_amini Differential Revision: https://reviews.llvm.org/D117919
2022-01-26 16:44:32 -08:00
":BufferizationTransforms",
":ConversionPasses",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
":FuncTransforms",
[mlir][sparse][gpu] sparse GPU code generator pipeline setup Reviewed By: Peiming Differential Revision: https://reviews.llvm.org/D147571
2023-04-05 12:57:23 -07:00
":GPUDialect",
":GPUToNVVMTransforms",
":GPUTransforms",
[mlir][sparse] Adding standard pipeline for tests. Addresses https://bugs.llvm.org/show_bug.cgi?id=52409 aka https://github.com/llvm/llvm-project/issues/51751 Reviewed By: aartbik, mehdi_amini Differential Revision: https://reviews.llvm.org/D117919
2022-01-26 16:44:32 -08:00
":LinalgTransforms",
[bazel] Add a missing dependency after 786cbb09ed
2022-12-02 16:56:05 +01:00
":MemRefTransforms",
[mlir][sparse][gpu] sparse GPU code generator pipeline setup Reviewed By: Peiming Differential Revision: https://reviews.llvm.org/D147571
2023-04-05 12:57:23 -07:00
":NVVMDialect",
[mlir][sparse] Adding standard pipeline for tests. Addresses https://bugs.llvm.org/show_bug.cgi?id=52409 aka https://github.com/llvm/llvm-project/issues/51751 Reviewed By: aartbik, mehdi_amini Differential Revision: https://reviews.llvm.org/D117919
2022-01-26 16:44:32 -08:00
":Pass",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":SparseTensorDialect",
[mlir][sparse] Adding standard pipeline for tests. Addresses https://bugs.llvm.org/show_bug.cgi?id=52409 aka https://github.com/llvm/llvm-project/issues/51751 Reviewed By: aartbik, mehdi_amini Differential Revision: https://reviews.llvm.org/D117919
2022-01-26 16:44:32 -08:00
":SparseTensorTransforms",
":TensorTransforms",
[Bazel] Port eb65327fe9d8b1a6824937e2b76f759101cd6416
2022-09-09 15:35:52 -07:00
":Transforms",
[mlir][sparse] Updating sparse-compiler pipeline for python usage Explicitly nests passes for FuncOp, adds more options to the sparse-compiler pipeline, and updates python integration tests. This should be sufficient to close https://github.com/llvm/llvm-project/issues/51751 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D118658
2022-02-01 14:19:11 -08:00
":VectorToLLVM",
[mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg Bufferize Differential Revision: https://reviews.llvm.org/D119444
2022-02-15 21:16:50 +09:00
":VectorTransforms",
[mlir] fix mismerge in bazel
2023-04-13 14:58:11 +02:00
] + if_cuda_available([
":SerializeToCubin",
]),
[mlir][sparse] Adding standard pipeline for tests. Addresses https://bugs.llvm.org/show_bug.cgi?id=52409 aka https://github.com/llvm/llvm-project/issues/51751 Reviewed By: aartbik, mehdi_amini Differential Revision: https://reviews.llvm.org/D117919
2022-01-26 16:44:32 -08:00
)
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
##---------------------------------------------------------------------------##
# NVGPU dialect.
##---------------------------------------------------------------------------##
td_library(
name = "NVGPUTdFiles",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
srcs = ["include/mlir/Dialect/NVGPU/IR/NVGPU.td"],
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
includes = ["include"],
deps = [
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "NVGPUIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=nvgpu",
],
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
"include/mlir/Dialect/NVGPU/IR/NVGPUDialect.h.inc",
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
),
(
[
"-gen-dialect-defs",
"-dialect=nvgpu",
],
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
"include/mlir/Dialect/NVGPU/IR/NVGPUDialect.cpp.inc",
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
),
(
["-gen-op-decls"],
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
"include/mlir/Dialect/NVGPU/IR/NVGPU.h.inc",
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
),
(
["-gen-op-defs"],
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
"include/mlir/Dialect/NVGPU/IR/NVGPU.cpp.inc",
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
),
(
["-gen-op-doc"],
"g3doc/Dialects/NVGPU/NVGPU.md",
),
[mlir][Bazel] Port afb0b21f24adecfa0b302c79366980453b15ba7b
2022-09-26 09:37:32 +02:00
(
["-gen-typedef-decls"],
"include/mlir/Dialect/NVGPU/IR/NVGPUTypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/NVGPU/IR/NVGPUTypes.cpp.inc",
),
[mlir] Update bazel build after rG70c2e0618a0f3c09ed7149d88b4987b932eb6705
2023-07-21 13:16:31 +02:00
(
["-gen-enum-decls"],
"include/mlir/Dialect/NVGPU/IR/NVGPUEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/NVGPU/IR/NVGPUEnums.cpp.inc",
),
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/NVGPU/IR/NVGPUAttrDefs.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/NVGPU/IR/NVGPUAttrDefs.cpp.inc",
),
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
],
tblgen = ":mlir-tblgen",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
td_file = "include/mlir/Dialect/NVGPU/IR/NVGPU.td",
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
deps = [":NVGPUTdFiles"],
)
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
gentbl_cc_library(
name = "NVGPUPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=NVGPU",
],
[mlir][NVGPU][NFC] Clean up code structure * Move passes to `Transforms` directory. * Add `Utils.h` (will be utilized in a subsequent change). Differential Revision: https://reviews.llvm.org/D155427
2023-07-17 10:30:06 +02:00
"include/mlir/Dialect/NVGPU/Transforms/Passes.h.inc",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
),
],
tblgen = ":mlir-tblgen",
[mlir][NVGPU][NFC] Clean up code structure * Move passes to `Transforms` directory. * Add `Utils.h` (will be utilized in a subsequent change). Differential Revision: https://reviews.llvm.org/D155427
2023-07-17 10:30:06 +02:00
td_file = "include/mlir/Dialect/NVGPU/Transforms/Passes.td",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
deps = [":PassBaseTdFiles"],
)
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "NVGPUDialect",
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
srcs = ["lib/Dialect/NVGPU/IR/NVGPUDialect.cpp"],
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
hdrs = ["include/mlir/Dialect/NVGPU/IR/NVGPUDialect.h"],
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
includes = ["include"],
deps = [
[mlir][gpu] Move async copy ops to NVGPU and add caching hints Move async copy operations to NVGPU as they only exist on NV target and are designed to match ptx semantic. This allows us to also add more fine grain caching hint attribute to the op. Add hint to bypass L1 and hook it up to NVVM op. Differential Revision: https://reviews.llvm.org/D125244
2022-05-09 17:18:21 +00:00
":GPUDialect",
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
":IR",
":NVGPUIncGen",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
":SideEffectInterfaces",
"//llvm:Core",
"//llvm:Support",
],
)
Revert "Revert "[mlir][Transform] Add support for mma.sync m16n8k16 f16 rewrite." and "[mlir][Transform] Introduce nvgpu transform extensions"" This reverts commit 6506692fe619ef8a1f7c6ea829d9a9eceb31622d. Differential Revision: https://reviews.llvm.org/D153845
2023-06-27 09:02:31 +00:00
cc_library(
name = "NVGPUTransformOps",
srcs = glob([
"lib/Dialect/NVGPU/TransformOps/*.cpp",
]),
hdrs = glob([
"include/mlir/Dialect/NVGPU/TransformOps/*.h",
]),
includes = ["include"],
deps = [
[mlir][Bazel] Add missing dependencies after 564713c47175d9f61fe8e18e750fb7f7b486d533
2023-07-03 13:16:28 +02:00
":AffineDialect",
[mlir][nvgpu] add simple pipelining for shared memory copies Add a simple transform operation to the NVGPU extension that performs software pipelining of copies to shared memory. The functionality is extremely minimalistic in this version and only supports copies from global to shared memory inside an `scf.for` loop with either `vector.transfer` or `nvgpu.device_async_copy` operations when pipelining preconditions are already satisfied in the IR. This is the minimally useful version that uses the more general loop pipeliner in an NVGPU-specific way. Further extensions and orthogonalizations will be necessary. This required a change to the loop pipeliner itself to properly propagate errors should the predicate generator fail. This is loosely inspired from the vesion in IREE, but has less unsafe assumptions and more principled way of communicating decisions. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D155223
2023-07-13 17:55:36 +00:00
":Analysis",
Revert "Revert "[mlir][Transform] Add support for mma.sync m16n8k16 f16 rewrite." and "[mlir][Transform] Introduce nvgpu transform extensions"" This reverts commit 6506692fe619ef8a1f7c6ea829d9a9eceb31622d. Differential Revision: https://reviews.llvm.org/D153845
2023-06-27 09:02:31 +00:00
":ArithDialect",
":ArithUtils",
":DialectUtils",
":GPUDialect",
":IR",
":LinalgDialect",
":MemRefDialect",
":NVGPUDialect",
":NVGPUTransformOpsIncGen",
[mlir][NVGPU][transform] Add `create_async_groups` transform op This transform looks for suitable vector transfers from global memory to shared memory and converts them to async device copies. Differential Revision: https://reviews.llvm.org/D155569
2023-07-18 14:13:31 +02:00
":NVGPUTransforms",
[mlir][nvgpu] add simple pipelining for shared memory copies Add a simple transform operation to the NVGPU extension that performs software pipelining of copies to shared memory. The functionality is extremely minimalistic in this version and only supports copies from global to shared memory inside an `scf.for` loop with either `vector.transfer` or `nvgpu.device_async_copy` operations when pipelining preconditions are already satisfied in the IR. This is the minimally useful version that uses the more general loop pipeliner in an NVGPU-specific way. Further extensions and orthogonalizations will be necessary. This required a change to the loop pipeliner itself to properly propagate errors should the predicate generator fail. This is loosely inspired from the vesion in IREE, but has less unsafe assumptions and more principled way of communicating decisions. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D155223
2023-07-13 17:55:36 +00:00
":SCFDialect",
":SCFTransforms",
Revert "Revert "[mlir][Transform] Add support for mma.sync m16n8k16 f16 rewrite." and "[mlir][Transform] Introduce nvgpu transform extensions"" This reverts commit 6506692fe619ef8a1f7c6ea829d9a9eceb31622d. Differential Revision: https://reviews.llvm.org/D153845
2023-06-27 09:02:31 +00:00
":Support",
":TransformDialect",
":VectorDialect",
"//llvm:Support",
],
)
td_library(
name = "NVGPUTransformOpsTdFiles",
srcs = glob([
"include/mlir/Dialect/NVGPU/TransformOps/*.td",
]),
includes = ["include"],
deps = [
":TransformDialectTdFiles",
],
)
gentbl_cc_library(
name = "NVGPUTransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/NVGPU/TransformOps/NVGPUTransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/NVGPU/TransformOps/NVGPUTransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/NVGPU/TransformOps/NVGPUTransformOps.td",
deps = [
":NVGPUTransformOpsTdFiles",
],
)
Add NVGPUUtil bazel lib
2023-01-17 15:22:30 -08:00
cc_library(
name = "NVGPUUtils",
srcs = ["lib/Dialect/NVGPU/Utils/MMAUtils.cpp"],
hdrs = ["include/mlir/Dialect/NVGPU/Utils/MMAUtils.h"],
includes = ["include"],
deps = [
":AffineDialect",
":ArithDialect",
":IR",
":NVGPUDialect",
":NVVMDialect",
":VectorDialect",
],
)
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
cc_library(
name = "NVGPUTransforms",
[mlir][NVGPU][NFC] Clean up code structure * Move passes to `Transforms` directory. * Add `Utils.h` (will be utilized in a subsequent change). Differential Revision: https://reviews.llvm.org/D155427
2023-07-17 10:30:06 +02:00
srcs = glob([
"lib/Dialect/NVGPU/Transforms/*.cpp",
]),
hdrs = glob([
"include/mlir/Dialect/NVGPU/Transforms/*.h",
]),
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
includes = ["include"],
deps = [
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][sparse] Inline the definition of LLVM_ATTRIBUTE_WEAK This library is supposed not to have a dependency on LLVM, and linking LLVMSupport into it breaks its shared library setup.
2022-06-17 22:40:00 +02:00
":FuncDialect",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
":GPUDialect",
[mlir][nvgpu] Fix Bazel BUILD file Differential Revision: https://reviews.llvm.org/D128078
2022-06-17 14:31:44 -04:00
":IR",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
":MemRefDialect",
":NVGPUDialect",
[mlir][nvgpu] Fix Bazel BUILD file Differential Revision: https://reviews.llvm.org/D128078
2022-06-17 14:31:44 -04:00
":NVGPUPassIncGen",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
":Pass",
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
":SideEffectInterfaces",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
":Support",
":Transforms",
":VectorDialect",
[mlir][nvgpu] Add NVGPU dialect (architectural specific gpu dialect) This introduce a new dialect for vendro specific ptx operations. This also adds the first operation ldmatrix as an example. More operations will be added in follow up patches. This new dialect is meant to be a bridge between GPU and Vector dialectis and NVVM dialect. This is based on the RFC proposed here: https://discourse.llvm.org/t/rfc-add-nv-gpu-dialect-hw-specific-extension-of-gpu-dialect-for-nvidia-gpus/61466/8 Differential Revision: https://reviews.llvm.org/D123266
2022-04-07 07:14:39 +00:00
"//llvm:Core",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
name = "FuncTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = [
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir/Dialect/Func/IR/FuncOps.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
[mlir] Fix BUILD issues and dependencies. Differential Revision: https://reviews.llvm.org/D123868
2022-04-15 18:59:03 +00:00
":AttrTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":CallInterfacesTdFiles",
":CastInterfacesTdFiles",
":ControlFlowInterfacesTdFiles",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FunctionInterfacesTdFiles",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
":VectorInterfacesTdFiles",
],
)
gentbl_cc_library(
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
name = "FuncIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir/Dialect/Func/IR/FuncOps.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-op-defs"],
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir/Dialect/Func/IR/FuncOps.cpp.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-dialect-decls"],
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir/Dialect/Func/IR/FuncOpsDialect.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir/Dialect/Func/IR/FuncOpsDialect.cpp.inc",
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-enum-decls"],
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir/Dialect/Func/IR/FuncOpsEnums.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-enum-defs"],
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir/Dialect/Func/IR/FuncOpsEnums.cpp.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
],
tblgen = ":mlir-tblgen",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
td_file = "include/mlir/Dialect/Func/IR/FuncOps.td",
deps = [":FuncTdFiles"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
cc_library(
name = "Dialect",
srcs = glob([
"lib/Dialect/*.cpp",
"lib/Dialect/*.h",
]),
hdrs = glob([
"include/mlir/Dialect/*.h",
]),
includes = ["include"],
deps = [
":IR",
"//llvm:Support",
],
)
[mlir] Add IteratorType enum to StructuredOpsUtils. Summary: Use the new enum in TilingIterface and verify that `iterator_type` attribute in LinalgOp interface is compatible with the enum values. Later IteratorType enum will be used in LinalgInterface to replace the current `iterator_type` attribute array of string. Existing enums in Linalg are moved into a separate td file and tablegen build target. This is necessary, have one I32EnumAttr in a shared space that generated enum class definition and EnumAttrs is dialect-specific location. Otherwise there might be a conflict that I32EnumAttr generates enum definitions in multiple places. Differential Revision: https://reviews.llvm.org/D134634
2022-09-26 10:33:08 +00:00
td_library(
name = "DialectUtilsTdFiles",
srcs = [
"include/mlir/Dialect/Utils/StructuredOpsUtils.td",
],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
gentbl_cc_library(
name = "DialectUtilsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-enum-decls"],
"include/mlir/Dialect/Utils/DialectUtilsEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/Utils/DialectUtilsEnums.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Utils/StructuredOpsUtils.td",
deps = [":DialectUtilsTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "DialectUtils",
srcs = glob([
"lib/Dialect/Utils/*.cpp",
"lib/Dialect/Utils/*.h",
]),
hdrs = glob([
"include/mlir/Dialect/Utils/*.h",
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithUtils",
[mlir] Add IteratorType enum to StructuredOpsUtils. Summary: Use the new enum in TilingIterface and verify that `iterator_type` attribute in LinalgOp interface is compatible with the enum values. Later IteratorType enum will be used in LinalgInterface to replace the current `iterator_type` attribute array of string. Existing enums in Linalg are moved into a separate td file and tablegen build target. This is necessary, have one I32EnumAttr in a shared space that generated enum class definition and EnumAttrs is dialect-specific location. Otherwise there might be a conflict that I32EnumAttr generates enum definitions in multiple places. Differential Revision: https://reviews.llvm.org/D134634
2022-09-26 10:33:08 +00:00
":DialectUtilsIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":Support",
"//llvm:Support",
],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "AffineDialect",
srcs = glob([
"lib/Dialect/Affine/IR/*.cpp",
"lib/Dialect/Affine/IR/*.h",
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Dialect/Affine/IR/*.h",
]),
includes = ["include"],
deps = [
":AffineMemoryOpInterfacesIncGen",
":AffineOpsIncGen",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[bazel] Port ead1107257e6
2022-02-03 01:41:03 +01:00
":ControlFlowInterfaces",
[mlir][Bazel] Fix Bazel build Differential Revision: https://reviews.llvm.org/D132264
2022-08-21 01:01:04 -07:00
":DialectUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":LoopLikeInterface",
":MemRefDialect",
[mlir][interfaces] Add ShapedDimOpInterface This interface is implemented by memref.dim and tensor.dim. This change makes it possible to remove a build dependency of the Affine dialect on the Tensor dialect (and maybe also the MemRef dialect in the future). Differential Revision: https://reviews.llvm.org/D133595
2022-10-03 13:53:16 +09:00
":ShapedOpInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfaces",
[bazel] Add missing dependency after 687f78210d0
2023-02-10 10:22:18 +01:00
":Support",
[mlir][linalg] ValueBoundsOpInterface: Add affine.apply op Differential Revision: https://reviews.llvm.org/D145694
2023-04-06 10:15:11 +09:00
":ValueBoundsOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "EmitCDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = glob([
"lib/Dialect/EmitC/IR/*.cpp",
]),
hdrs = glob([
"include/mlir/Dialect/EmitC/IR/*.h",
]),
includes = ["include"],
deps = [
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":EmitCAttributesIncGen",
":EmitCOpsIncGen",
":IR",
":SideEffectInterfaces",
"//llvm:Support",
],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "AsyncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = glob([
"lib/Dialect/Async/IR/*.cpp",
]),
hdrs = glob([
"include/mlir/Dialect/Async/IR/*.h",
]),
includes = ["include"],
deps = [
":AsyncOpsIncGen",
":ControlFlowInterfaces",
":IR",
[async] Get the number of worker threads from the runtime. Reviewed By: ezhulenev Differential Revision: https://reviews.llvm.org/D117751
2022-01-31 12:00:03 -08:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfaces",
"//llvm:Support",
],
)
cc_library(
name = "AsyncTransforms",
srcs = glob([
"lib/Dialect/Async/Transforms/*.cpp",
"lib/Dialect/Async/Transforms/*.h",
]),
Make AsyncParallelForRewrite parameterizable with a cost model which drives deciding the parallelization granularity. Reviewed By: ezhulenev, mehdi_amini Differential Revision: https://reviews.llvm.org/D115423
2021-12-19 08:35:37 -08:00
hdrs = [
"include/mlir/Dialect/Async/Passes.h",
"include/mlir/Dialect/Async/Transforms.h",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AsyncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AsyncPassIncGen",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":Pass",
":SCFDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":SCFToControlFlow",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":TransformUtils",
[mlir][Vector] Add InferTypeOpInterface to vector dialect and remove the now redundant builders
2022-02-18 01:15:14 +01:00
":Transforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TransformsPassIncGen",
"//llvm:Core",
"//llvm:Support",
],
)
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
cc_library(
name = "AffineAnalysis",
srcs = glob([
"lib/Dialect/Affine/Analysis/*.cpp",
"lib/Dialect/Affine/Analysis/*.h",
]),
hdrs = glob(["include/mlir/Dialect/Affine/Analysis/*.h"]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][affine][NFC] Extract core functionality of `canonicalizeMinMaxOp` Move code from SCF to Affine: Add a new helper function `simplifyConstrainedMinMaxOp` to Affine/Analysis/Utils.h. `canonicalizeMinMaxOp` was originally designed for loop peeling, but it is not SCF-specific and can be used to simplify any affine.min/max ops. Various functions in SCF/Transforms are simplified by dropping unnecessary parameters. Differential Revision: https://reviews.llvm.org/D140962
2023-01-04 10:56:43 +01:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
":IR",
[mlir] Remove `Transforms/SideEffectUtils.h` and move the methods into `Interface/SideEffectInterfaces.h`. The methods in `SideEffectUtils.h` (and their implementations in `SideEffectUtils.cpp`) seem to have similar intent to methods already existing in `SideEffectInterfaces.h`. Move the decleration (and implementation) from `SideEffectUtils.h` (and `SideEffectUtils.cpp`) into `SideEffectInterfaces.h` (and `SideEffectInterface.cpp`). Also drop the `SideEffectInterface::hasNoEffect` method in favor of `mlir::isMemoryEffectFree` which actually recurses into the operation instead of just relying on the `hasRecursiveMemoryEffectTrait` exclusively. Differential Revision: https://reviews.llvm.org/D137857
2022-11-11 17:10:38 +00:00
":SideEffectInterfaces",
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
":Support",
More bazel fixes
2022-03-04 04:20:08 +00:00
":ViewLikeInterface",
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "AffineUtils",
srcs = glob(
[
"lib/Dialect/Affine/Utils/*.cpp",
"lib/Dialect/Affine/Utils/*.h",
],
),
[bazel] Update MLIR deps I believe this is due to D117839, D117848, maybe others. Latest build failure: https://buildkite.com/llvm-project/upstream-bazel-rbe/builds/18053
2022-01-24 21:21:37 -08:00
hdrs = [
"include/mlir/Dialect/Affine/LoopFusionUtils.h",
"include/mlir/Dialect/Affine/LoopUtils.h",
"include/mlir/Dialect/Affine/Utils.h",
[mlir] NFC: move mergeOffsetsSizesAndStrides into Affine/Utils So that these utility functions can also be used ViewLikeInterface ops not in the memref dialect. Reviewed By: mravishankar, christopherbate Differential Revision: https://reviews.llvm.org/D134487
2022-09-23 13:20:10 -04:00
"include/mlir/Dialect/Affine/ViewLikeInterfaceUtils.h",
[bazel] Update MLIR deps I believe this is due to D117839, D117848, maybe others. Latest build failure: https://buildkite.com/llvm-project/upstream-bazel-rbe/builds/18053
2022-01-24 21:21:37 -08:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
":AffineAnalysis",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithUtils",
[bazel] Fix build after 33468a5 Add missing dependency.
2023-04-14 20:44:19 +02:00
":DialectUtils",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir] Update BUILD rule for AffineUtils bc657b2eef82f604e5bfb8da421cbdfc80156739 adds the dependency to MemRefDialect. Reviewed By: GMNGeoffrey, mehdi_amini Differential Revision: https://reviews.llvm.org/D115558
2021-12-10 23:50:27 +00:00
":MemRefDialect",
[bazel] Update MLIR deps I believe this is due to D117839, D117848, maybe others. Latest build failure: https://buildkite.com/llvm-project/upstream-bazel-rbe/builds/18053
2022-01-24 21:21:37 -08:00
":SCFDialect",
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TransformUtils",
[mlir] NFC: move mergeOffsetsSizesAndStrides into Affine/Utils So that these utility functions can also be used ViewLikeInterface ops not in the memref dialect. Reviewed By: mravishankar, christopherbate Differential Revision: https://reviews.llvm.org/D134487
2022-09-23 13:20:10 -04:00
":ViewLikeInterface",
[bazel] Update MLIR deps I believe this is due to D117839, D117848, maybe others. Latest build failure: https://buildkite.com/llvm-project/upstream-bazel-rbe/builds/18053
2022-01-24 21:21:37 -08:00
"//llvm:Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
gentbl_cc_library(
name = "AffinePassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Affine",
],
"include/mlir/Dialect/Affine/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Affine/Passes.td",
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "AffineTransforms",
srcs = glob([
"lib/Dialect/Affine/Transforms/*.cpp",
"lib/Dialect/Affine/Transforms/*.h",
]),
[mlir][Bazel] Adjust build after 0fa20ecafe0c3c7ffde413800eb4b1551b432273
2023-03-14 13:45:53 +01:00
hdrs = [
"include/mlir/Dialect/Affine/Passes.h",
"include/mlir/Dialect/Affine/Transforms/Transforms.h",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
":AffineAnalysis",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AffinePassIncGen",
":AffineUtils",
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":MemRefDialect",
":Pass",
":SCFDialect",
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
":SCFUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
[mlir][Interfaces] Add ValueBoundsOpInterface Ops can implement this interface to specify lower/upper bounds for their result values and block arguments. Bounds can be specified for: * Index-type values * Dimension sizes of shapes values The bounds are added to a constraint set. Users can query this constraint set to compute bounds wrt. to a user-specified set of values. Only EQ bounds are supported at the moment. This revision also contains interface implementations for various tensor dialect ops, which illustrates how to implement this interface. Differential Revision: https://reviews.llvm.org/D145681
2023-04-06 02:41:15 +02:00
":TensorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
[mlir][Interfaces] Add ValueBoundsOpInterface Ops can implement this interface to specify lower/upper bounds for their result values and block arguments. Bounds can be specified for: * Index-type values * Dimension sizes of shapes values The bounds are added to a constraint set. Users can query this constraint set to compute bounds wrt. to a user-specified set of values. Only EQ bounds are supported at the moment. This revision also contains interface implementations for various tensor dialect ops, which illustrates how to implement this interface. Differential Revision: https://reviews.llvm.org/D145681
2023-04-06 02:41:15 +02:00
":ValueBoundsOpInterface",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":VectorUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
gentbl_cc_library(
name = "ConversionPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Conversion",
],
"include/mlir/Conversion/Passes.h.inc",
),
(
[
"-gen-pass-capi-header",
"--prefix=Conversion",
],
"include/mlir/Conversion/Passes.capi.h.inc",
),
(
[
"-gen-pass-capi-impl",
"--prefix=Conversion",
],
"include/mlir/Conversion/Passes.capi.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Conversion/Passes.td",
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "ConversionPasses",
hdrs = ["include/mlir/Conversion/Passes.h"],
includes = ["include"],
deps = [
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
":AMDGPUToROCDL",
[mlir][linalg] Add lowering of named ops on complex numbers This lets linalg.dot and friends lower to a complex muladd using ops from the complex dialect. Differential Revision: https://reviews.llvm.org/D125461
2022-05-12 13:35:27 +02:00
":AffineToStandard",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToLLVM",
":ArithToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ArmNeon2dToIntr",
":AsyncToLLVM",
[mlir] Add bazel build for BufferizationToMemRef.
2021-11-30 13:27:24 +01:00
":BufferizationToMemRef",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ComplexToLLVM",
[mlir][complex] Add pow/sqrt/tanh ops and lowering to libm Lowering through libm gives us a baseline version, even though it's not going to be particularly fast. This is similar to what we do for some math dialect ops. Differential Revision: https://reviews.llvm.org/D125550
2022-05-13 17:12:11 +02:00
":ComplexToLibm",
[mlir][spirv][complex] Convert complex ops to SPIR-V ops This commit adds conversion from complex construction and extraction ops to SPIR-V. Other arithemtic ops can be done via ComplexToStandard patterns. Reviewed By: kuhar Differential Revision: https://reviews.llvm.org/D147193
2023-03-30 08:49:57 -07:00
":ComplexToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ComplexToStandard",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":ControlFlowToLLVM",
":ControlFlowToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
More bazel fixes.
2022-03-03 02:05:11 +00:00
":FuncToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUToGPURuntimeTransforms",
":GPUToNVVMTransforms",
":GPUToROCDLTransforms",
":GPUToSPIRV",
":GPUToVulkanTransforms",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
":IndexToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LinalgToStandard",
[mlir][Bazel] Fix bazel build.
2022-08-26 09:04:50 +02:00
":MathToFuncs",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MathToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MathToLibm",
[mlir] NFC: split Math to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107093
2021-07-29 16:21:43 -04:00
":MathToSPIRV",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MemRefToLLVM",
[mlir] NFC: split MemRef to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107094
2021-07-29 16:22:46 -04:00
":MemRefToSPIRV",
[mlir] Update bazel file after adding nvgpu to nvvm conversion
2022-04-15 04:07:46 +00:00
":NVGPUToNVVM",
Fix bazel build file for D154060. Differential Revision: https://reviews.llvm.org/D154976
2023-07-11 17:03:48 +02:00
":NVVMToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpenACCToLLVM",
":OpenACCToSCF",
":OpenMPToLLVM",
":PDLToPDLInterp",
[mlir] Factor type reconciliation out of Standard-to-LLVM conversion Conversion to the LLVM dialect is being refactored to be more progressive and is now performed as a series of independent passes converting different dialects. These passes may produce `unrealized_conversion_cast` operations that represent pending conversions between built-in and LLVM dialect types. Historically, a more monolithic Standard-to-LLVM conversion pass did not need these casts as all operations were converted in one shot. Previous refactorings have led to the requirement of running the Standard-to-LLVM conversion pass to clean up `unrealized_conversion_cast`s even though the IR had no standard operations in it. The pass must have been also run the last among all to-LLVM passes, in contradiction with the partial conversion logic. Additionally, the way it was set up could produce invalid operations by removing casts between LLVM and built-in types even when the consumer did not accept the uncasted type, or could lead to cryptic conversion errors (recursive application of the rewrite pattern on `unrealized_conversion_cast` as a means to indicate failure to eliminate casts). In fact, the need to eliminate A->B->A `unrealized_conversion_cast`s is not specific to to-LLVM conversions and can be factored out into a separate type reconciliation pass, which is achieved in this commit. While the cast operation itself has a folder pattern, it is insufficient in most conversion passes as the folder only applies to the second cast. Without complex legality setup in the conversion target, the conversion infra will either consider the cast operations valid and not fold them (a separate canonicalization would be necessary to trigger the folding), or consider the first cast invalid upon generation and stop with error. The pattern provided by the reconciliation pass applies to the first cast operation instead. Furthermore, having a separate pass makes it clear when `unrealized_conversion_cast`s could not have been eliminated since it is the only reason why this pass can fail. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D109507
2021-09-09 16:06:10 +02:00
":ReconcileUnrealizedCasts",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":SCFToControlFlow",
":SCFToGPU",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SCFToOpenMP",
":SCFToSPIRV",
":SPIRVToLLVM",
":ShapeToStandard",
[mlir] Add TensorToLinalgPass This pass is to handle computationally complex operations like tensor.pad which are not simply lowered to the exact same operation in the memref dialect. Differential Revision: https://reviews.llvm.org/D125384
2022-05-11 14:10:12 +02:00
":TensorToLinalg",
More bazel fixes.
2022-03-03 02:05:11 +00:00
":TensorToSPIRV",
[tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard Apply scale should be optionally disabled when lowering via TosaToStandard. In most cases it should persist until the lowering to specific backend. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D122948
2022-04-04 12:22:09 -07:00
":TosaToArith",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TosaToLinalg",
":TosaToSCF",
[tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard Apply scale should be optionally disabled when lowering via TosaToStandard. In most cases it should persist until the lowering to specific backend. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D122948
2022-04-04 12:22:09 -07:00
":TosaToTensor",
[bazel][mlir] Port 0f446adf678212b6daedebc5eb1bd2849b1efad1
2023-07-24 19:21:18 +02:00
":UBToLLVM",
[bazel] Add UPToSPIRV library/dep Added in D156163
2023-07-24 19:33:33 -07:00
":UBToSPIRV",
[mlir][Bazel] Add VectorToArmSME target.
2023-07-18 16:02:15 +02:00
":VectorToArmSME",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":VectorToGPU",
":VectorToLLVM",
":VectorToSCF",
":VectorToSPIRV",
],
)
cc_library(
name = "AsyncToLLVM",
srcs = glob([
"lib/Conversion/AsyncToLLVM/*.cpp",
"lib/Conversion/AsyncToLLVM/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob(["include/mlir/Conversion/AsyncToLLVM/*.h"]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AsyncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
":FuncTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
":Pass",
":Support",
":Transforms",
"//llvm:Support",
],
)
cc_library(
name = "AffineToStandard",
srcs = glob([
"lib/Conversion/AffineToStandard/*.cpp",
"lib/Conversion/AffineToStandard/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob(["include/mlir/Conversion/AffineToStandard/*.h"]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Update bazel after 8d12bf4ac102ea5887705ca467163aecea13b6f1
2022-02-10 10:12:04 +01:00
":AffineUtils",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":MemRefDialect",
":Pass",
":SCFDialect",
":Support",
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
# SDBM dialect only contains attribute components that can be constructed given
# a dialect object, so whenever it is used it must also be registered. Therefore
# we don't split out the registration library for it.
cc_library(
name = "SDBM",
srcs = glob([
"lib/Dialect/SDBM/*.cpp",
"lib/Dialect/SDBM/*.h",
]),
hdrs = glob([
"include/mlir/Dialect/SDBM/*.h",
]),
includes = ["include"],
deps = [
":IR",
":Support",
"//llvm:Support",
],
)
cc_library(
name = "SCFDialect",
srcs = glob(
[
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
"lib/Dialect/SCF/IR/*.cpp",
"lib/Dialect/SCF/IR/*.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
),
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
hdrs = glob(
[
[mlir] move SCF headers to SCF/{IR,Transforms} respectively This aligns the SCF dialect file layout with the majority of the dialects. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D128049
2022-06-17 15:47:15 +02:00
"include/mlir/Dialect/SCF/IR/*.h",
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
],
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ControlFlowInterfaces",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[MLIR] Add InferTypeOpInterface to scf.if op Differential Revision: https://reviews.llvm.org/D142049
2023-01-19 13:18:22 -05:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LoopLikeInterface",
[Bazel] Add more mlir dependencies after D124298 The Bazel layering_check feature compiles libraries with `-fmodule-name=X -fmodules-strict-decluse` which require #include to be in deps.
2022-04-23 13:06:54 -07:00
":MemRefDialect",
[fix/build] bazel rule for ParallelCombiningOpInterface
2022-07-01 10:33:48 +02:00
":ParallelCombiningOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
[mlir] Introduce device mapper attribute for `thread_dim_map` and `mapped to dims` `scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [0, 1]} } { thread_dim_mapping = [0, 1]} ``` It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation. The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]} } { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]} ``` Reviewed By: ftynse, nicolasvasilache Differential Revision: https://reviews.llvm.org/D137413
2022-11-10 17:55:49 +01:00
":SCFDeviceMappingInterfacesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SCFIncGen",
":SCFPassIncGen",
":Support",
[Bazel] Add more mlir dependencies after D124298 The Bazel layering_check feature compiles libraries with `-fmodule-name=X -fmodules-strict-decluse` which require #include to be in deps.
2022-04-23 13:06:54 -07:00
":TensorDialect",
[mlir] fix mismerge in bazel
2023-04-13 14:58:11 +02:00
":ValueBoundsOpInterface",
Fix bazel build after 59b273a166b9d1165015b5cf10c21641cc01d682. Reviewed By: tpopp Differential Revision: https://reviews.llvm.org/D126765
2022-06-01 12:06:28 +02:00
":ViewLikeInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
cc_library(
name = "SCFUtils",
srcs = glob(
[
"lib/Dialect/SCF/Utils/*.cpp",
],
),
hdrs = glob(
[
"include/mlir/Dialect/SCF/Utils/*.h",
],
),
includes = ["include"],
deps = [
":AffineAnalysis",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
":IR",
":SCFDialect",
[mlir] Remove `Transforms/SideEffectUtils.h` and move the methods into `Interface/SideEffectInterfaces.h`. The methods in `SideEffectUtils.h` (and their implementations in `SideEffectUtils.cpp`) seem to have similar intent to methods already existing in `SideEffectInterfaces.h`. Move the decleration (and implementation) from `SideEffectUtils.h` (and `SideEffectUtils.cpp`) into `SideEffectInterfaces.h` (and `SideEffectInterface.cpp`). Also drop the `SideEffectInterface::hasNoEffect` method in favor of `mlir::isMemoryEffectFree` which actually recurses into the operation instead of just relying on the `hasRecursiveMemoryEffectTrait` exclusively. Differential Revision: https://reviews.llvm.org/D137857
2022-11-11 17:10:38 +00:00
":SideEffectInterfaces",
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
":Support",
":Transforms",
"//llvm:Support",
],
)
Revert "Revert "[MLIR] Update bazel build file after 455305624884cf9237143e2ba0635fcc5ba5206"" This reverts commit 9ef7ae507387f133cf34ca0bc0edd3f009a3ff71.
2023-01-20 21:56:25 +01:00
cc_library(
name = "InferIntRangeCommon",
srcs = [
"lib/Interfaces/Utils/InferIntRangeCommon.cpp",
],
hdrs = ["include/mlir/Interfaces/Utils/InferIntRangeCommon.h"],
includes = ["include"],
deps = [
":IR",
":InferIntRangeInterface",
[bazel] Fix up dependency
2023-01-20 21:58:56 +01:00
"//llvm:Support",
Revert "Revert "[MLIR] Update bazel build file after 455305624884cf9237143e2ba0635fcc5ba5206"" This reverts commit 9ef7ae507387f133cf34ca0bc0edd3f009a3ff71.
2023-01-20 21:56:25 +01:00
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "DataLayoutInterfaces",
srcs = ["lib/Interfaces/DataLayoutInterfaces.cpp"],
hdrs = ["include/mlir/Interfaces/DataLayoutInterfaces.h"],
includes = ["include"],
deps = [
":DataLayoutInterfacesIncGen",
":IR",
"//llvm:Support",
],
)
cc_library(
name = "LoopLikeInterface",
srcs = ["lib/Interfaces/LoopLikeInterface.cpp"],
hdrs = ["include/mlir/Interfaces/LoopLikeInterface.h"],
includes = ["include"],
deps = [
":IR",
":LoopLikeInterfaceIncGen",
[bazel] Update MLIR deps I believe this is due to D117839, D117848, maybe others. Latest build failure: https://buildkite.com/llvm-project/upstream-bazel-rbe/builds/18053
2022-01-24 21:21:37 -08:00
"//llvm:Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
cc_library(
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
name = "MemorySlotInterfaces",
srcs = ["lib/Interfaces/MemorySlotInterfaces.cpp"],
hdrs = ["include/mlir/Interfaces/MemorySlotInterfaces.h"],
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
includes = ["include"],
deps = [
":IR",
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
":MemorySlotInterfacesIncGen",
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
"//llvm:Support",
],
)
[mlir][interfaces] Add ShapedDimOpInterface This interface is implemented by memref.dim and tensor.dim. This change makes it possible to remove a build dependency of the Affine dialect on the Tensor dialect (and maybe also the MemRef dialect in the future). Differential Revision: https://reviews.llvm.org/D133595
2022-10-03 13:53:16 +09:00
cc_library(
name = "ShapedOpInterfaces",
srcs = ["lib/Interfaces/ShapedOpInterfaces.cpp"],
hdrs = ["include/mlir/Interfaces/ShapedOpInterfaces.h"],
includes = ["include"],
deps = [
":IR",
":ShapedOpInterfacesIncGen",
"//llvm:Support",
],
)
[fix/build] bazel rule for ParallelCombiningOpInterface
2022-07-01 10:33:48 +02:00
cc_library(
name = "ParallelCombiningOpInterface",
srcs = ["lib/Interfaces/ParallelCombiningOpInterface.cpp"],
hdrs = ["include/mlir/Interfaces/ParallelCombiningOpInterface.h"],
includes = ["include"],
deps = [
":IR",
":ParallelCombiningOpInterfaceIncGen",
"//llvm:Support",
],
)
[mlir] Add RuntimeVerifiableOpInterface and transform Static op verification cannot detect cases where an op is valid at compile time but may be invalid at runtime. An example of such an op is `memref::ExpandShapeOp`. Invalid at compile time: `memref.expand_shape %m [[0, 1]] : memref<11xf32> into memref<2x5xf32>` Valid at compile time (because we do not know any better): `memref.expand_shape %m [[0, 1]] : memref<?xf32> into memref<?x5xf32>`. This op may or may not be valid at runtime depending on the runtime shape of `%m`. Invalid runtime ops such as the one above are hard to debug because they can crash the program execution at a seemingly unrelated position or (even worse) compute an invalid result without crashing. This revision adds a new op interface `RuntimeVerifiableOpInterface` that can be implemented by ops that provide additional runtime verification. Such runtime verification can be computationally expensive, so it is only generated on an opt-in basis by running `-generate-runtime-verification`. A simple runtime verifier for `memref::ExpandShapeOp` is provided as an example. Differential Revision: https://reviews.llvm.org/D138576
2022-12-21 10:51:10 +01:00
cc_library(
name = "RuntimeVerifiableOpInterface",
srcs = ["lib/Interfaces/RuntimeVerifiableOpInterface.cpp"],
hdrs = ["include/mlir/Interfaces/RuntimeVerifiableOpInterface.h"],
includes = ["include"],
deps = [
":IR",
":RuntimeVerifiableOpInterfaceIncGen",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "VectorInterfaces",
srcs = ["lib/Interfaces/VectorInterfaces.cpp"],
hdrs = ["include/mlir/Interfaces/VectorInterfaces.h"],
includes = ["include"],
deps = [
":IR",
":VectorInterfacesIncGen",
],
)
cc_library(
name = "ViewLikeInterface",
srcs = ["lib/Interfaces/ViewLikeInterface.cpp"],
hdrs = ["include/mlir/Interfaces/ViewLikeInterface.h"],
includes = ["include"],
deps = [
[Bazel] Update for 0813700de1 Updates Bazel BUILD files for changes in https://github.com/llvm/llvm-project/commit/0813700de1 Differential Revision: https://reviews.llvm.org/D104990
2021-06-27 08:30:54 -07:00
":DialectUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":ViewLikeInterfaceIncGen",
],
)
cc_library(
name = "CopyOpInterface",
srcs = ["lib/Interfaces/CopyOpInterface.cpp"],
hdrs = ["include/mlir/Interfaces/CopyOpInterface.h"],
includes = ["include"],
deps = [
":CopyOpInterfaceIncGen",
":IR",
],
)
td_library(
name = "ShapeOpsTdFiles",
srcs = [
"include/mlir/Dialect/Shape/IR/ShapeBase.td",
"include/mlir/Dialect/Shape/IR/ShapeOps.td",
],
includes = ["include"],
deps = [
[mlir] Add shape.func Add shape func op for use (primarily) in shape function_library op. Allows setting default dialect for some simpler authoring. This is a minimal version of the ops needed. Differential Revision: https://reviews.llvm.org/D124055
2022-04-22 11:35:34 -07:00
":CallInterfacesTdFiles",
Add missing BUILD dependency to ShapeOps
2022-02-08 09:00:03 +01:00
":CastInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ControlFlowInterfacesTdFiles",
[mlir] Add shape.func Add shape func op for use (primarily) in shape function_library op. Allows setting default dialect for some simpler authoring. This is a minimal version of the ops needed. Differential Revision: https://reviews.llvm.org/D124055
2022-04-22 11:35:34 -07:00
":FunctionInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":InferTypeOpInterfaceTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "ShapeOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Shape/IR/ShapeOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Shape/IR/ShapeOps.cpp.inc",
),
(
["-gen-dialect-decls"],
"include/mlir/Dialect/Shape/IR/ShapeOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
"include/mlir/Dialect/Shape/IR/ShapeOpsDialect.cpp.inc",
),
[mlir][shape] Switch types to ODS generated (NFC) These were already pretty simple, so just switching to generated.
2022-06-25 09:06:52 -07:00
(
["-gen-typedef-decls"],
"include/mlir/Dialect/Shape/IR/ShapeOpsTypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/Shape/IR/ShapeOpsTypes.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Shape/IR/ShapeOps.td",
deps = [":ShapeOpsTdFiles"],
)
gentbl_cc_library(
name = "MLIRShapeCanonicalizationIncGen",
strip_include_prefix = "include/mlir/Dialect/Shape/IR",
tbl_outs = [
(
["-gen-rewriters"],
"include/mlir/Dialect/Shape/IR/ShapeCanonicalization.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "lib/Dialect/Shape/IR/ShapeCanonicalization.td",
deps = [
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ShapeOpsTdFiles",
":TensorOpsTdFiles",
],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "ShapeDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = glob(["lib/Dialect/Shape/IR/*.cpp"]),
hdrs = ["include/mlir/Dialect/Shape/IR/Shape.h"],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ControlFlowInterfaces",
":Dialect",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":InferTypeOpInterface",
":MLIRShapeCanonicalizationIncGen",
":ShapeOpsIncGen",
":SideEffectInterfaces",
":TensorDialect",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "ShapeToStandardGen",
strip_include_prefix = "lib/Conversion/ShapeToStandard",
tbl_outs = [
(
["-gen-rewriters"],
"lib/Conversion/ShapeToStandard/ShapeToStandard.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "lib/Conversion/ShapeToStandard/ShapeToStandard.td",
deps = [":ShapeOpsTdFiles"],
)
cc_library(
name = "ShapeToStandard",
srcs = glob([
"lib/Conversion/ShapeToStandard/*.cpp",
"lib/Conversion/ShapeToStandard/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = ["include/mlir/Conversion/ShapeToStandard/ShapeToStandard.h"],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":MemRefDialect",
":Pass",
":SCFDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ShapeDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ShapeToStandardGen",
":Support",
":TensorDialect",
":Transforms",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "ShapeTransformsPassIncGen",
strip_include_prefix = "include",
tbl_outs = [(
[
"-gen-pass-decls",
"-name=Shape",
],
"include/mlir/Dialect/Shape/Transforms/Passes.h.inc",
)],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Shape/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "ShapeTransforms",
srcs = glob([
"lib/Dialect/Shape/Transforms/*.cpp",
"lib/Dialect/Shape/Transforms/*.h",
]),
[mlir][shape] Migrate bufferization to BufferizableOpInterface Differential Revision: https://reviews.llvm.org/D121043
2022-03-07 21:27:53 +09:00
hdrs = [
[Bazel] fixes for 9f77909.
2022-10-03 09:06:42 +02:00
"include/mlir/Dialect/Shape/Analysis/ShapeMappingAnalysis.h",
[mlir][shape] Migrate bufferization to BufferizableOpInterface Differential Revision: https://reviews.llvm.org/D121043
2022-03-07 21:27:53 +09:00
"include/mlir/Dialect/Shape/Transforms/BufferizableOpInterfaceImpl.h",
"include/mlir/Dialect/Shape/Transforms/Passes.h",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationDialect",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":BufferizationTransforms",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":MemRefDialect",
":Pass",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ShapeDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ShapeTransformsPassIncGen",
[Bazel] fixes for 9f77909.
2022-10-03 09:06:42 +02:00
":TensorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
[Bazel] fixes for 9f77909.
2022-10-03 09:06:42 +02:00
"//llvm:Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
td_library(
name = "ControlFlowOpsTdFiles",
srcs = [
"include/mlir/Dialect/ControlFlow/IR/ControlFlowOps.td",
],
includes = ["include"],
deps = [
[mlir] Fix BUILD issues and dependencies. Differential Revision: https://reviews.llvm.org/D123868
2022-04-15 18:59:03 +00:00
":AttrTdFiles",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":CallInterfacesTdFiles",
":CastInterfacesTdFiles",
":ControlFlowInterfacesTdFiles",
":InferTypeOpInterfaceTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "ControlFlowOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/ControlFlow/IR/ControlFlowOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/ControlFlow/IR/ControlFlowOps.cpp.inc",
),
(
["-gen-dialect-decls"],
"include/mlir/Dialect/ControlFlow/IR/ControlFlowOpsDialect.h.inc",
),
(
["-gen-dialect-defs"],
"include/mlir/Dialect/ControlFlow/IR/ControlFlowOpsDialect.cpp.inc",
),
(
["-gen-enum-decls"],
"include/mlir/Dialect/ControlFlow/IR/ControlFlowOpsEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/ControlFlow/IR/ControlFlowOpsEnums.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/ControlFlow/IR/ControlFlowOps.td",
[mlir] Fix BUILD issues and dependencies. Differential Revision: https://reviews.llvm.org/D123868
2022-04-15 18:59:03 +00:00
deps = [
":AttrTdFiles",
":ControlFlowOpsTdFiles",
],
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
)
cc_library(
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
name = "ControlFlowDialect",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
srcs = glob(
[
"lib/Dialect/ControlFlow/IR/*.cpp",
"lib/Dialect/ControlFlow/IR/*.h",
],
),
hdrs = glob([
"include/mlir/Dialect/ControlFlow/IR/*.h",
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":CommonFolders",
":ControlFlowInterfaces",
":ControlFlowOpsIncGen",
":IR",
":SideEffectInterfaces",
":Support",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
name = "FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = glob(
[
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"lib/Dialect/Func/IR/*.cpp",
"lib/Dialect/Func/IR/*.h",
"lib/Dialect/Func/Utils/*.cpp",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
),
hdrs = glob([
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir/Dialect/Func/IR/*.h",
"include/mlir/Dialect/Func/Utils/*.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
]) + ["include/mlir/Transforms/InliningUtils.h"],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":CallOpInterfaces",
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":CommonFolders",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ControlFlowInterfaces",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
":FuncIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfaces",
":Support",
"//llvm:Support",
],
)
[bazel][NFC] Add Dialect/Func/Extensions library and deps Added in D120368
2023-06-09 17:04:41 -07:00
cc_library(
name = "FuncExtensions",
srcs = glob(["lib/Dialect/Func/Extensions/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/Func/Extensions/*.h"]),
includes = ["include"],
deps = [
":ControlFlowDialect",
":FuncDialect",
":IR",
":InferTypeOpInterface",
],
)
cc_library(
name = "AllExtensions",
hdrs = ["include/mlir/InitAllExtensions.h"],
deps = [
":FuncExtensions",
],
)
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
# TODO(zinenko): remove this after updating users.
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
name = "FuncTransformsPassIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
strip_include_prefix = "include",
tbl_outs = [(
[
"-gen-pass-decls",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"-name=Func",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"include/mlir/Dialect/Func/Transforms/Passes.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)],
tblgen = ":mlir-tblgen",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
td_file = "include/mlir/Dialect/Func/Transforms/Passes.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [":PassBaseTdFiles"],
)
cc_library(
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
name = "FuncTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = glob([
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"lib/Dialect/Func/Transforms/*.cpp",
"lib/Dialect/Func/Transforms/*.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
]),
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
hdrs = glob(["include/mlir/Dialect/Func/Transforms/*.h"]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationDialect",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":BufferizationTransforms",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
":FuncTransformsPassIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Add more mlir dependencies after D124298 The Bazel layering_check feature compiles libraries with `-fmodule-name=X -fmodules-strict-decluse` which require #include to be in deps.
2022-04-23 13:06:54 -07:00
":MemRefDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":SCFDialect",
":Support",
":Transforms",
"//llvm:Support",
],
)
cc_library(
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
name = "VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = glob(
[
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
"lib/Dialect/Vector/IR/*.cpp",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
),
hdrs = glob([
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
"include/mlir/Dialect/Vector/IR/*.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[mlir][vector] Add operations used for Vector distribution Add vector op warp_execute_on_lane_0 that will be used to do incremental vector distribution in order to target warp level vector programming for architectures with GPU-like SIMT programming model. The idea behing the op is discussed further on discourse: https://discourse.llvm.org/t/vector-vector-distribution-large-vector-to-small-vector/1983/23 Differential Revision: https://reviews.llvm.org/D123703
2022-04-13 18:38:11 +00:00
":ControlFlowInterfaces",
[MLIR][Vector] Add a dependency on `:DataLayoutInterfaces` Reviewed By: tpopp Differential Revision: https://reviews.llvm.org/D144631
2023-02-23 12:11:04 +01:00
":DataLayoutInterfaces",
[mlir][vector][bufferize] Implement DestinationStyleOpInterface on TransferWriteOp This simplifies the BufferizableOpInterface implementation of vector.transfer_write. Differential Revision: https://reviews.llvm.org/D136348
2022-10-27 10:59:52 +02:00
":DestinationStyleOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":DialectUtils",
":IR",
[mlir][Bazel] Add missing dependencies. When building with layering_check enabled, there needs to be a dependency for each header include.
2022-04-25 09:16:16 +02:00
":InferTypeOpInterface",
[mlir][Vector] Introduce the MaskingOpInterface This MaskingOpInterface provides masking cababilitites to those operations that implement it. For only is only implemented by the `vector.mask` operation and it's used to break the dependency between the Vector dialect (where the `vector.mask` op lives) and operations implementing the MaskableOpInterface. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D136734
2022-10-26 00:58:56 +00:00
":MaskableOpInterface",
":MaskingOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MemRefDialect",
":SideEffectInterfaces",
":Support",
":TensorDialect",
":VectorInterfaces",
":VectorOpsIncGen",
":ViewLikeInterface",
"//llvm:Support",
],
)
[mlir][Vector] Vector transform skeleton Differential Revision: https://reviews.llvm.org/D134722
2022-06-21 06:52:01 -07:00
cc_library(
name = "VectorTransformOps",
srcs = [
"lib/Dialect/Vector/TransformOps/VectorTransformOps.cpp",
],
hdrs = [
"include/mlir/Dialect/Vector/TransformOps/VectorTransformOps.h",
],
includes = ["include"],
deps = [
":AffineDialect",
":ArithDialect",
":AsmParser",
":IR",
[mlir][Vector][Transforms] Improve the control over individual vector lowerings and transforms This revision adds vector transform operations that allow us to better inspect the composition of various lowerings that were previously very opaque. This commit is NFC in that it does not change patterns beyond adding `rewriter.notifyFailure` messages and it does not change the tests beyond breaking them into pieces and using transforms instead of throwaway opaque test passes. Reviewed By: ftynse, springerm Co-authored-by: Alex Zinenko <zinenko@google.com> Differential Revision: https://reviews.llvm.org/D146755
2023-03-24 12:09:50 +00:00
":LLVMDialect",
[mlir][Vector] Vector transform skeleton Differential Revision: https://reviews.llvm.org/D134722
2022-06-21 06:52:01 -07:00
":SideEffectInterfaces",
":TransformDialect",
":TransformUtils",
[mlir][Vector] Add a LowerVectorsOp to VectorTransformOps This op significantly improves transfor dialect usage when using vector abstractions. It also brings us closer to writing simple end-to-end unit tests that guard against subtle regressions in how patterns combine. Differential Revision: https://reviews.llvm.org/D138723
2022-11-25 08:13:08 -08:00
":VectorDialect",
Fix bazel build overlay.
2023-01-17 13:17:56 +01:00
":VectorEnumsIncGen",
[mlir][Vector] Add a LowerVectorsOp to VectorTransformOps This op significantly improves transfor dialect usage when using vector abstractions. It also brings us closer to writing simple end-to-end unit tests that guard against subtle regressions in how patterns combine. Differential Revision: https://reviews.llvm.org/D138723
2022-11-25 08:13:08 -08:00
":VectorToSCF",
":VectorTransformOpsIncGen",
":VectorTransforms",
":X86VectorTransforms",
[mlir][Vector] Vector transform skeleton Differential Revision: https://reviews.llvm.org/D134722
2022-06-21 06:52:01 -07:00
"//llvm:Support",
],
)
Fix bazel build overlay.
2023-01-17 13:17:56 +01:00
td_library(
name = "VectorTransformsTdFiles",
srcs = ["include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
gentbl_cc_library(
name = "VectorEnumsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-enum-decls"],
"include/mlir/Dialect/Vector/Transforms/VectorTransformsEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/Vector/Transforms/VectorTransformsEnums.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Vector/Transforms/VectorTransformsBase.td",
deps = [
":OpBaseTdFiles",
],
)
[mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg Bufferize Differential Revision: https://reviews.llvm.org/D119444
2022-02-15 21:16:50 +09:00
gentbl_cc_library(
name = "VectorPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Vector",
],
"include/mlir/Dialect/Vector/Transforms/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Vector/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
cc_library(
name = "VectorTransforms",
srcs = glob(
[
"lib/Dialect/Vector/Transforms/*.cpp",
[mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg Bufferize Differential Revision: https://reviews.llvm.org/D119444
2022-02-15 21:16:50 +09:00
"lib/Dialect/Vector/Transforms/*.h",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
],
),
hdrs = glob([
"include/mlir/Dialect/Vector/Transforms/*.h",
]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Fix bazel build for 5a1cdcbd8698cd263696b38e2672fccac9ec793c
2023-07-11 14:26:32 -07:00
":ArithTransforms",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithUtils",
[mlir][bufferize][NFC] Move vector BufferizableOpInterface impl to vector dialect Differential Revision: https://reviews.llvm.org/D118540
2022-02-01 00:27:13 +09:00
":BufferizationDialect",
[mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg Bufferize Differential Revision: https://reviews.llvm.org/D119444
2022-02-15 21:16:50 +09:00
":BufferizationTransforms",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":DialectUtils",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":IR",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":MemRefDialect",
[mlir][MemRef] Move narrow type emulation common methods to MemRefUtils. It also unifies the computation of StridedLayoutAttr. If the stride is static known value, we can just use it. Differential Revision: https://reviews.llvm.org/D155017
2023-07-11 14:13:07 -07:00
":MemRefUtils",
[mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg Bufferize Differential Revision: https://reviews.llvm.org/D119444
2022-02-15 21:16:50 +09:00
":Pass",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":SCFDialect",
[mlir] Remove `Transforms/SideEffectUtils.h` and move the methods into `Interface/SideEffectInterfaces.h`. The methods in `SideEffectUtils.h` (and their implementations in `SideEffectUtils.cpp`) seem to have similar intent to methods already existing in `SideEffectInterfaces.h`. Move the decleration (and implementation) from `SideEffectUtils.h` (and `SideEffectUtils.cpp`) into `SideEffectInterfaces.h` (and `SideEffectInterface.cpp`). Also drop the `SideEffectInterface::hasNoEffect` method in favor of `mlir::isMemoryEffectFree` which actually recurses into the operation instead of just relying on the `hasRecursiveMemoryEffectTrait` exclusively. Differential Revision: https://reviews.llvm.org/D137857
2022-11-11 17:10:38 +00:00
":SideEffectInterfaces",
[mlir][bazel] fix bazel build on VectorTransforms
2022-06-06 21:51:11 +00:00
":Support",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":TensorDialect",
[mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg Bufferize Differential Revision: https://reviews.llvm.org/D119444
2022-02-15 21:16:50 +09:00
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Fix bazel build overlay.
2023-01-17 13:17:56 +01:00
":VectorEnumsIncGen",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":VectorInterfaces",
[mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg Bufferize Differential Revision: https://reviews.llvm.org/D119444
2022-02-15 21:16:50 +09:00
":VectorPassIncGen",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":VectorUtils",
"//llvm:Support",
],
)
cc_library(
name = "VectorUtils",
srcs = glob(
[
"lib/Dialect/Vector/Utils/*.cpp",
],
),
hdrs = glob([
"include/mlir/Dialect/Vector/Utils/*.h",
]),
includes = ["include"],
deps = [
":AffineAnalysis",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[bazel] Add missing dependencies after 7a69a9d7ae
2022-11-23 10:45:15 +01:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":IR",
":MemRefDialect",
":Support",
":TensorDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "Support",
[MLIR] Fix bazel BUILD file after Parser.h moved. See D121024, D121025, D121026. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D121094
2022-03-07 13:04:05 +01:00
srcs = glob([
"lib/Support/*.cpp",
"lib/Support/*.h",
]),
hdrs = glob(["include/mlir/Support/*.h"]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
[bazel] add TargetParser
2022-12-20 20:09:00 +01:00
deps = [
"//llvm:Support",
"//llvm:TargetParser",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
[bazel] Fix build after 28d04c5
2023-03-06 17:24:26 +01:00
cc_library(
name = "Debug",
srcs = glob([
"lib/Debug/*.cpp",
"lib/Debug/*.h",
[bazel][mlir] Fold :Observers and :BreakpointManagers into :Debug To avoid circular deps. While at it, add missing dep on CAPIIR.
2023-04-25 00:53:03 -04:00
"lib/Debug/BreakpointManagers/*.cpp",
"lib/Debug/BreakpointManagers/*.h",
"lib/Debug/Observers/*.cpp",
"lib/Debug/Observers/*.h",
[bazel] Fix build after 28d04c5
2023-03-06 17:24:26 +01:00
]),
[bazel] Fix bazel build for 7f069f5ef4fee00520ed0c350dca42c3c4b72b61
2023-04-22 13:59:45 +02:00
hdrs = glob([
"include/mlir/Debug/*.h",
[bazel][mlir] Fold :Observers and :BreakpointManagers into :Debug To avoid circular deps. While at it, add missing dep on CAPIIR.
2023-04-25 00:53:03 -04:00
"include/mlir/Debug/BreakpointManagers/*.h",
"include/mlir/Debug/Observers/*.h",
[mlir][Bazel] Adjust BUILD file for 930744fcdad7b326dd0337622b6e8dc99efcfa60
2023-03-20 15:10:12 +01:00
]),
includes = ["include"],
deps = [
[bazel][mlir] Fold :Observers and :BreakpointManagers into :Debug To avoid circular deps. While at it, add missing dep on CAPIIR.
2023-04-25 00:53:03 -04:00
":CAPIIR",
[bazel] Fix build after 28d04c5
2023-03-06 17:24:26 +01:00
":IR",
[bazel][mlir] Fold :Observers and :BreakpointManagers into :Debug To avoid circular deps. While at it, add missing dep on CAPIIR.
2023-04-25 00:53:03 -04:00
":Support",
[bazel] Fix build after 28d04c5
2023-03-06 17:24:26 +01:00
"//llvm:Support",
],
)
[mlir][Bazel] Adapt BUILD.bazel for move of lsp-server support files. Differential Revision: https://reviews.llvm.org/D121891
2022-03-17 09:52:56 +01:00
cc_library(
name = "MlirLspServerSupportLib",
srcs = glob(
[
"lib/Tools/lsp-server-support/*.cpp",
],
),
hdrs = glob(
[
[bazel] Adjust lsp-server-support include paths for 305d71853965
2023-01-28 11:04:57 +01:00
"include/mlir/Tools/lsp-server-support/*.h",
[mlir][Bazel] Adapt BUILD.bazel for move of lsp-server support files. Differential Revision: https://reviews.llvm.org/D121891
2022-03-17 09:52:56 +01:00
],
),
deps = [
":Support",
[mlir][Bazel] Add missing dependency. Differential Revision: https://reviews.llvm.org/D121893
2022-03-17 10:24:59 +01:00
"//llvm:Support",
[bazel] add TargetParser
2022-12-20 20:09:00 +01:00
"//llvm:TargetParser",
[mlir][Bazel] Adapt BUILD.bazel for move of lsp-server support files. Differential Revision: https://reviews.llvm.org/D121891
2022-03-17 09:52:56 +01:00
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "MlirLspServerLib",
srcs = glob(
[
"lib/Tools/mlir-lsp-server/*.cpp",
"lib/Tools/mlir-lsp-server/*.h",
"lib/Tools/mlir-lsp-server/lsp/*.cpp",
"lib/Tools/mlir-lsp-server/lsp/*.h",
],
),
hdrs = glob(
["include/mlir/Tools/mlir-lsp-server/*.h"],
),
includes = ["include"],
deps = [
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
":AsmParser",
[bazel] Port f7b8a70e7a1738e0fc6574e3cf8faa4fa1f34eba
2022-09-03 09:55:20 +02:00
":BytecodeWriter",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir][Bazel] Adapt BUILD.bazel for move of lsp-server support files. Differential Revision: https://reviews.llvm.org/D121891
2022-03-17 09:52:56 +01:00
":MlirLspServerSupportLib",
[bazel] Port f7b8a70e7a1738e0fc6574e3cf8faa4fa1f34eba
2022-09-03 09:55:20 +02:00
":Parser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
"//llvm:Support",
],
)
[bazel][mlir] Add MLIR PDLL LSP server target Add targets for PDLL LSP server.
2022-03-19 17:53:37 -07:00
cc_library(
name = "MlirPdllLspServerLib",
srcs = glob(
[
"lib/Tools/mlir-pdll-lsp-server/*.cpp",
"lib/Tools/mlir-pdll-lsp-server/*.h",
],
),
hdrs = glob(
["include/mlir/Tools/mlir-pdll-lsp-server/*.h"],
),
includes = ["include"],
deps = [
Fix bazel build After 1c2edb026ed67ddbb30ebe3e2d2f4f17a882a881
2022-05-31 10:15:12 +02:00
":IR",
[bazel][mlir] Add MLIR PDLL LSP server target Add targets for PDLL LSP server.
2022-03-19 17:53:37 -07:00
":MlirLspServerSupportLib",
":PDLLAST",
Fix bazel build After 1c2edb026ed67ddbb30ebe3e2d2f4f17a882a881
2022-05-31 10:15:12 +02:00
":PDLLCodeGen",
[bazel][mlir] Add MLIR PDLL LSP server target Add targets for PDLL LSP server.
2022-03-19 17:53:37 -07:00
":PDLLODS",
":PDLLParser",
":Support",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
name = "AsmParserTokenKinds",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
# strip_include_prefix does not apply to textual_hdrs.
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
hdrs = ["lib/AsmParser/TokenKinds.def"],
strip_include_prefix = "lib/AsmParser",
textual_hdrs = ["lib/AsmParser/TokenKinds.def"],
)
cc_library(
name = "AsmParser",
srcs = glob([
"lib/AsmParser/*.cpp",
"lib/AsmParser/*.h",
]),
hdrs = glob([
"include/mlir/AsmParser/*.h",
]),
includes = ["include"],
deps = [
":AsmParserTokenKinds",
":IR",
":Support",
"//llvm:Support",
[bazel] add TargetParser
2022-12-20 20:09:00 +01:00
"//llvm:TargetParser",
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
[mlir][Bazel] Fix bazel build Add fixes for https://reviews.llvm.org/D131747
2022-08-22 12:06:47 +02:00
cc_library(
name = "BytecodeReader",
srcs = glob([
"lib/Bytecode/Reader/*.cpp",
"lib/Bytecode/Reader/*.h",
"lib/Bytecode/*.h",
]),
hdrs = glob([
"include/mlir/Bytecode/*.h",
]),
includes = ["include"],
deps = [
":AsmParser",
[mlir][bazel] Port for 660f714e26999d266232a1fbb02712bb879bd34e, second attempt.
2023-05-27 08:37:45 +02:00
":BytecodeOpInterfaceIncGen",
[mlir][Bazel] Fix bazel build Add fixes for https://reviews.llvm.org/D131747
2022-08-22 12:06:47 +02:00
":IR",
":Support",
"//llvm:Support",
],
)
cc_library(
name = "BytecodeWriter",
srcs = glob([
"lib/Bytecode/Writer/*.cpp",
"lib/Bytecode/Writer/*.h",
"lib/Bytecode/*.h",
]),
hdrs = glob([
"include/mlir/Bytecode/*.h",
]),
includes = ["include"],
deps = [
[mlir][bazel] Port for 660f714e26999d266232a1fbb02712bb879bd34e, second attempt.
2023-05-27 08:37:45 +02:00
":BytecodeOpInterfaceIncGen",
[mlir][Bazel] Fix bazel build Add fixes for https://reviews.llvm.org/D131747
2022-08-22 12:06:47 +02:00
":IR",
":Support",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "Parser",
srcs = glob([
"lib/Parser/*.cpp",
"lib/Parser/*.h",
]),
hdrs = glob([
"include/mlir/Parser/*.h",
[MLIR] Fix bazel BUILD file after Parser.h moved. See D121024, D121025, D121026. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D121094
2022-03-07 13:04:05 +01:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
":AsmParser",
[mlir][Bazel] Fix bazel build. To avoid a dependency cycle, add BytecodeImplementation.h header to the "IR" target.
2022-08-24 08:51:44 +02:00
":BytecodeReader",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":Support",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "LLVMDialectInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
[mlir] Port bazel for 3948f0a0b5e5fecddf315b8de321c6a44ee7ff5c
2023-02-28 09:32:30 +01:00
"include/mlir/Dialect/LLVMIR/LLVMInterfaces.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-op-interface-defs"],
[mlir] Port bazel for 3948f0a0b5e5fecddf315b8de321c6a44ee7ff5c
2023-02-28 09:32:30 +01:00
"include/mlir/Dialect/LLVMIR/LLVMInterfaces.cpp.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
[mlir] add an interface to support custom types in LLVM dialect pointers This may be necessary in partial multi-stage conversion when a container type from dialect A containing types from dialect B goes through the conversion where only dialect A is converted to the LLVM dialect. We will need to keep a pointer-to-non-LLVM type in the IR until a further conversion can convert dialect B types to LLVM types. Reviewed By: wsmoses Differential Revision: https://reviews.llvm.org/D106076
2021-07-15 18:16:07 +02:00
(
["-gen-type-interface-decls"],
"include/mlir/Dialect/LLVMIR/LLVMTypeInterfaces.h.inc",
),
(
["-gen-type-interface-defs"],
"include/mlir/Dialect/LLVMIR/LLVMTypeInterfaces.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
[mlir] Port bazel for 3948f0a0b5e5fecddf315b8de321c6a44ee7ff5c
2023-02-28 09:32:30 +01:00
td_file = "include/mlir/Dialect/LLVMIR/LLVMInterfaces.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [":LLVMOpsTdFiles"],
)
cc_library(
name = "LLVMDialect",
srcs = glob(
[
"lib/Dialect/LLVMIR/IR/*.cpp",
"lib/Dialect/LLVMIR/IR/*.h",
],
exclude = [
"lib/Dialect/LLVMIR/IR/*AMX*.cpp",
"lib/Dialect/LLVMIR/IR/*AMX*.h",
"lib/Dialect/LLVMIR/IR/*ArmSVE*.cpp",
"lib/Dialect/LLVMIR/IR/*ArmSVE*.h",
"lib/Dialect/LLVMIR/IR/NVVM*.cpp",
"lib/Dialect/LLVMIR/IR/NVVM*.h",
"lib/Dialect/LLVMIR/IR/ROCDL*.cpp",
"lib/Dialect/LLVMIR/IR/ROCDL*.h",
"lib/Dialect/LLVMIR/IR/*X86Vector*.cpp",
"lib/Dialect/LLVMIR/IR/*X86Vector*.h",
],
),
hdrs = glob(
["include/mlir/Dialect/LLVMIR/*.h"],
exclude = [
"include/mlir/Dialect/LLVMIR/*AMX*.h",
"include/mlir/Dialect/LLVMIR/*ArmSVE*.h",
"include/mlir/Dialect/LLVMIR/NVVM*.h",
"include/mlir/Dialect/LLVMIR/ROCDL*.h",
"include/mlir/Dialect/LLVMIR/*X86Vector*.h",
],
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
) + [
"include/mlir/Transforms/InliningUtils.h",
"include/mlir/Transforms/Mem2Reg.h",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir][Bazel] Fix Bazel build after a2158374ba1a6f81f4cce3eb54d0bc44f3ab75e0
2022-07-06 08:47:21 +02:00
":CallOpInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ControlFlowInterfaces",
":DataLayoutInterfaces",
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
":IR",
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialectInterfaceIncGen",
[mlir] Split intrinsics out of LLVMOps.td Move the operations that correspond to LLVM IR intrinsics in a separate .td file. This makes it easier to maintain the intrinsics and decreases the compile time of LLVMDialect.cpp by ~25%. Depends On D123310 Reviewed By: wsmoses, jacquesguan Differential Revision: https://reviews.llvm.org/D123315
2022-04-13 18:29:47 +02:00
":LLVMIntrinsicOpsIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMOpsIncGen",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
":LLVMTypesIncGen",
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
":MemorySlotInterfaces",
":MemorySlotInterfacesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfaces",
":Support",
"//llvm:AsmParser",
[mlir][Bazel] Port 3bef1e0f4c1d7fd901b58abe0c5c0e67da764ba2
2022-10-24 09:10:30 +02:00
"//llvm:BinaryFormat",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:BitReader",
"//llvm:BitWriter",
"//llvm:Core",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "LLVMPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=LLVM",
],
"include/mlir/Dialect/LLVMIR/Transforms/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "LLVMIRTransforms",
srcs = glob([
"lib/Dialect/LLVMIR/Transforms/*.cpp",
"lib/Dialect/LLVMIR/Transforms/*.h",
]),
hdrs = glob(["include/mlir/Dialect/LLVMIR/Transforms/*.h"]),
includes = ["include"],
deps = [
[mlir] replace 'emit_c_wrappers' func->llvm conversion option with a pass The 'emit_c_wrappers' option in the FuncToLLVM conversion requests C interface wrappers to be emitted for every builtin function in the module. While this has been useful to bootstrap the interface, it is problematic in the longer term as it may unintentionally affect the functions that should retain their existing interface, e.g., libm functions obtained by lowering math operations (see D126964 for an example). Since D77314, we have a finer-grain control over interface generation via an attribute that avoids the problem entirely. Remove the 'emit_c_wrappers' option. Introduce the '-llvm-request-c-wrappers' pass that can be run in any pipeline that needs blanket emission of functions to annotate all builtin functions with the attribute before performing the usual lowering that accounts for the attribute. Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D127952
2022-06-16 12:53:49 +02:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":LLVMDialect",
":LLVMPassIncGen",
Recommit: "[MLIR][NVVM] Replace fdiv on fp16 with promoted (fp32) multiplication with reciprocal plus one (conditional) Newton iteration." This change rolls bcfc0a9051014437b55ab932d9aca5ecdca6776b forward (i.e., reverting 369ce54bb302f209239b8ebc77ad824add9df089) with fixed CMakeLists.txt.
2022-06-04 12:33:42 +02:00
":NVVMDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
Recommit: "[MLIR][NVVM] Replace fdiv on fp16 with promoted (fp32) multiplication with reciprocal plus one (conditional) Newton iteration." This change rolls bcfc0a9051014437b55ab932d9aca5ecdca6776b forward (i.e., reverting 369ce54bb302f209239b8ebc77ad824add9df089) with fixed CMakeLists.txt.
2022-06-04 12:33:42 +02:00
":Transforms",
[mlir][Bazel] Add missing dependencies for 72429a42ac33564fa82449d99dc234da32a05498
2023-02-15 08:30:57 +01:00
"//llvm:BinaryFormat",
"//llvm:Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
td_library(
name = "GPUOpsTdFiles",
srcs = [
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUBase.td",
"include/mlir/Dialect/GPU/IR/GPUOps.td",
"include/mlir/Dialect/GPU/IR/ParallelLoopMapperAttr.td",
[mlir] Introduce device mapper attribute for `thread_dim_map` and `mapped to dims` `scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [0, 1]} } { thread_dim_mapping = [0, 1]} ``` It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation. The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]} } { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]} ``` Reviewed By: ftynse, nicolasvasilache Differential Revision: https://reviews.llvm.org/D137413
2022-11-10 17:55:49 +01:00
"include/mlir/Dialect/GPU/TransformOps/GPUDeviceMappingAttr.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
":DLTIDialectTdFiles",
":DataLayoutInterfacesTdFiles",
[bazel] Port 7ceffae18c43
2022-01-19 14:14:36 +01:00
":FunctionInterfacesTdFiles",
[mlir] Add InferIntRangeInterface to gpu.launch Infers block/grid dimensions/indices or ranges of such dimensions/indices. Reviewed By: krzysz00 Differential Revision: https://reviews.llvm.org/D129036
2022-07-04 08:11:30 +02:00
":InferIntRangeInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMOpsTdFiles",
":OpBaseTdFiles",
[mlir] Introduce device mapper attribute for `thread_dim_map` and `mapped to dims` `scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [0, 1]} } { thread_dim_mapping = [0, 1]} ``` It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation. The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]} } { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]} ``` Reviewed By: ftynse, nicolasvasilache Differential Revision: https://reviews.llvm.org/D137413
2022-11-10 17:55:49 +01:00
":SCFTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfacesTdFiles",
],
)
[mlir] Introduce device mapper attribute for `thread_dim_map` and `mapped to dims` `scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [0, 1]} } { thread_dim_mapping = [0, 1]} ``` It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation. The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]} } { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]} ``` Reviewed By: ftynse, nicolasvasilache Differential Revision: https://reviews.llvm.org/D137413
2022-11-10 17:55:49 +01:00
gentbl_cc_library(
name = "GPUDeviceMapperEnumsGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-enum-decls"],
"include/mlir/Dialect/GPU/TransformOps/GPUDeviceMapperEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/GPU/TransformOps/GPUDeviceMapperEnums.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/GPU/TransformOps/GPUDeviceMappingAttr.td",
deps = [
":GPUOpsTdFiles",
":OpBaseTdFiles",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "GPUBaseIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOpInterfaces.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-op-interface-defs"],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOpInterfaces.cpp.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
],
tblgen = ":mlir-tblgen",
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
td_file = "include/mlir/Dialect/GPU/IR/GPUBase.td",
[mlir][Bazel] Fix wrong dependency for GPUBaseIncGen.
2022-02-23 10:52:24 +01:00
deps = [":OpBaseTdFiles"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
gentbl_cc_library(
name = "GPUOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
[mlir] Fix bazel build of GPU dialect
2022-01-18 22:39:24 +00:00
(
[
"-gen-dialect-decls",
"-dialect=gpu",
],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOpsDialect.h.inc",
[mlir] Fix bazel build of GPU dialect
2022-01-18 22:39:24 +00:00
),
(
[
"-gen-dialect-defs",
"-dialect=gpu",
],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOpsDialect.cpp.inc",
[mlir] Fix bazel build of GPU dialect
2022-01-18 22:39:24 +00:00
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-decls"],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOps.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-op-defs"],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOps.cpp.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
[mlir][gpu] Add basic support to do elementwise ops on mma matrix type In order to support fusion with mma matrix type we need to be able to execute elementwise operations on them. This add an op to be able to support some basic elementwise operations. This is a is not a full solution as it only supports a limited scope or operations. Ideally we would want to be able to fuse with more kind of operations. Differential Revision: https://reviews.llvm.org/D112857
2021-11-01 11:43:54 -07:00
(
["-gen-enum-decls"],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOpsEnums.h.inc",
[mlir][gpu] Add basic support to do elementwise ops on mma matrix type In order to support fusion with mma matrix type we need to be able to execute elementwise operations on them. This add an op to be able to support some basic elementwise operations. This is a is not a full solution as it only supports a limited scope or operations. Ideally we would want to be able to fuse with more kind of operations. Differential Revision: https://reviews.llvm.org/D112857
2021-11-01 11:43:54 -07:00
),
(
["-gen-enum-defs"],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOpsEnums.cpp.inc",
[mlir][gpu] Add basic support to do elementwise ops on mma matrix type In order to support fusion with mma matrix type we need to be able to execute elementwise operations on them. This add an op to be able to support some basic elementwise operations. This is a is not a full solution as it only supports a limited scope or operations. Ideally we would want to be able to fuse with more kind of operations. Differential Revision: https://reviews.llvm.org/D112857
2021-11-01 11:43:54 -07:00
),
[mlir] Replace StrEnumAttr -> EnumAttr in core dialects Removes uses of `StrEnumAttr` in core dialects Reviewed By: mehdi_amini, rriddle Differential Revision: https://reviews.llvm.org/D117514
2022-01-18 16:53:55 +00:00
(
["-gen-attrdef-decls"],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOpsAttributes.h.inc",
[mlir] Replace StrEnumAttr -> EnumAttr in core dialects Removes uses of `StrEnumAttr` in core dialects Reviewed By: mehdi_amini, rriddle Differential Revision: https://reviews.llvm.org/D117514
2022-01-18 16:53:55 +00:00
),
(
["-gen-attrdef-defs"],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/IR/GPUOpsAttributes.cpp.inc",
[mlir] Replace StrEnumAttr -> EnumAttr in core dialects Removes uses of `StrEnumAttr` in core dialects Reviewed By: mehdi_amini, rriddle Differential Revision: https://reviews.llvm.org/D117514
2022-01-18 16:53:55 +00:00
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
td_file = "include/mlir/Dialect/GPU/IR/GPUOps.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [
":DLTIDialectTdFiles",
":GPUOpsTdFiles",
],
)
cc_library(
name = "GPUDialect",
srcs = glob(
[
"lib/Dialect/GPU/IR/*.cpp",
"lib/Dialect/GPU/IR/*.h",
],
),
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
hdrs = glob(["include/mlir/Dialect/GPU/IR/*.h"]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":DLTIDialect",
":GPUBaseIncGen",
":GPUOpsIncGen",
":IR",
[mlir] Add InferIntRangeInterface to gpu.launch Infers block/grid dimensions/indices or ranges of such dimensions/indices. Reviewed By: krzysz00 Differential Revision: https://reviews.llvm.org/D129036
2022-07-04 08:11:30 +02:00
":InferIntRangeInterface",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
":MemRefDialect",
[mlir] Introduce device mapper attribute for `thread_dim_map` and `mapped to dims` `scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [0, 1]} } { thread_dim_mapping = [0, 1]} ``` It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation. The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]} } { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]} ``` Reviewed By: ftynse, nicolasvasilache Differential Revision: https://reviews.llvm.org/D137413
2022-11-10 17:55:49 +01:00
":SCFDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfaces",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "GPUPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=GPU",
],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/Transforms/Passes.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
[
"-gen-pass-capi-header",
"--prefix=GPU",
],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/Transforms/Passes.capi.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
[
"-gen-pass-capi-impl",
"--prefix=GPU",
],
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
"include/mlir/Dialect/GPU/Transforms/Passes.capi.cpp.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
],
tblgen = ":mlir-tblgen",
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
td_file = "include/mlir/Dialect/GPU/Transforms/Passes.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "GPUTransforms",
srcs = glob(
[
"lib/Dialect/GPU/Transforms/*.cpp",
"lib/Dialect/GPU/Transforms/*.h",
],
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
exclude = [
"lib/Dialect/GPU/Transforms/SerializeToCubin.cpp",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
[mlir][gpu] Move GPU headers into IR/ and Transforms/ Depends on D127350 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127352
2022-06-09 21:33:41 +00:00
hdrs = glob(["include/mlir/Dialect/GPU/Transforms/*.h"]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
[mlir][bazel] NFC: change MLIR_GPU_TO_CUBIN_PASS_ENABLE from `defines` to `local_defines`.
2022-11-14 12:21:59 +01:00
local_defines = if_cuda_available(["MLIR_GPU_TO_CUBIN_PASS_ENABLE"]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [
[mlir][bazel] reformat BUILD.bazel with buildifier Reviewed By: jreiffers Differential Revision: https://reviews.llvm.org/D147244
2023-03-30 10:10:57 -07:00
":AffineUtils",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
":AsmParser",
[mlir] Sort the libraties in BUILD.bazel.
2022-07-26 16:32:40 +02:00
":AsyncDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
[mlir][GPU] Extend GPU kernel outlining to generate DL specification This patch extends the GPU kernel outlining pass so that it can take in an optional data layout specification that will be attached to the GPU module operation generated. If the data layout specification is not provided the default data layout is used instead. Reviewed By: herhut, mehdi_amini Differential Revision: https://reviews.llvm.org/D115722
2021-12-16 09:47:41 +00:00
":DLTIDialect",
[bazel] add missing dep for GPUTransforms
2023-06-05 09:20:08 +02:00
":ExecutionEngineUtils",
[mlir][bazel] reformat BUILD.bazel with buildifier Reviewed By: jreiffers Differential Revision: https://reviews.llvm.org/D147244
2023-03-30 10:10:57 -07:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":GPUPassIncGen",
[mlir][bazel] reformat BUILD.bazel with buildifier Reviewed By: jreiffers Differential Revision: https://reviews.llvm.org/D147244
2023-03-30 10:10:57 -07:00
":GPUToLLVMIRTranslation",
[bazel] add missing dep for GPUTransforms
2023-06-05 09:20:08 +02:00
":IR",
[mlir][Bazel] Adjust BUILD files for a9d003ef855ff7ed1bf4f8229ee9944b55936e6f
2023-06-05 09:57:01 +02:00
":IndexDialect",
[mlir][bazel] reformat BUILD.bazel with buildifier Reviewed By: jreiffers Differential Revision: https://reviews.llvm.org/D147244
2023-03-30 10:10:57 -07:00
":LLVMToLLVMIRTranslation",
":MemRefDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":ROCDLToLLVMIRTranslation",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SCFDialect",
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
":SerializeToCubin_stub",
[mlir] Remove `Transforms/SideEffectUtils.h` and move the methods into `Interface/SideEffectInterfaces.h`. The methods in `SideEffectUtils.h` (and their implementations in `SideEffectUtils.cpp`) seem to have similar intent to methods already existing in `SideEffectInterfaces.h`. Move the decleration (and implementation) from `SideEffectUtils.h` (and `SideEffectUtils.cpp`) into `SideEffectInterfaces.h` (and `SideEffectInterface.cpp`). Also drop the `SideEffectInterface::hasNoEffect` method in favor of `mlir::isMemoryEffectFree` which actually recurses into the operation instead of just relying on the `hasRecursiveMemoryEffectTrait` exclusively. Differential Revision: https://reviews.llvm.org/D137857
2022-11-11 17:10:38 +00:00
":SideEffectInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":ToLLVMIRTranslation",
[mlir][bazel] reformat BUILD.bazel with buildifier Reviewed By: jreiffers Differential Revision: https://reviews.llvm.org/D147244
2023-03-30 10:10:57 -07:00
":Transforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
Adds //mlir:GPUTransforms dependency to //llvm:MC as the former includes headers from the latter. Differential Revision: https://reviews.llvm.org/D111876
2021-10-15 12:29:48 +02:00
"//llvm:MC",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
"//llvm:Target",
] + if_cuda_available([
":NVVMToLLVMIRTranslation",
"//llvm:NVPTXCodeGen",
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
]),
)
cc_library(
name = "SerializeToCubin",
srcs = [
"lib/Dialect/GPU/Transforms/SerializeToCubin.cpp",
],
local_defines = if_cuda_available(["MLIR_GPU_TO_CUBIN_PASS_ENABLE"]),
deps = [
":GPUDialect",
":GPUPassIncGen",
":GPUTransforms",
[mlir][bazel] reformat BUILD.bazel with buildifier Reviewed By: jreiffers Differential Revision: https://reviews.llvm.org/D147244
2023-03-30 10:10:57 -07:00
":NVVMToLLVMIRTranslation",
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
":Pass",
":Support",
[mlir][bazel] reformat BUILD.bazel with buildifier Reviewed By: jreiffers Differential Revision: https://reviews.llvm.org/D147244
2023-03-30 10:10:57 -07:00
":ToLLVMIRTranslation",
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
"//llvm:Support",
] + if_cuda_available([
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"@cuda//:cuda_headers",
"@cuda//:libcuda",
[mlir][Bazel] Add missing dependencies.
2023-03-20 09:28:25 +01:00
]),
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
)
write_file(
name = "SerializeToCubin_stub_cc",
out = "SerializeToCubin_stub.cc",
content = [
[mlir][Bazel] Add missing dependencies.
2023-03-20 09:28:25 +01:00
"""
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
#include "mlir/Dialect/GPU/Transforms/Passes.h"
// Provide a weak registration stub in case the real SerializeToCubin is not
// linked in.
[mlir][Bazel] Avoid __attribute__((weak)) for MSVC. Differential Revision: https://reviews.llvm.org/D146405
2023-03-20 08:28:48 +01:00
#if defined(_MSC_VER)
// This might not work correctly, but it avoids a compilation error because
// MSVC does not support __attribute__((weak)).
void mlir::registerGpuSerializeToCubinPass() {}
#else
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
__attribute__((weak)) void mlir::registerGpuSerializeToCubinPass() {}
[mlir][Bazel] Avoid __attribute__((weak)) for MSVC. Differential Revision: https://reviews.llvm.org/D146405
2023-03-20 08:28:48 +01:00
#endif
[mlir][Bazel] Add missing dependencies.
2023-03-20 09:28:25 +01:00
""",
],
Separate SerializeToCubin from GPUTransforms. SerializeToCubin depends on CUDA at *runtime* which is undesirable for MLIR's general use case, as compilation should be doable on any host, regardless of whether it has a GPU. SerializeToCubin is needed to run some GPU tests, so when we build mlir-opt, SerializeToCubin pass is linked in directly into mlir-opt. Differential Revision: https://reviews.llvm.org/D146330
2023-03-17 13:47:46 -07:00
)
cc_library(
name = "SerializeToCubin_stub",
srcs = [":SerializeToCubin_stub_cc"],
hdrs = glob(["include/mlir/Dialect/GPU/Transforms/*.h"]),
includes = ["include"],
deps = [
":GPUDialect",
":GPUPassIncGen",
":Pass",
":Support",
"//llvm:Support",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
[mlir][transform] Create GPU transform dialect This revision adds GPU transform dialect. It also introduce a prefix such as "transform.gpu" for all ops related to this dialect. MLIR already had two GPU transform op in linalg. This revision moves these ops into GPUTransformOps. The Ops are as follows: `transform.structured.map_nested_foreach_thread_to_gpu_blocks` -> `transform.gpu.map_foreach_to_blocks` This op selects the outermost (toplevel) foreach_thread and parallelize across GPU blocks. It can also generate `gpu_launch`. `transform.structured.map_nested_foreach_thread_to_gpu_threads` -> `transform.gpu.map_nested_foreach_to_threads` This op parallelizes nested foreach_thread that are inside `gpu_launch` across GPU threads. It doesn't add new functionality, but there are some minor refactoring of the code. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D134800
2022-10-03 09:56:42 +02:00
td_library(
name = "GPUTransformOpsTdFiles",
srcs = [
"include/mlir/Dialect/GPU/TransformOps/GPUTransformOps.td",
],
includes = ["include"],
deps = [
":TransformDialectTdFiles",
],
)
gentbl_cc_library(
name = "GPUTransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/GPU/TransformOps/GPUTransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/GPU/TransformOps/GPUTransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/GPU/TransformOps/GPUTransformOps.td",
deps = [
":GPUTransformOpsTdFiles",
],
)
cc_library(
name = "GPUTransformOps",
[mlir][linalg] NFC - Move some utils in preparation for revamping mapping of scf.forall
2023-07-25 00:50:42 +02:00
srcs = glob([
"lib/Dialect/GPU/TransformOps/*.cpp",
]),
hdrs = glob([
"include/mlir/Dialect/GPU/TransformOps/*.h",
]),
[mlir][transform] Create GPU transform dialect This revision adds GPU transform dialect. It also introduce a prefix such as "transform.gpu" for all ops related to this dialect. MLIR already had two GPU transform op in linalg. This revision moves these ops into GPUTransformOps. The Ops are as follows: `transform.structured.map_nested_foreach_thread_to_gpu_blocks` -> `transform.gpu.map_foreach_to_blocks` This op selects the outermost (toplevel) foreach_thread and parallelize across GPU blocks. It can also generate `gpu_launch`. `transform.structured.map_nested_foreach_thread_to_gpu_threads` -> `transform.gpu.map_nested_foreach_to_threads` This op parallelizes nested foreach_thread that are inside `gpu_launch` across GPU threads. It doesn't add new functionality, but there are some minor refactoring of the code. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D134800
2022-10-03 09:56:42 +02:00
includes = ["include"],
deps = [
[mlir][Bazel] Add missing dependencies.
2023-03-20 09:28:25 +01:00
":AffineDialect",
[mlir][transform] Create GPU transform dialect This revision adds GPU transform dialect. It also introduce a prefix such as "transform.gpu" for all ops related to this dialect. MLIR already had two GPU transform op in linalg. This revision moves these ops into GPUTransformOps. The Ops are as follows: `transform.structured.map_nested_foreach_thread_to_gpu_blocks` -> `transform.gpu.map_foreach_to_blocks` This op selects the outermost (toplevel) foreach_thread and parallelize across GPU blocks. It can also generate `gpu_launch`. `transform.structured.map_nested_foreach_thread_to_gpu_threads` -> `transform.gpu.map_nested_foreach_to_threads` This op parallelizes nested foreach_thread that are inside `gpu_launch` across GPU threads. It doesn't add new functionality, but there are some minor refactoring of the code. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D134800
2022-10-03 09:56:42 +02:00
":ArithDialect",
":AsmParser",
":ControlFlowDialect",
[mlir][Bazel] Add missing dependencies.
2023-03-20 09:28:25 +01:00
":DialectUtils",
":FuncDialect",
[mlir][transform] Create GPU transform dialect This revision adds GPU transform dialect. It also introduce a prefix such as "transform.gpu" for all ops related to this dialect. MLIR already had two GPU transform op in linalg. This revision moves these ops into GPUTransformOps. The Ops are as follows: `transform.structured.map_nested_foreach_thread_to_gpu_blocks` -> `transform.gpu.map_foreach_to_blocks` This op selects the outermost (toplevel) foreach_thread and parallelize across GPU blocks. It can also generate `gpu_launch`. `transform.structured.map_nested_foreach_thread_to_gpu_threads` -> `transform.gpu.map_nested_foreach_to_threads` This op parallelizes nested foreach_thread that are inside `gpu_launch` across GPU threads. It doesn't add new functionality, but there are some minor refactoring of the code. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D134800
2022-10-03 09:56:42 +02:00
":GPUDialect",
":GPUTransformOpsIncGen",
":GPUTransforms",
":IR",
[mlir] add a simple gpu barrier elimination mechanism GPU code generation, and specifically the shared memory copy insertion may introduce spurious barriers guarding read-after-read dependencies or read-after-write on non-aliasing data, which degrades performance due to unnecessary synchronization. Add a pattern and transform op that removes such barriers by analyzing memory effects that the barrier actually guards that are not also guarded by other barriers. The code is adapted from the Polygeist incubator project. Co-authored-by: William Moses <gh@wsmoses.com> Co-authored-by: Ivan Radanov Ivanov <ivanov.i.aa@m.titech.ac.jp> Reviewed By: nicolasvasilache, wsmoses Differential Revision: https://reviews.llvm.org/D154720
2023-07-07 14:35:04 +00:00
":MemRefDialect",
[mlir][transform] Create GPU transform dialect This revision adds GPU transform dialect. It also introduce a prefix such as "transform.gpu" for all ops related to this dialect. MLIR already had two GPU transform op in linalg. This revision moves these ops into GPUTransformOps. The Ops are as follows: `transform.structured.map_nested_foreach_thread_to_gpu_blocks` -> `transform.gpu.map_foreach_to_blocks` This op selects the outermost (toplevel) foreach_thread and parallelize across GPU blocks. It can also generate `gpu_launch`. `transform.structured.map_nested_foreach_thread_to_gpu_threads` -> `transform.gpu.map_nested_foreach_to_threads` This op parallelizes nested foreach_thread that are inside `gpu_launch` across GPU threads. It doesn't add new functionality, but there are some minor refactoring of the code. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D134800
2022-10-03 09:56:42 +02:00
":Parser",
":SCFDialect",
":SideEffectInterfaces",
[mlir][Bazel] Adjust build after aafb52d7c9226cd9925bd5135309bd02b6e3b59d
2023-03-14 13:54:01 +01:00
":Support",
[mlir][transform] Create GPU transform dialect This revision adds GPU transform dialect. It also introduce a prefix such as "transform.gpu" for all ops related to this dialect. MLIR already had two GPU transform op in linalg. This revision moves these ops into GPUTransformOps. The Ops are as follows: `transform.structured.map_nested_foreach_thread_to_gpu_blocks` -> `transform.gpu.map_foreach_to_blocks` This op selects the outermost (toplevel) foreach_thread and parallelize across GPU blocks. It can also generate `gpu_launch`. `transform.structured.map_nested_foreach_thread_to_gpu_threads` -> `transform.gpu.map_nested_foreach_to_threads` This op parallelizes nested foreach_thread that are inside `gpu_launch` across GPU threads. It doesn't add new functionality, but there are some minor refactoring of the code. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D134800
2022-10-03 09:56:42 +02:00
":TransformDialect",
":TransformUtils",
[mlir] add a simple gpu barrier elimination mechanism GPU code generation, and specifically the shared memory copy insertion may introduce spurious barriers guarding read-after-read dependencies or read-after-write on non-aliasing data, which degrades performance due to unnecessary synchronization. Add a pattern and transform op that removes such barriers by analyzing memory effects that the barrier actually guards that are not also guarded by other barriers. The code is adapted from the Polygeist incubator project. Co-authored-by: William Moses <gh@wsmoses.com> Co-authored-by: Ivan Radanov Ivanov <ivanov.i.aa@m.titech.ac.jp> Reviewed By: nicolasvasilache, wsmoses Differential Revision: https://reviews.llvm.org/D154720
2023-07-07 14:35:04 +00:00
":VectorDialect",
[bazel][mlir] Port ff8775f3fffea787922eb238c0af3ed5715485d9
2023-07-25 19:55:58 +02:00
":VectorTransforms",
[mlir][transform] Create GPU transform dialect This revision adds GPU transform dialect. It also introduce a prefix such as "transform.gpu" for all ops related to this dialect. MLIR already had two GPU transform op in linalg. This revision moves these ops into GPUTransformOps. The Ops are as follows: `transform.structured.map_nested_foreach_thread_to_gpu_blocks` -> `transform.gpu.map_foreach_to_blocks` This op selects the outermost (toplevel) foreach_thread and parallelize across GPU blocks. It can also generate `gpu_launch`. `transform.structured.map_nested_foreach_thread_to_gpu_threads` -> `transform.gpu.map_nested_foreach_to_threads` This op parallelizes nested foreach_thread that are inside `gpu_launch` across GPU threads. It doesn't add new functionality, but there are some minor refactoring of the code. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D134800
2022-10-03 09:56:42 +02:00
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "LLVMOpsTdFiles",
srcs = [
[bazel] Port dd38f899803465dd2765d1601b3989df3bd53863 and fix 7732c97f52e72a0737aaafd19d1f3be9f26d1a20.
2022-10-17 16:54:37 -07:00
"include/mlir/Dialect/LLVMIR/LLVMAttrDefs.td",
Fix bazel build after e5f50bd26807
2023-04-14 11:11:39 +02:00
"include/mlir/Dialect/LLVMIR/LLVMDialect.td",
Add missing dependency to fix build.bazel. Reviewed By: Peiming, aartbik Differential Revision: https://reviews.llvm.org/D136516
2022-10-21 16:58:30 -07:00
"include/mlir/Dialect/LLVMIR/LLVMEnums.td",
[mlir][Bazel] Adjustments for fa51c1753a274fbb7a71d8fe91fd4e5caf2fa4d3 Fix bazel build. Differential Revision: https://reviews.llvm.org/D145903
2023-03-13 08:54:29 +01:00
"include/mlir/Dialect/LLVMIR/LLVMInterfaces.td",
[mlir] Split intrinsics out of LLVMOps.td Move the operations that correspond to LLVM IR intrinsics in a separate .td file. This makes it easier to maintain the intrinsics and decreases the compile time of LLVMDialect.cpp by ~25%. Depends On D123310 Reviewed By: wsmoses, jacquesguan Differential Revision: https://reviews.llvm.org/D123315
2022-04-13 18:29:47 +02:00
"include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/Dialect/LLVMIR/LLVMOpBase.td",
"include/mlir/Dialect/LLVMIR/LLVMOps.td",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
"include/mlir/Dialect/LLVMIR/LLVMTypes.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
[mlir][Bazel] Fix Bazel build after a2158374ba1a6f81f4cce3eb54d0bc44f3ab75e0
2022-07-06 08:47:21 +02:00
":CallInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ControlFlowInterfacesTdFiles",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
":DataLayoutInterfacesTdFiles",
[bazel] Port 7ceffae18c43
2022-01-19 14:14:36 +01:00
":FunctionInterfacesTdFiles",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterfaceTdFiles",
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
":MemorySlotInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
cc_library(
name = "GPUCommonTransforms",
srcs = [
"lib/Conversion/GPUCommon/GPUOpsLowering.cpp",
],
hdrs = [
[bazel] Tweak dependency spaghetti after 70c2e0618a0f3c09ed7149d88b4987b932eb6705
2023-07-21 13:51:32 +02:00
"include/mlir/Conversion/GPUCommon/GPUCommonPass.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"lib/Conversion/GPUCommon/GPUOpsLowering.h",
"lib/Conversion/GPUCommon/IndexIntrinsicsOpLowering.h",
"lib/Conversion/GPUCommon/OpToFuncCallLowering.h",
],
deps = [
[bazel] Tweak dependency spaghetti after 70c2e0618a0f3c09ed7149d88b4987b932eb6705
2023-07-21 13:51:32 +02:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
[bazel] Tweak dependency spaghetti after 70c2e0618a0f3c09ed7149d88b4987b932eb6705
2023-07-21 13:51:32 +02:00
":GPUTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
[bazel] Tweak dependency spaghetti after 70c2e0618a0f3c09ed7149d88b4987b932eb6705
2023-07-21 13:51:32 +02:00
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
gentbl_cc_library(
name = "GPUToNVVMGen",
strip_include_prefix = "lib/Conversion/GPUToNVVM",
tbl_outs = [
(
["-gen-rewriters"],
"lib/Conversion/GPUToNVVM/GPUToNVVM.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "lib/Conversion/GPUToNVVM/GPUToNVVM.td",
deps = [
":GPUOpsTdFiles",
":NVVMOpsTdFiles",
],
)
cc_library(
name = "GPUToNVVMTransforms",
srcs = glob([
"lib/Conversion/GPUToNVVM/*.cpp",
"lib/Conversion/GPUToNVVM/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/GPUToNVVM/*.h",
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithToLLVM",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":ControlFlowToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUCommonTransforms",
":GPUDialect",
":GPUToNVVMGen",
":GPUTransforms",
":IR",
[Bazel] Fixes for b5d847b1b95750d0af40cfc8c71a8fec50bb8613 and 6412a13539ab2914eed8e1df83c399b9a16e3408
2021-07-07 16:50:23 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
":MathDialect",
":MemRefDialect",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MemRefToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":NVVMDialect",
":Pass",
":Transforms",
"//llvm:Support",
],
)
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
cc_library(
name = "AMDGPUToROCDL",
srcs = glob([
"lib/Conversion/AMDGPUToROCDL/*.cpp",
"lib/Conversion/AMDGPUToROCDL/*.h",
[bazel] Fix mlir buildifier issues
2023-03-28 13:11:47 -07:00
]) + ["include/mlir/Conversion/GPUToROCDL/Runtimes.h"],
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
hdrs = glob([
"include/mlir/Conversion/AMDGPUToROCDL/*.h",
]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AMDGPUDialect",
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
":AMDGPUUtils",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
":ConversionPassIncGen",
":IR",
":LLVMCommonConversion",
bazel fixes for c55b41d5199d2394dd6cdb8f52180d8b81d809d4
2022-09-01 08:19:50 +02:00
":LLVMDialect",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
":Pass",
[mlir][linalg] Add lowering of named ops on complex numbers This lets linalg.dot and friends lower to a complex muladd using ops from the complex dialect. Differential Revision: https://reviews.llvm.org/D125461
2022-05-12 13:35:27 +02:00
":ROCDLDialect",
[mlir][bzl] Update for 1a92dbcfa88a857bf735c897125d9a2f9be53e9e and cab44c515c63db876a7f626fdc36be4a9157ebad
2022-07-07 17:36:28 -07:00
":Support",
[MLIR][AMDGPU] Add AMDGPU dialect, wrappers around raw buffer intrinsics By analogy with the NVGPU dialect, introduce an AMDGPU dialect for AMD-specific intrinsic wrappers. The dialect initially includes wrappers around the raw buffer intrinsics. On AMD GPUs, a memref can be converted to a "buffer descriptor" that allows more precise control of memory access, such as by allowing for out of bounds loads/stores to be replaced by 0/ignored without adding additional conditional logic, which is important for performance. The repository currently contains a limited conversion from transfer_read/transfer_write to Mubuf intrinsics, which are an older, deprecated intrinsic for the same functionality. The new amdgpu.raw_buffer_* ops allow these operations to be used explicitly and for including metadata such as whether the target chipset is an RDNA chip or not (which impacts the interpretation of some bits in the buffer descriptor), while still maintaining an MLIR-like interface. (This change also exposes the floating-point atomic add intrinsic.) Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D122765
2022-03-30 21:56:19 +00:00
":Transforms",
"//llvm:Support",
],
)
[mlir] Update bazel file after adding nvgpu to nvvm conversion
2022-04-15 04:07:46 +00:00
cc_library(
name = "NVGPUToNVVM",
srcs = glob([
"lib/Conversion/NVGPUToNVVM/*.cpp",
"lib/Conversion/NVGPUToNVVM/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
[mlir] Update bazel file after adding nvgpu to nvvm conversion
2022-04-15 04:07:46 +00:00
hdrs = glob([
"include/mlir/Conversion/NVGPUToNVVM/*.h",
]),
includes = ["include"],
deps = [
":ConversionPassIncGen",
[bazel] Tweak dependency spaghetti after 70c2e0618a0f3c09ed7149d88b4987b932eb6705
2023-07-21 13:51:32 +02:00
":GPUCommonTransforms",
[mlir][gpu] Move async copy ops to NVGPU and add caching hints Move async copy operations to NVGPU as they only exist on NV target and are designed to match ptx semantic. This allows us to also add more fine grain caching hint attribute to the op. Add hint to bypass L1 and hook it up to NVVM op. Differential Revision: https://reviews.llvm.org/D125244
2022-05-09 17:18:21 +00:00
":GPUDialect",
[mlir] Fix build after D155680
2023-07-21 13:33:54 +02:00
":GPUToGPURuntimeTransforms",
[mlir] Update bazel file after adding nvgpu to nvvm conversion
2022-04-15 04:07:46 +00:00
":IR",
":LLVMCommonConversion",
[bazel][mlir] NVGPUToNVVM fixes for 708185f03ff
2022-11-07 14:21:06 -05:00
":LLVMDialect",
Fix bazel build for 5a1cdcbd8698cd263696b38e2672fccac9ec793c
2023-07-11 14:26:32 -07:00
":MemRefDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":NVGPUDialect",
[mlir] Update bazel file after adding nvgpu to nvvm conversion
2022-04-15 04:07:46 +00:00
":NVVMDialect",
":Pass",
[bazel] Tweak dependency spaghetti after 70c2e0618a0f3c09ed7149d88b4987b932eb6705
2023-07-21 13:51:32 +02:00
"//llvm:Support",
[mlir] Update bazel file after adding nvgpu to nvvm conversion
2022-04-15 04:07:46 +00:00
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "VectorToSPIRV",
srcs = glob([
"lib/Conversion/VectorToSPIRV/*.cpp",
"lib/Conversion/VectorToSPIRV/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/VectorToSPIRV/*.h",
]),
includes = ["include"],
deps = [
[mlir][spirv][vector] Add pattern to convert reduction to SPIR-V dot prod This converts a specific form of `vector.reduction` to SPIR-V integer dot product ops. Add a new test pass to excercise this outside of the main vector to spirv conversion pass. Reviewed By: antiagainst Differential Revision: https://reviews.llvm.org/D145760
2023-03-10 13:54:16 -05:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir][spirv] Fix missing dependencies in BUILD.bazel
2022-02-04 16:51:05 -05:00
":IR",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":SPIRVConversion",
":SPIRVDialect",
[mlir][Bazel] Adjustments for fa51c1753a274fbb7a71d8fe91fd4e5caf2fa4d3 Fix bazel build. Differential Revision: https://reviews.llvm.org/D145903
2023-03-13 08:54:29 +01:00
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir][spirv] Fix missing dependencies in BUILD.bazel
2022-02-04 16:51:05 -05:00
"//llvm:Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
gentbl_cc_library(
name = "GPUToROCDLTGen",
strip_include_prefix = "lib/Conversion/GPUToROCDL",
tbl_outs = [
(
["-gen-rewriters"],
"lib/Conversion/GPUToROCDL/GPUToROCDL.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "lib/Conversion/GPUToROCDL/GPUToROCDL.td",
deps = [
":GPUOpsTdFiles",
":ROCDLOpsTdFiles",
],
)
cc_library(
name = "GPUToROCDLTransforms",
srcs = [
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
"include/mlir/Conversion/GPUToROCDL/Runtimes.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"lib/Conversion/GPUToROCDL/LowerGpuOpsToROCDLOps.cpp",
],
hdrs = ["include/mlir/Conversion/GPUToROCDL/GPUToROCDLPass.h"],
includes = ["include"],
deps = [
[mlir][AMDGPU] Add AMDGPU conversion patterns to ConvertGPUToROCDL This ensures that attributes such as the index bitwidth propagate correctly to the AMDGPUToROCDL patterns. Differential Revision: https://reviews.llvm.org/D125320
2022-05-10 15:37:53 +00:00
":AMDGPUToROCDL",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToLLVM",
[bazel] Add missing dependencies for 4f1e244eb5
2023-01-22 10:58:47 +01:00
":ControlFlowDialect",
[mlir][bazel] Update post 24a1
2022-02-08 08:48:10 -08:00
":ControlFlowToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUCommonTransforms",
":GPUDialect",
":GPUToROCDLTGen",
":GPUTransforms",
[mlir] Update mlir/BUILD.bazel.
2022-04-05 10:14:28 +02:00
":IR",
[Bazel] Fixes for b5d847b1b95750d0af40cfc8c71a8fec50bb8613 and 6412a13539ab2914eed8e1df83c399b9a16e3408
2021-07-07 16:50:23 -07:00
":LLVMCommonConversion",
Fix the broken bazel build for f6076bd81fce06bb29f83d370b8298110ecc2f99
2023-01-03 00:02:15 +01:00
":LLVMDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MathDialect",
[bazel] Add missing dependencies for 4f1e244eb5
2023-01-22 10:58:47 +01:00
":MemRefDialect",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MemRefToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":ROCDLDialect",
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":VectorToLLVM",
":VectorToSCF",
"//llvm:Support",
],
)
cc_library(
name = "GPUToVulkanTransforms",
srcs = [
"lib/Conversion/GPUToVulkan/ConvertGPULaunchFuncToVulkanLaunchFunc.cpp",
"lib/Conversion/GPUToVulkan/ConvertLaunchFuncToVulkanCalls.cpp",
],
hdrs = ["include/mlir/Conversion/GPUToVulkan/ConvertGPUToVulkanPass.h"],
includes = ["include"],
deps = [
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":IR",
":LLVMDialect",
":Pass",
":SPIRVDialect",
":SPIRVSerialization",
":Support",
"//llvm:Support",
],
)
cc_library(
name = "GPUToGPURuntimeTransforms",
srcs = [
"lib/Conversion/GPUCommon/GPUToLLVMConversion.cpp",
],
hdrs = ["include/mlir/Conversion/GPUCommon/GPUCommonPass.h"],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToLLVM",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AsyncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AsyncToLLVM",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":ControlFlowToLLVM",
[mlir][Vector] Add InferTypeOpInterface to vector dialect and remove the now redundant builders
2022-02-18 01:15:14 +01:00
":ConversionPassIncGen",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":GPUTransforms",
":IR",
Update Bazel overlay in GPUToGPURuntimeTransforms.
2021-07-08 21:36:44 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
[bazel] Add missing dependency after 16f8d17f7b
2023-01-20 20:40:06 +01:00
":MemRefDialect",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MemRefToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":VectorToLLVM",
"//llvm:Support",
],
)
cc_library(
name = "GPUToSPIRV",
srcs = glob([
"lib/Conversion/GPUToSPIRV/*.cpp",
"lib/Conversion/GPUToSPIRV/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/GPUToSPIRV/*.h",
]),
includes = [
"include",
"lib/Conversions/GPUToSPIRV",
],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":IR",
[mlir] NFC: split MemRef to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107094
2021-07-29 16:22:46 -04:00
":MemRefToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":SCFDialect",
":SCFToSPIRV",
":SPIRVConversion",
":SPIRVDialect",
":Support",
":Transforms",
":VectorToSPIRV",
"//llvm:Support",
],
)
cc_library(
name = "PDLToPDLInterp",
srcs = glob([
"lib/Conversion/PDLToPDLInterp/*.cpp",
"lib/Conversion/PDLToPDLInterp/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
[NFC][bazel] Enable layering_check for mlir/unittests
2023-01-19 03:40:59 -08:00
hdrs = [
"include/mlir/Conversion/PDLToPDLInterp/PDLToPDLInterp.h",
"lib/Conversion/PDLToPDLInterp/RootOrdering.h",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":ConversionPassIncGen",
":IR",
":InferTypeOpInterface",
":PDLDialect",
":PDLInterpDialect",
":Pass",
":Support",
"//llvm:Support",
],
)
cc_library(
name = "SPIRVToLLVM",
srcs = glob([
"lib/Conversion/SPIRVToLLVM/*.cpp",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/SPIRVToLLVM/*.h",
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":IR",
[Bazel] Fixes for b5d847b1b95750d0af40cfc8c71a8fec50bb8613 and 6412a13539ab2914eed8e1df83c399b9a16e3408
2021-07-07 16:50:23 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MemRefToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":SPIRVDialect",
":SPIRVUtils",
":Support",
":Transforms",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "LLVMOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/LLVMIR/LLVMOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/LLVMIR/LLVMOps.cpp.inc",
),
(
["-gen-dialect-decls"],
"include/mlir/Dialect/LLVMIR/LLVMOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
"include/mlir/Dialect/LLVMIR/LLVMOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-enum-decls"],
"include/mlir/Dialect/LLVMIR/LLVMOpsEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/LLVMIR/LLVMOpsEnums.cpp.inc",
),
[bazel] Port dd38f899803465dd2765d1601b3989df3bd53863 and fix 7732c97f52e72a0737aaafd19d1f3be9f26d1a20.
2022-10-17 16:54:37 -07:00
(
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
[
"--gen-attrdef-decls",
"-attrdefs-dialect=llvm",
],
[bazel] Port dd38f899803465dd2765d1601b3989df3bd53863 and fix 7732c97f52e72a0737aaafd19d1f3be9f26d1a20.
2022-10-17 16:54:37 -07:00
"include/mlir/Dialect/LLVMIR/LLVMOpsAttrDefs.h.inc",
),
(
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
[
"--gen-attrdef-defs",
"-attrdefs-dialect=llvm",
],
[bazel] Port dd38f899803465dd2765d1601b3989df3bd53863 and fix 7732c97f52e72a0737aaafd19d1f3be9f26d1a20.
2022-10-17 16:54:37 -07:00
"include/mlir/Dialect/LLVMIR/LLVMOpsAttrDefs.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/LLVMOps.td",
deps = [":LLVMOpsTdFiles"],
)
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
gentbl_cc_library(
name = "LLVMTypesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-typedef-decls",
"-typedefs-dialect=llvm",
],
"include/mlir/Dialect/LLVMIR/LLVMTypes.h.inc",
),
(
[
"-gen-typedef-defs",
"-typedefs-dialect=llvm",
],
"include/mlir/Dialect/LLVMIR/LLVMTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/LLVMTypes.td",
deps = [":LLVMOpsTdFiles"],
)
[mlir] Split intrinsics out of LLVMOps.td Move the operations that correspond to LLVM IR intrinsics in a separate .td file. This makes it easier to maintain the intrinsics and decreases the compile time of LLVMDialect.cpp by ~25%. Depends On D123310 Reviewed By: wsmoses, jacquesguan Differential Revision: https://reviews.llvm.org/D123315
2022-04-13 18:29:47 +02:00
gentbl_cc_library(
name = "LLVMIntrinsicOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td",
deps = [":LLVMOpsTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "LLVMConversionIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-llvmir-conversions"],
"include/mlir/Dialect/LLVMIR/LLVMConversions.inc",
),
(
["-gen-enum-to-llvmir-conversions"],
"include/mlir/Dialect/LLVMIR/LLVMConversionEnumsToLLVM.inc",
),
(
["-gen-enum-from-llvmir-conversions"],
"include/mlir/Dialect/LLVMIR/LLVMConversionEnumsFromLLVM.inc",
),
[mlir][Bazel] Port 8446f24ef03928f511ab59dae17a8aa581b0ab62
2022-10-12 14:07:31 +02:00
(
["-gen-op-from-llvmir-conversions"],
"include/mlir/Dialect/LLVMIR/LLVMOpFromLLVMIRConversions.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/LLVMOps.td",
deps = [":LLVMOpsTdFiles"],
)
[mlir] Split intrinsics out of LLVMOps.td Move the operations that correspond to LLVM IR intrinsics in a separate .td file. This makes it easier to maintain the intrinsics and decreases the compile time of LLVMDialect.cpp by ~25%. Depends On D123310 Reviewed By: wsmoses, jacquesguan Differential Revision: https://reviews.llvm.org/D123315
2022-04-13 18:29:47 +02:00
gentbl_cc_library(
name = "LLVMIntrinsicConversionIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-llvmir-conversions"],
"include/mlir/Dialect/LLVMIR/LLVMIntrinsicConversions.inc",
),
[mlir][Bazel] Adjust BUILD.bazel file
2022-05-31 13:35:49 +02:00
(
[mlir][Bazel] Port d85f6e5d57f38f0cbbc756484e3a93ae89b11195
2022-10-07 13:49:28 +02:00
["-gen-intr-from-llvmir-conversions"],
"include/mlir/Dialect/LLVMIR/LLVMIntrinsicFromLLVMIRConversions.inc",
),
(
["-gen-convertible-llvmir-intrinsics"],
"include/mlir/Dialect/LLVMIR/LLVMConvertibleLLVMIRIntrinsics.inc",
[mlir][Bazel] Adjust BUILD.bazel file
2022-05-31 13:35:49 +02:00
),
[mlir] Split intrinsics out of LLVMOps.td Move the operations that correspond to LLVM IR intrinsics in a separate .td file. This makes it easier to maintain the intrinsics and decreases the compile time of LLVMDialect.cpp by ~25%. Depends On D123310 Reviewed By: wsmoses, jacquesguan Differential Revision: https://reviews.llvm.org/D123315
2022-04-13 18:29:47 +02:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td",
deps = [":LLVMOpsTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "NVVMDialect",
srcs = ["lib/Dialect/LLVMIR/IR/NVVMDialect.cpp"],
hdrs = ["include/mlir/Dialect/LLVMIR/NVVMDialect.h"],
includes = ["include"],
deps = [
[mlir] Fix the bazel build for 3ac17449cf988bfcde804a4cc532420ed1657595
2022-10-28 14:47:01 +02:00
":DialectUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":LLVMDialect",
":NVVMOpsIncGen",
":SideEffectInterfaces",
[mlir][test] Fix linking error post test-lower-to-nvvm This fixes builds for 7e78ecfe10ea9071234de8d385b87d338d280266 (both cmake and bazel) as well as trim unnecessary dependencies. This is achieved by moving the functionality to test/lib/GPU which is a more natural landing pad.
2023-07-17 18:08:00 +02:00
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:AsmParser",
"//llvm:Core",
"//llvm:Support",
],
)
td_library(
name = "NVVMOpsTdFiles",
srcs = ["include/mlir/Dialect/LLVMIR/NVVMOps.td"],
includes = ["include"],
deps = [
":LLVMOpsTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "NVVMOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/LLVMIR/NVVMOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/LLVMIR/NVVMOps.cpp.inc",
),
(
[
"-gen-dialect-decls",
"-dialect=nvvm",
],
"include/mlir/Dialect/LLVMIR/NVVMOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=nvvm",
],
"include/mlir/Dialect/LLVMIR/NVVMOpsDialect.cpp.inc",
),
[mlir][nvvm] Generalize wmma ops to handle more types and shapes wmma intrinsics have a large number of combinations, ideally we want to be able to target all the different variants. To avoid a combinatorial explosion in the number of mlir op we use attributes to represent the different variation of load/store/mma ops. We also can generate with tablegen helpers to know which combinations are available. Using this we can avoid having too hardcode a path for specific shapes and can support more types. This patch also adds boiler plates for tf32 op support. Differential Revision: https://reviews.llvm.org/D112689
2021-10-27 22:43:10 -07:00
(
["-gen-enum-decls"],
"include/mlir/Dialect/LLVMIR/NVVMOpsEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/LLVMIR/NVVMOpsEnums.cpp.inc",
),
[mlir] Replace StrEnumAttr -> EnumAttr in core dialects Removes uses of `StrEnumAttr` in core dialects Reviewed By: mehdi_amini, rriddle Differential Revision: https://reviews.llvm.org/D117514
2022-01-18 16:53:55 +00:00
(
[
"-gen-attrdef-decls",
"-attrdefs-dialect=nvvm",
],
"include/mlir/Dialect/LLVMIR/NVVMOpsAttributes.h.inc",
),
(
[
"-gen-attrdef-defs",
"-attrdefs-dialect=nvvm",
],
"include/mlir/Dialect/LLVMIR/NVVMOpsAttributes.cpp.inc",
),
Fix bazel build file for D154060. Differential Revision: https://reviews.llvm.org/D154976
2023-07-11 17:03:48 +02:00
(
[
"-gen-op-interface-decls",
"-attrdefs-dialect=nvvm",
],
"include/mlir/Dialect/LLVMIR/NVVMOpsInterface.h.inc",
),
(
[
"-gen-op-interface-defs",
"-attrdefs-dialect=nvvm",
],
"include/mlir/Dialect/LLVMIR/NVVMOpsInterface.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/NVVMOps.td",
deps = [":NVVMOpsTdFiles"],
)
gentbl_cc_library(
name = "NVVMConversionIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-llvmir-conversions"],
"include/mlir/Dialect/LLVMIR/NVVMConversions.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/NVVMOps.td",
deps = [":NVVMOpsTdFiles"],
)
Fix bazel build file for D154060. Differential Revision: https://reviews.llvm.org/D154976
2023-07-11 17:03:48 +02:00
cc_library(
name = "NVVMToLLVM",
srcs = glob(["lib/Conversion/NVVMToLLVM/NVVMToLLVM.cpp"]),
hdrs = glob(["include/mlir/Conversion/NVVMToLLVM/*.h"]),
includes = ["include"],
deps = [
":ConversionPassIncGen",
":FuncDialect",
":GPUDialect",
":IR",
":LLVMCommonConversion",
":LLVMDialect",
":MemRefDialect",
":NVVMDialect",
":NVVMOpsIncGen",
":Pass",
":Support",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "ROCDLDialect",
srcs = ["lib/Dialect/LLVMIR/IR/ROCDLDialect.cpp"],
hdrs = ["include/mlir/Dialect/LLVMIR/ROCDLDialect.h"],
includes = ["include"],
deps = [
":IR",
":LLVMDialect",
":ROCDLOpsIncGen",
":SideEffectInterfaces",
"//llvm:AsmParser",
"//llvm:Core",
"//llvm:Support",
],
)
td_library(
name = "ROCDLOpsTdFiles",
srcs = ["include/mlir/Dialect/LLVMIR/ROCDLOps.td"],
includes = ["include"],
deps = [
":LLVMOpsTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "ROCDLOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/LLVMIR/ROCDLOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/LLVMIR/ROCDLOps.cpp.inc",
),
(
[
"-gen-dialect-decls",
"-dialect=rocdl",
],
"include/mlir/Dialect/LLVMIR/ROCDLOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=rocdl",
],
"include/mlir/Dialect/LLVMIR/ROCDLOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/ROCDLOps.td",
deps = [":ROCDLOpsTdFiles"],
)
gentbl_cc_library(
name = "ROCDLConversionIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-llvmir-conversions"],
"include/mlir/Dialect/LLVMIR/ROCDLConversions.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/LLVMIR/ROCDLOps.td",
deps = [":ROCDLOpsTdFiles"],
)
cc_library(
name = "PDLDialect",
srcs = glob([
"lib/Dialect/PDL/IR/*.cpp",
"lib/Dialect/PDL/IR/*.h",
]),
hdrs = glob([
"include/mlir/Dialect/PDL/IR/*.h",
]),
includes = ["include"],
deps = [
":IR",
":InferTypeOpInterface",
":PDLOpsIncGen",
":PDLTypesIncGen",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":SideEffectInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
td_library(
name = "PDLDialectTdFiles",
srcs = [
"include/mlir/Dialect/PDL/IR/PDLDialect.td",
"include/mlir/Dialect/PDL/IR/PDLOps.td",
"include/mlir/Dialect/PDL/IR/PDLTypes.td",
],
deps = [
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "PDLOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/PDL/IR/PDLOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/PDL/IR/PDLOps.cpp.inc",
),
(
["-gen-dialect-decls"],
"include/mlir/Dialect/PDL/IR/PDLOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
"include/mlir/Dialect/PDL/IR/PDLOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/PDL/IR/PDLOps.td",
deps = [":PDLDialectTdFiles"],
)
gentbl_cc_library(
name = "PDLTypesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-typedef-decls"],
"include/mlir/Dialect/PDL/IR/PDLOpsTypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/PDL/IR/PDLOpsTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/PDL/IR/PDLTypes.td",
deps = [":PDLDialectTdFiles"],
)
cc_library(
name = "PDLInterpDialect",
srcs = glob([
"lib/Dialect/PDLInterp/IR/*.cpp",
"lib/Dialect/PDLInterp/IR/*.h",
]),
hdrs = glob([
"include/mlir/Dialect/PDLInterp/IR/*.h",
]),
includes = ["include"],
deps = [
":IR",
":InferTypeOpInterface",
":PDLDialect",
":PDLInterpOpsIncGen",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":SideEffectInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
td_library(
name = "PDLInterpOpsTdFiles",
srcs = ["include/mlir/Dialect/PDLInterp/IR/PDLInterpOps.td"],
includes = ["include"],
deps = [
Updated mlir-tablegen dependencies for f96a8675cd29776ca39bb76133ee9167f5001f17 Differential Revision: https://reviews.llvm.org/D121764
2022-03-15 19:41:33 -07:00
":FunctionInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":PDLDialectTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "PDLInterpOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/PDLInterp/IR/PDLInterpOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/PDLInterp/IR/PDLInterpOps.cpp.inc",
),
(
[
"-gen-dialect-decls",
"-dialect=pdl_interp",
],
"include/mlir/Dialect/PDLInterp/IR/PDLInterpOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=pdl_interp",
],
"include/mlir/Dialect/PDLInterp/IR/PDLInterpOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/PDLInterp/IR/PDLInterpOps.td",
deps = [":PDLInterpOpsTdFiles"],
)
td_library(
name = "SPIRVOpsTdFiles",
srcs = glob(["include/mlir/Dialect/SPIRV/IR/*.td"]),
includes = ["include"],
deps = [
[mlir][bazel] Fix bazel build files This adds a missing dependency on builtin dialect td files. Differential Revision: https://reviews.llvm.org/D130890
2022-08-01 11:40:39 +02:00
":BuiltinDialectTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":CallInterfacesTdFiles",
":ControlFlowInterfacesTdFiles",
[bazel] Port 7ceffae18c43
2022-01-19 14:14:36 +01:00
":FunctionInterfacesTdFiles",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "SPIRVOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/SPIRV/IR/SPIRVOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/SPIRV/IR/SPIRVOps.cpp.inc",
),
(
["-gen-dialect-decls"],
"include/mlir/Dialect/SPIRV/IR/SPIRVOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
"include/mlir/Dialect/SPIRV/IR/SPIRVOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-doc"],
"g3doc/Dialects/SPIRV/SPIRVOps.md",
),
(
["-gen-enum-decls"],
"include/mlir/Dialect/SPIRV/IR/SPIRVEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/SPIRV/IR/SPIRVEnums.cpp.inc",
),
(
["-gen-spirv-enum-avail-decls"],
"include/mlir/Dialect/SPIRV/IR/SPIRVEnumAvailability.h.inc",
),
(
["-gen-spirv-enum-avail-defs"],
"include/mlir/Dialect/SPIRV/IR/SPIRVEnumAvailability.cpp.inc",
),
(
["-gen-spirv-capability-implication"],
"include/mlir/Dialect/SPIRV/IR/SPIRVCapabilityImplication.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SPIRV/IR/SPIRVOps.td",
deps = [":SPIRVOpsTdFiles"],
)
[mlir][spirv] Replace StructAttrs with AttrDefs Depends on D127370 Reviewed By: antiagainst Differential Revision: https://reviews.llvm.org/D127373
2022-06-09 21:35:32 +00:00
gentbl_cc_library(
name = "SPIRVAttributesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/SPIRV/IR/SPIRVAttributes.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/SPIRV/IR/SPIRVAttributes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SPIRV/IR/SPIRVAttributes.td",
deps = [":SPIRVOpsTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "SPIRVCanonicalizationIncGen",
strip_include_prefix = "lib/Dialect/SPIRV/IR",
tbl_outs = [
(
["-gen-rewriters"],
"lib/Dialect/SPIRV/IR/SPIRVCanonicalization.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "lib/Dialect/SPIRV/IR/SPIRVCanonicalization.td",
deps = [":SPIRVOpsTdFiles"],
)
gentbl_cc_library(
name = "SPIRVAvailabilityIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-avail-interface-decls"],
"include/mlir/Dialect/SPIRV/IR/SPIRVAvailability.h.inc",
),
(
["-gen-avail-interface-defs"],
"include/mlir/Dialect/SPIRV/IR/SPIRVAvailability.cpp.inc",
),
(
["-gen-spirv-avail-impls"],
"include/mlir/Dialect/SPIRV/IR/SPIRVOpAvailabilityImpl.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SPIRV/IR/SPIRVOps.td",
deps = [":SPIRVOpsTdFiles"],
)
gentbl_cc_library(
name = "SPIRVAttrUtilsGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-spirv-attr-utils"],
"include/mlir/Dialect/SPIRV/IR/SPIRVAttrUtils.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SPIRV/IR/SPIRVBase.td",
deps = [":SPIRVOpsTdFiles"],
)
gentbl_cc_library(
name = "SPIRVSerializationGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-spirv-serialization"],
"include/mlir/Dialect/SPIRV/IR/SPIRVSerialization.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SPIRV/IR/SPIRVOps.td",
deps = [":SPIRVOpsTdFiles"],
)
cc_library(
name = "SPIRVDialect",
srcs = glob([
"lib/Dialect/SPIRV/IR/*.cpp",
"lib/Dialect/SPIRV/IR/*.h",
]) + ["include/mlir/Transforms/InliningUtils.h"],
hdrs = glob([
"include/mlir/Dialect/SPIRV/IR/*.h",
]),
includes = ["include"],
deps = [
":CommonFolders",
":ControlFlowInterfaces",
":IR",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Parser",
":SPIRVAttrUtilsGen",
[mlir][spirv] Replace StructAttrs with AttrDefs Depends on D127370 Reviewed By: antiagainst Differential Revision: https://reviews.llvm.org/D127373
2022-06-09 21:35:32 +00:00
":SPIRVAttributesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SPIRVAvailabilityIncGen",
":SPIRVCanonicalizationIncGen",
":SPIRVOpsIncGen",
":SPIRVSerializationGen",
":SideEffectInterfaces",
":Support",
":Transforms",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "SPIRVPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=SPIRV",
],
"include/mlir/Dialect/SPIRV/Transforms/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/SPIRV/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "SPIRVUtils",
srcs = glob([
"lib/Dialect/SPIRV/Utils/*.cpp",
]),
hdrs = glob([
"include/mlir/Dialect/SPIRV/Utils/*.h",
]),
includes = ["include"],
deps = [
":SPIRVDialect",
"//llvm:Support",
],
)
cc_library(
name = "SPIRVConversion",
srcs = ["lib/Dialect/SPIRV/Transforms/SPIRVConversion.cpp"],
hdrs = ["include/mlir/Dialect/SPIRV/Transforms/SPIRVConversion.h"],
includes = ["include"],
deps = [
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SPIRVDialect",
":TransformUtils",
"//llvm:Support",
],
)
cc_library(
name = "SPIRVTransforms",
srcs = glob(
[
"lib/Dialect/SPIRV/Transforms/*.cpp",
"lib/Dialect/SPIRV/Transforms/*.h",
],
exclude = ["lib/Dialect/SPIRV/Transforms/SPIRVConversion.cpp"],
),
hdrs = glob(
["include/mlir/Dialect/SPIRV/Transforms/*.h"],
exclude = ["include/mlir/Dialect/SPIRV/Transforms/SPIRVConversion.h"],
),
includes = ["include"],
deps = [
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":Pass",
":SPIRVConversion",
":SPIRVDialect",
":SPIRVPassIncGen",
":SPIRVUtils",
[bazel] Add missing :Support dependency after 1b8224537070
2023-01-06 12:51:01 +01:00
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
"//llvm:Support",
],
)
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
cc_library(
name = "SPIRVCommonConversion",
hdrs = ["lib/Conversion/SPIRVCommon/Pattern.h"],
includes = ["include"],
deps = [
[bazel] add missing dependency to //mlir:SPIRVCommonConversion
2023-03-17 15:00:50 -07:00
":IR",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":SPIRVDialect",
":Transforms",
[Bazel] Add llvm:Support to Bazel deps of SPIRVCommonConversion Differential Revision: https://reviews.llvm.org/D137119
2022-10-31 15:26:58 -07:00
"//llvm:Support",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
],
)
[mlir] NFC: split Math to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107093
2021-07-29 16:21:43 -04:00
cc_library(
name = "MathToSPIRV",
srcs = glob([
"lib/Conversion/MathToSPIRV/*.cpp",
"lib/Conversion/MathToSPIRV/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
[mlir] NFC: split Math to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107093
2021-07-29 16:21:43 -04:00
hdrs = glob([
"include/mlir/Conversion/MathToSPIRV/*.h",
]),
includes = [
"include",
"lib/Conversion/MathToSPIRV",
],
deps = [
":ConversionPassIncGen",
":IR",
":MathDialect",
":Pass",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":SPIRVCommonConversion",
[mlir] NFC: split Math to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107093
2021-07-29 16:21:43 -04:00
":SPIRVConversion",
":SPIRVDialect",
":Support",
":Transforms",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
Attempt fix for Bazel build from 3ba66435d9377cb99ee1718807cbf6a7c4b0d6d3
2022-03-02 21:45:25 +00:00
name = "FuncToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = glob([
Attempt fix for Bazel build from 3ba66435d9377cb99ee1718807cbf6a7c4b0d6d3
2022-03-02 21:45:25 +00:00
"lib/Conversion/FuncToSPIRV/*.cpp",
"lib/Conversion/FuncToSPIRV/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
Attempt fix for Bazel build from 3ba66435d9377cb99ee1718807cbf6a7c4b0d6d3
2022-03-02 21:45:25 +00:00
"include/mlir/Conversion/FuncToSPIRV/*.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
]),
includes = [
"include",
Attempt fix for Bazel build from 3ba66435d9377cb99ee1718807cbf6a7c4b0d6d3
2022-03-02 21:45:25 +00:00
"lib/Conversion/FuncToSPIRV",
],
deps = [
":ControlFlowToSPIRV",
":ConversionPassIncGen",
":FuncDialect",
":IR",
":MathToSPIRV",
":Pass",
":SPIRVCommonConversion",
":SPIRVConversion",
":SPIRVDialect",
":SPIRVUtils",
":Support",
":TensorDialect",
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Attempt fix for Bazel build from 3ba66435d9377cb99ee1718807cbf6a7c4b0d6d3
2022-03-02 21:45:25 +00:00
"//llvm:Support",
],
)
[mlir] Add TensorToLinalgPass This pass is to handle computationally complex operations like tensor.pad which are not simply lowered to the exact same operation in the memref dialect. Differential Revision: https://reviews.llvm.org/D125384
2022-05-11 14:10:12 +02:00
cc_library(
name = "TensorToLinalg",
srcs = glob([
"lib/Conversion/TensorToLinalg/*.cpp",
"lib/Conversion/TensorToLinalg/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
[mlir] Add TensorToLinalgPass This pass is to handle computationally complex operations like tensor.pad which are not simply lowered to the exact same operation in the memref dialect. Differential Revision: https://reviews.llvm.org/D125384
2022-05-11 14:10:12 +02:00
hdrs = glob([
"include/mlir/Conversion/TensorToLinalg/*.h",
]),
includes = [
"include",
"lib/Conversion/TensorToLinalg",
],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir] Add TensorToLinalgPass This pass is to handle computationally complex operations like tensor.pad which are not simply lowered to the exact same operation in the memref dialect. Differential Revision: https://reviews.llvm.org/D125384
2022-05-11 14:10:12 +02:00
":ConversionPassIncGen",
":FuncDialect",
":IR",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
[mlir] Add TensorToLinalgPass This pass is to handle computationally complex operations like tensor.pad which are not simply lowered to the exact same operation in the memref dialect. Differential Revision: https://reviews.llvm.org/D125384
2022-05-11 14:10:12 +02:00
":LinalgTransforms",
":Pass",
":Support",
":TensorDialect",
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir] Add TensorToLinalgPass This pass is to handle computationally complex operations like tensor.pad which are not simply lowered to the exact same operation in the memref dialect. Differential Revision: https://reviews.llvm.org/D125384
2022-05-11 14:10:12 +02:00
"//llvm:Support",
],
)
Attempt fix for Bazel build from 3ba66435d9377cb99ee1718807cbf6a7c4b0d6d3
2022-03-02 21:45:25 +00:00
cc_library(
name = "TensorToSPIRV",
srcs = glob([
"lib/Conversion/TensorToSPIRV/*.cpp",
"lib/Conversion/TensorToSPIRV/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Attempt fix for Bazel build from 3ba66435d9377cb99ee1718807cbf6a7c4b0d6d3
2022-03-02 21:45:25 +00:00
hdrs = glob([
"include/mlir/Conversion/TensorToSPIRV/*.h",
]),
includes = [
"include",
"lib/Conversion/TensorToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToSPIRV",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":ControlFlowToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
More attempts at bazel fix
2022-03-03 00:18:54 +00:00
":FuncToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":MathToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":SPIRVCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SPIRVConversion",
":SPIRVDialect",
":SPIRVUtils",
":Support",
":TensorDialect",
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
cc_library(
name = "SPIRVBinaryUtils",
srcs = ["lib/Target/SPIRV/SPIRVBinaryUtils.cpp"],
hdrs = ["include/mlir/Target/SPIRV/SPIRVBinaryUtils.h"],
includes = ["include"],
deps = [
":SPIRVAttrUtilsGen",
":SPIRVDialect",
":SPIRVOpsIncGen",
":Support",
"//llvm:Support",
],
)
cc_library(
name = "SPIRVSerialization",
srcs = [
"lib/Target/SPIRV/Serialization/Serialization.cpp",
"lib/Target/SPIRV/Serialization/SerializeOps.cpp",
"lib/Target/SPIRV/Serialization/Serializer.cpp",
"lib/Target/SPIRV/Serialization/Serializer.h",
],
hdrs = ["include/mlir/Target/SPIRV/Serialization.h"],
includes = ["include"],
deps = [
":IR",
":SPIRVAttrUtilsGen",
":SPIRVBinaryUtils",
":SPIRVDialect",
":SPIRVOpsIncGen",
":SPIRVSerializationGen",
":Support",
":Transforms",
"//llvm:Support",
],
)
cc_library(
name = "SPIRVDeserialization",
srcs = glob([
"lib/Target/SPIRV/Deserialization/*.cpp",
"lib/Target/SPIRV/Deserialization/*.h",
]),
hdrs = ["include/mlir/Target/SPIRV/Deserialization.h"],
includes = ["include"],
deps = [
":IR",
":SPIRVAttrUtilsGen",
":SPIRVBinaryUtils",
":SPIRVDialect",
":SPIRVOpsIncGen",
":SPIRVSerializationGen",
":Support",
":Transforms",
"//llvm:Support",
],
)
cc_library(
name = "SPIRVModuleCombiner",
srcs = glob(
["lib/Dialect/SPIRV/Linking/ModuleCombiner/*.cpp"],
),
hdrs = ["include/mlir/Dialect/SPIRV/Linking/ModuleCombiner.h"],
includes = ["include"],
deps = [
":IR",
":SPIRVDialect",
":Support",
"//llvm:Support",
],
)
cc_library(
name = "SPIRVTranslateRegistration",
srcs = ["lib/Target/SPIRV/TranslateRegistration.cpp"],
includes = ["include"],
deps = [
":IR",
":Parser",
":SPIRVDeserialization",
":SPIRVDialect",
":SPIRVSerialization",
":Support",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":TranslateLib",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
td_library(
name = "TensorOpsTdFiles",
srcs = [
"include/mlir/Dialect/Tensor/IR/TensorBase.td",
"include/mlir/Dialect/Tensor/IR/TensorOps.td",
],
includes = ["include"],
deps = [
":CastInterfacesTdFiles",
":ControlFlowInterfacesTdFiles",
[mlir][tensor] Implement DestinationStyleOpInterface for tensor.insert/insert_slice Also allow unranked tensors/memrefs with destination style op outputs. This allows for a simpler implementation of the BufferizableOpInterface (in a subsequent commit). Differential Revision: https://reviews.llvm.org/D136346
2022-10-27 10:27:11 +02:00
":DestinationStyleOpInterfaceTdFiles",
[mlir][Tensor] Implement `reifyReturnTypeShapesPerResultDim` for `tensor.insert_slice`. Differential Revision: https://reviews.llvm.org/D105852
2021-07-13 14:51:20 -07:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
[mlir][Tensor] Move ParallelInsertSlice to the tensor dialect This is moslty NFC and will allow tensor.parallel_insert_slice to gain rank-reducing semantics by reusing the vast majority of the tensor.insert_slice impl. Depends on D128857 Differential Revision: https://reviews.llvm.org/D128920
2022-06-30 04:27:41 -07:00
":ParallelCombiningOpInterfaceTdFiles",
[mlir][interfaces] Add ShapedDimOpInterface This interface is implemented by memref.dim and tensor.dim. This change makes it possible to remove a build dependency of the Affine dialect on the Tensor dialect (and maybe also the MemRef dialect in the future). Differential Revision: https://reviews.llvm.org/D133595
2022-10-03 13:53:16 +09:00
":ShapedOpInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfacesTdFiles",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
":TilingInterfaceTdFiles",
Update Bazel BUILD files up to be9a87fe9b Differential Revision: https://reviews.llvm.org/D104791
2021-06-23 09:25:36 -07:00
":ViewLikeInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
gentbl_cc_library(
name = "TensorOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=tensor",
],
"include/mlir/Dialect/Tensor/IR/TensorOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=tensor",
],
"include/mlir/Dialect/Tensor/IR/TensorOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-decls"],
"include/mlir/Dialect/Tensor/IR/TensorOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Tensor/IR/TensorOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Tensor/IR/TensorOps.td",
deps = [":TensorOpsTdFiles"],
)
cc_library(
name = "TensorDialect",
[mlir] Move `linalg.tensor_expand/collapse_shape` to TensorDialect. RFC: https://llvm.discourse.group/t/rfc-reshape-ops-restructuring/3310 linalg.fill gets a canonicalizer, because `FoldFillWithTensorReshape` cannot be moved to tensorops (it uses linalg::FillOp inside). Before it was listed as a canonicalization pattern for the reshape operations, now it became a canonicalization for FillOp. Differential Revision: https://reviews.llvm.org/D115502
2021-12-10 12:03:47 +01:00
srcs = [
"include/mlir/Transforms/InliningUtils.h",
"lib/Dialect/Tensor/IR/TensorDialect.cpp",
"lib/Dialect/Tensor/IR/TensorOps.cpp",
[mlir][Interfaces] Add ValueBoundsOpInterface Ops can implement this interface to specify lower/upper bounds for their result values and block arguments. Bounds can be specified for: * Index-type values * Dimension sizes of shapes values The bounds are added to a constraint set. Users can query this constraint set to compute bounds wrt. to a user-specified set of values. Only EQ bounds are supported at the moment. This revision also contains interface implementations for various tensor dialect ops, which illustrates how to implement this interface. Differential Revision: https://reviews.llvm.org/D145681
2023-04-06 02:41:15 +02:00
"lib/Dialect/Tensor/IR/ValueBoundsOpInterfaceImpl.cpp",
],
hdrs = [
"include/mlir/Dialect/Tensor/IR/Tensor.h",
"include/mlir/Dialect/Tensor/IR/ValueBoundsOpInterfaceImpl.h",
[mlir] Move `linalg.tensor_expand/collapse_shape` to TensorDialect. RFC: https://llvm.discourse.group/t/rfc-reshape-ops-restructuring/3310 linalg.fill gets a canonicalizer, because `FoldFillWithTensorReshape` cannot be moved to tensorops (it uses linalg::FillOp inside). Before it was listed as a canonicalization pattern for the reshape operations, now it became a canonicalization for FillOp. Differential Revision: https://reviews.llvm.org/D115502
2021-12-10 12:03:47 +01:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir][tensor][NFC] Rename linalg.init_tensor to tensor.empty tensor.empty/linalg.init_tensor produces an uninititalized tensor that can be used as a destination operand for destination-style ops (ops that implement `DestinationStyleOpInterface`). This change makes it possible to implement `TilingInterface` for non-destination-style ops without depending on the Linalg dialect. RFC: https://discourse.llvm.org/t/rfc-add-tensor-from-shape-operation/65101 Differential Revision: https://reviews.llvm.org/D135129
2022-10-04 17:06:00 +09:00
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfaces",
[mlir][Bazel][cmake] Add missing dependency. After 2c7b0685e179 this dependency is needed.
2022-01-28 08:20:24 +01:00
":ComplexDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ControlFlowInterfaces",
[mlir][interfaces] Remove getDestinationOperands from TilingInterface `getDestinationOperands` was almost a duplicate of `DestinationStyleOpInterface::getOutputOperands`. Now that the interface has been moved to mlir/Interfaces, it is no longer needed. Differential Revision: https://reviews.llvm.org/D136240
2022-10-24 09:17:44 +02:00
":DestinationStyleOpInterface",
[Bazel] Update for 0813700de1 Updates Bazel BUILD files for changes in https://github.com/llvm/llvm-project/commit/0813700de1 Differential Revision: https://reviews.llvm.org/D104990
2021-06-27 08:30:54 -07:00
":DialectUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir][Tensor] Implement `reifyReturnTypeShapesPerResultDim` for `tensor.insert_slice`. Differential Revision: https://reviews.llvm.org/D105852
2021-07-13 14:51:20 -07:00
":InferTypeOpInterface",
[mlir][Tensor] Move ParallelInsertSlice to the tensor dialect This is moslty NFC and will allow tensor.parallel_insert_slice to gain rank-reducing semantics by reusing the vast majority of the tensor.insert_slice impl. Depends on D128857 Differential Revision: https://reviews.llvm.org/D128920
2022-06-30 04:27:41 -07:00
":ParallelCombiningOpInterface",
[mlir][interfaces] Add ShapedDimOpInterface This interface is implemented by memref.dim and tensor.dim. This change makes it possible to remove a build dependency of the Affine dialect on the Tensor dialect (and maybe also the MemRef dialect in the future). Differential Revision: https://reviews.llvm.org/D133595
2022-10-03 13:53:16 +09:00
":ShapedOpInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfaces",
[bazel] Add missing dependency after 9aa505a28d
2022-11-22 10:46:52 +01:00
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TensorOpsIncGen",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
":TilingInterface",
[mlir][Interfaces] Add ValueBoundsOpInterface Ops can implement this interface to specify lower/upper bounds for their result values and block arguments. Bounds can be specified for: * Index-type values * Dimension sizes of shapes values The bounds are added to a constraint set. Users can query this constraint set to compute bounds wrt. to a user-specified set of values. Only EQ bounds are supported at the moment. This revision also contains interface implementations for various tensor dialect ops, which illustrates how to implement this interface. Differential Revision: https://reviews.llvm.org/D145681
2023-04-06 02:41:15 +02:00
":ValueBoundsOpInterface",
Update Bazel BUILD files up to be9a87fe9b Differential Revision: https://reviews.llvm.org/D104791
2021-06-23 09:25:36 -07:00
":ViewLikeInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
[mlir] Move `linalg.tensor_expand/collapse_shape` to TensorDialect. RFC: https://llvm.discourse.group/t/rfc-reshape-ops-restructuring/3310 linalg.fill gets a canonicalizer, because `FoldFillWithTensorReshape` cannot be moved to tensorops (it uses linalg::FillOp inside). Before it was listed as a canonicalization pattern for the reshape operations, now it became a canonicalization for FillOp. Differential Revision: https://reviews.llvm.org/D115502
2021-12-10 12:03:47 +01:00
cc_library(
name = "TensorInferTypeOpInterfaceImpl",
srcs = ["lib/Dialect/Tensor/IR/TensorInferTypeOpInterfaceImpl.cpp"],
hdrs = ["include/mlir/Dialect/Tensor/IR/TensorInferTypeOpInterfaceImpl.h"],
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Fix Bazel build post c888a0ce8846e7ebf30914d4959125da80b3f566
2023-03-01 08:10:49 -08:00
":ArithUtils",
[MLIR] Expose `getAsValues` in `StaticValueUtils.h` (NFC) [reland] The utility function should live in `StaticValueUtils.h` as it provides a convenient way to convert a vector of OpFoldResults into a vector of Values. Reviewed By: nicolasvasilache, cota Differential Revision: https://reviews.llvm.org/D134451
2022-09-27 11:16:43 -04:00
":DialectUtils",
[mlir] Move `linalg.tensor_expand/collapse_shape` to TensorDialect. RFC: https://llvm.discourse.group/t/rfc-reshape-ops-restructuring/3310 linalg.fill gets a canonicalizer, because `FoldFillWithTensorReshape` cannot be moved to tensorops (it uses linalg::FillOp inside). Before it was listed as a canonicalization pattern for the reshape operations, now it became a canonicalization for FillOp. Differential Revision: https://reviews.llvm.org/D115502
2021-12-10 12:03:47 +01:00
":IR",
":InferTypeOpInterface",
":TensorDialect",
"//llvm:Support",
],
)
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
cc_library(
name = "TensorTilingInterfaceImpl",
srcs = ["lib/Dialect/Tensor/IR/TensorTilingInterfaceImpl.cpp"],
hdrs = ["include/mlir/Dialect/Tensor/IR/TensorTilingInterfaceImpl.h"],
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir][tensor] Implement TilingInterface for tensor.pack op. We can compute the offsets and sizes for the slice of input because the iteration domain is defined over outer loops. If the dimension is tiled, the i-th index is the product of offset_i and inner_tile_i. Different from tiling a pad op, we do not have to deal with reading zero data from input. Because the tiling sizes are indicated to packed outer dimensions. We will read either the entire tile or partial tile for each packed tile. The scf.if and tensor.generate ops are not needed in this context. Co-authored-by: Lorenzo Chelini <l.chelini@icloud.com> Reviewed By: rengolin, mravishankar Differential Revision: https://reviews.llvm.org/D138631
2022-11-23 18:07:12 -08:00
":AffineUtils",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithUtils",
[mlir][tensor] Implement TilingInterface for tensor.pack op. We can compute the offsets and sizes for the slice of input because the iteration domain is defined over outer loops. If the dimension is tiled, the i-th index is the product of offset_i and inner_tile_i. Different from tiling a pad op, we do not have to deal with reading zero data from input. Because the tiling sizes are indicated to packed outer dimensions. We will read either the entire tile or partial tile for each packed tile. The scf.if and tensor.generate ops are not needed in this context. Co-authored-by: Lorenzo Chelini <l.chelini@icloud.com> Reviewed By: rengolin, mravishankar Differential Revision: https://reviews.llvm.org/D138631
2022-11-23 18:07:12 -08:00
":DialectUtils",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
":IR",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
[mlir][tensor] Implement TilingInterface for unpack op The main issue of tiling unpack op is about incomplete tile. Since all the dimensions are orthogonal, discussing 1-d unpack case is enough. The core idea is to make the input slice have complete tiles. In this case, a larger unpacked tile will be created. We'll need an extract_slice op to shift and truncate the output. Take Nn_to_N as an example. Say that N=32, n=8, and tiling_size=15. The coordinates of second tile (i.e., result[15..31]) are [(1, 7), (2, 0,), (2, 1) ... (3, 6), (3, 7)]. The first row and the last row are incomplete in terms of inputs. It's impossible to represent an unpack op using the coordinates. Because the input has higher rank and the math computation of coordinate is using mod and ceilDiv. That's very tricky. To represent the unpack op, we have to complete the rows. I.e., the input coordinates would start with (1, 0); end with (3, 7). In this context, the tiled unpack produces a (3 * n) elements because there are 3 rows in total. Follow by a tensor.extract_slice op, we can get the actual result. If the tiling sizes are multiple of inner tile sizes, it is a perfect tiling case. In this context, the larger input and output is not needed. Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D139362
2022-12-02 13:25:48 -08:00
":LinalgUtils",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
":SCFDialect",
":TensorDialect",
[mlir][tensor] Implement TilingInterface for tensor.pack op. We can compute the offsets and sizes for the slice of input because the iteration domain is defined over outer loops. If the dimension is tiled, the i-th index is the product of offset_i and inner_tile_i. Different from tiling a pad op, we do not have to deal with reading zero data from input. Because the tiling sizes are indicated to packed outer dimensions. We will read either the entire tile or partial tile for each packed tile. The scf.if and tensor.generate ops are not needed in this context. Co-authored-by: Lorenzo Chelini <l.chelini@icloud.com> Reviewed By: rengolin, mravishankar Differential Revision: https://reviews.llvm.org/D138631
2022-11-23 18:07:12 -08:00
":TensorUtils",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
":TilingInterface",
[mlir] fix mismerge in bazel
2023-04-13 14:58:11 +02:00
":ValueBoundsOpInterface",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
"//llvm:Support",
],
)
cc_library(
name = "TensorUtils",
srcs = ["lib/Dialect/Tensor/Utils/Utils.cpp"],
hdrs = ["include/mlir/Dialect/Tensor/Utils/Utils.h"],
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][Tensor] Add a FoldTensorSubsetOps pass and patterns These patterns follow FoldMemRefAliasOps which is further refactored for reuse. In the process, fix FoldMemRefAliasOps handling of strides for vector.transfer ops which was previously incorrect. These opt-in patterns generalize the existing canonicalizations on vector.transfer ops. In the future the blanket canonicalizations will be retired. They are kept for now to minimize porting disruptions. Differential Revision: https://reviews.llvm.org/D146624
2023-03-21 14:41:20 -07:00
":ArithUtils",
[mlir][Linalg] Refactor HoistPadding and add support for hoisting in the absence of packing loops. This revision cleans up the implementation of hoist padding and extends it to also work in the absence of packing loops. This allows better composition when hoisting the padded result of a DPS operation. A systematic usage of RewriterBase is applied to the implementation. Depends on: D144856 Differential Revision: https://reviews.llvm.org/D144855
2023-02-27 01:28:56 -08:00
":DialectUtils",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
":TensorDialect",
[mlir][tensor] Add pattern to drop redundant insert_slice rank expansion Drop insert_slice rank expansions if they are directly followed by an inverse rank reduction. Differential Revision: https://reviews.llvm.org/D151800
2023-06-01 08:47:00 +02:00
":ValueBoundsOpInterface",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "TensorPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Tensor",
],
"include/mlir/Dialect/Tensor/Transforms/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Tensor/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "TensorTransforms",
srcs = glob(
[
"lib/Dialect/Tensor/Transforms/*.cpp",
"lib/Dialect/Tensor/Transforms/*.h",
],
),
[mlir][tensor] Move BufferizableOpInterface impl to tensor dialect This is in preparation of unifying the existing bufferization with One-Shot bufferization. A subsequent commit will replace `tensor-bufferize`'s implementation with the BufferizableOpInterface-based implementation and move over missing test cases. Differential Revision: https://reviews.llvm.org/D117984
2022-01-24 23:16:29 +09:00
hdrs = [
"include/mlir/Dialect/Tensor/Transforms/BufferizableOpInterfaceImpl.h",
"include/mlir/Dialect/Tensor/Transforms/Passes.h",
[bazel] Port ba8424a251d7
2022-09-19 21:20:12 +02:00
"include/mlir/Dialect/Tensor/Transforms/TransformUtils.h",
[mlir][tensor] Add a pattern to split tensor.pad ops This commit adds a pattern to wrap a tensor.pad op with an scf.if op to separate the cases where we don't need padding (all pad sizes are actually zeros) and where we indeed need padding. This pattern is meant to handle padding inside tiled loops. Under such cases the padding sizes typically depend on the loop induction variables. Splitting them would allow treating perfect tiles and edge tiles separately. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D117018
2022-02-16 11:26:47 -05:00
"include/mlir/Dialect/Tensor/Transforms/Transforms.h",
[mlir][tensor] Move BufferizableOpInterface impl to tensor dialect This is in preparation of unifying the existing bufferization with One-Shot bufferization. A subsequent commit will replace `tensor-bufferize`'s implementation with the BufferizableOpInterface-based implementation and move over missing test cases. Differential Revision: https://reviews.llvm.org/D117984
2022-01-24 23:16:29 +09:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[tensor][bufferize] Use affine.apply instead of arith.addi in PadOp lowering Affine exprs compose better than arith ops. Differential Revision: https://reviews.llvm.org/D132456
2022-08-23 11:38:54 +02:00
":AffineDialect",
[mlir][tensor] Add transform to make tensor.pad loop-independent Add a transform to make `tensor.pad` and `tensor.empty` ops independent of SCF loop IVs. Such ops can then be hoisted. E.g.: ``` scf.for %iv = %lb to %ub step %step { %high = affine.apply affine_map<(d0)[s0] -> (s0 - d0)> (%i)[%ub] %p = tensor.pad %t low[5] high[%high] ... ... } ``` Is transformed to: ``` %high_new = affine.apply affine_map<()[s0, s1] -> (-s0 + s1)> ()[%lb, %ub] %p_hoistable = tensor.pad %t low[5] high[%high_new] %dim = tensor.dim %t, %c0 %size = affine.apply affine_map<(d0)[s0, s1] -> (-d0 + s0 + s1 + 5)>(%iv)[%ub, %dim] %slice = tensor.extract_slice %p_hoistable [0] [%size] [1] ``` Differential Revision: https://reviews.llvm.org/D143910
2023-04-28 10:34:03 +09:00
":AffineTransforms",
[mlir] NFC: move mergeOffsetsSizesAndStrides into Affine/Utils So that these utility functions can also be used ViewLikeInterface ops not in the memref dialect. Reviewed By: mravishankar, christopherbate Differential Revision: https://reviews.llvm.org/D134487
2022-09-23 13:20:10 -04:00
":AffineUtils",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationDialect",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":BufferizationTransforms",
[mlir][tensor] Add a pattern to split tensor.pad ops This commit adds a pattern to wrap a tensor.pad op with an scf.if op to separate the cases where we don't need padding (all pad sizes are actually zeros) and where we indeed need padding. This pattern is meant to handle padding inside tiled loops. Under such cases the padding sizes typically depend on the loop induction variables. Splitting them would allow treating perfect tiles and edge tiles separately. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D117018
2022-02-16 11:26:47 -05:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir][tensor][bufferize] Lower tensor.generate to linalg.map There is no memref equivalent of tensor.generate. The purpose of this change is to avoid creating scf.parallel loops during bufferization. Differential Revision: https://reviews.llvm.org/D136767
2022-10-27 11:54:01 +02:00
":LinalgDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MemRefDialect",
":Pass",
":SCFDialect",
":TensorDialect",
":TensorPassIncGen",
[mlir][tensor] Add pattern to drop redundant insert_slice rank expansion Drop insert_slice rank expansions if they are directly followed by an inverse rank reduction. Differential Revision: https://reviews.llvm.org/D151800
2023-06-01 08:47:00 +02:00
":TensorUtils",
[mlir][TilingInterface] Enable tile and fuse using TilingInterface. This patch implements tile and fuse transformation for ops that implement the tiling interface. To do so, - `TilingInterface` needs a new method that generates a tiled implementation of the operation based on the tile of the result needed. - A pattern is added that replaces a `tensor.extract_slice` whose source is defined by an operation that implements the `TilingInterface` with a tiled implementation that produces the extracted slice in-place (using the method added to `TilingInterface`). - A pattern is added that takes a sequence of operations that implement the `TilingInterface` (for now `LinalgOp`s), tiles the consumer, and greedily fuses its producers iteratively. Differential Revision: https://reviews.llvm.org/D127809
2022-06-21 16:59:26 +00:00
":TilingInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
[mlir][tensor] Add transform to make tensor.pad loop-independent Add a transform to make `tensor.pad` and `tensor.empty` ops independent of SCF loop IVs. Such ops can then be hoisted. E.g.: ``` scf.for %iv = %lb to %ub step %step { %high = affine.apply affine_map<(d0)[s0] -> (s0 - d0)> (%i)[%ub] %p = tensor.pad %t low[5] high[%high] ... ... } ``` Is transformed to: ``` %high_new = affine.apply affine_map<()[s0, s1] -> (-s0 + s1)> ()[%lb, %ub] %p_hoistable = tensor.pad %t low[5] high[%high_new] %dim = tensor.dim %t, %c0 %size = affine.apply affine_map<(d0)[s0, s1] -> (-d0 + s0 + s1 + 5)>(%iv)[%ub, %dim] %slice = tensor.extract_slice %p_hoistable [0] [%size] [1] ``` Differential Revision: https://reviews.llvm.org/D143910
2023-04-28 10:34:03 +09:00
":ValueBoundsOpInterface",
[mlir][Tensor] Add a FoldTensorSubsetOps pass and patterns These patterns follow FoldMemRefAliasOps which is further refactored for reuse. In the process, fix FoldMemRefAliasOps handling of strides for vector.transfer ops which was previously incorrect. These opt-in patterns generalize the existing canonicalizations on vector.transfer ops. In the future the blanket canonicalizations will be retired. They are kept for now to minimize porting disruptions. Differential Revision: https://reviews.llvm.org/D146624
2023-03-21 14:41:20 -07:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
[mlir][tensor] Add transform to make tensor.pad loop-independent Add a transform to make `tensor.pad` and `tensor.empty` ops independent of SCF loop IVs. Such ops can then be hoisted. E.g.: ``` scf.for %iv = %lb to %ub step %step { %high = affine.apply affine_map<(d0)[s0] -> (s0 - d0)> (%i)[%ub] %p = tensor.pad %t low[5] high[%high] ... ... } ``` Is transformed to: ``` %high_new = affine.apply affine_map<()[s0, s1] -> (-s0 + s1)> ()[%lb, %ub] %p_hoistable = tensor.pad %t low[5] high[%high_new] %dim = tensor.dim %t, %c0 %size = affine.apply affine_map<(d0)[s0, s1] -> (-d0 + s0 + s1 + 5)>(%iv)[%ub, %dim] %slice = tensor.extract_slice %p_hoistable [0] [%size] [1] ``` Differential Revision: https://reviews.llvm.org/D143910
2023-04-28 10:34:03 +09:00
td_library(
name = "TensorTransformOpsTdFiles",
srcs = [
"include/mlir/Dialect/Tensor/TransformOps/TensorTransformOps.td",
],
includes = ["include"],
deps = [
":TransformDialectTdFiles",
],
)
gentbl_cc_library(
name = "TensorTransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Tensor/TransformOps/TensorTransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Tensor/TransformOps/TensorTransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Tensor/TransformOps/TensorTransformOps.td",
deps = [
":TensorTransformOpsTdFiles",
],
)
[mlir][transform] Add TrackingListener This change makes it possible to use a greedy pattern rewrite as part of a transform op, even if the transform op does not invalidate the target handle (in particular transform ops without `FunctionalStyleTransformOpTrait`) and the targeted op is not isolated from above. The listener API allows us to track replacements of ops with values, but not ops with ops. Therefore, the TrackingListener is conservative: If an op is replaced with values that all have the same defining op and the defining op is of the same type as the original op, it is safe to assume that the op was replaced with an equivalent op. Otherwise, the op mapping is dropped. When this is not good enough, transforms can track values instead or provide a custom `findReplacementOp` function. Differential Revision: https://reviews.llvm.org/D147039
2023-03-29 11:02:26 +02:00
cc_library(
name = "TensorTransformOps",
srcs = glob(["lib/Dialect/Tensor/TransformOps/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/Tensor/TransformOps/*.h"]),
includes = ["include"],
deps = [
[mlir][tensor] Add transform to make tensor.pad loop-independent Add a transform to make `tensor.pad` and `tensor.empty` ops independent of SCF loop IVs. Such ops can then be hoisted. E.g.: ``` scf.for %iv = %lb to %ub step %step { %high = affine.apply affine_map<(d0)[s0] -> (s0 - d0)> (%i)[%ub] %p = tensor.pad %t low[5] high[%high] ... ... } ``` Is transformed to: ``` %high_new = affine.apply affine_map<()[s0, s1] -> (-s0 + s1)> ()[%lb, %ub] %p_hoistable = tensor.pad %t low[5] high[%high_new] %dim = tensor.dim %t, %c0 %size = affine.apply affine_map<(d0)[s0, s1] -> (-d0 + s0 + s1 + 5)>(%iv)[%ub, %dim] %slice = tensor.extract_slice %p_hoistable [0] [%size] [1] ``` Differential Revision: https://reviews.llvm.org/D143910
2023-04-28 10:34:03 +09:00
":AffineDialect",
[mlir][transform] Add TrackingListener This change makes it possible to use a greedy pattern rewrite as part of a transform op, even if the transform op does not invalidate the target handle (in particular transform ops without `FunctionalStyleTransformOpTrait`) and the targeted op is not isolated from above. The listener API allows us to track replacements of ops with values, but not ops with ops. Therefore, the TrackingListener is conservative: If an op is replaced with values that all have the same defining op and the defining op is of the same type as the original op, it is safe to assume that the op was replaced with an equivalent op. Otherwise, the op mapping is dropped. When this is not good enough, transforms can track values instead or provide a custom `findReplacementOp` function. Differential Revision: https://reviews.llvm.org/D147039
2023-03-29 11:02:26 +02:00
":IR",
[mlir][tensor] Add transform to make tensor.pad loop-independent Add a transform to make `tensor.pad` and `tensor.empty` ops independent of SCF loop IVs. Such ops can then be hoisted. E.g.: ``` scf.for %iv = %lb to %ub step %step { %high = affine.apply affine_map<(d0)[s0] -> (s0 - d0)> (%i)[%ub] %p = tensor.pad %t low[5] high[%high] ... ... } ``` Is transformed to: ``` %high_new = affine.apply affine_map<()[s0, s1] -> (-s0 + s1)> ()[%lb, %ub] %p_hoistable = tensor.pad %t low[5] high[%high_new] %dim = tensor.dim %t, %c0 %size = affine.apply affine_map<(d0)[s0, s1] -> (-d0 + s0 + s1 + 5)>(%iv)[%ub, %dim] %slice = tensor.extract_slice %p_hoistable [0] [%size] [1] ``` Differential Revision: https://reviews.llvm.org/D143910
2023-04-28 10:34:03 +09:00
":SCFDialect",
[mlir][transform] Add TrackingListener This change makes it possible to use a greedy pattern rewrite as part of a transform op, even if the transform op does not invalidate the target handle (in particular transform ops without `FunctionalStyleTransformOpTrait`) and the targeted op is not isolated from above. The listener API allows us to track replacements of ops with values, but not ops with ops. Therefore, the TrackingListener is conservative: If an op is replaced with values that all have the same defining op and the defining op is of the same type as the original op, it is safe to assume that the op was replaced with an equivalent op. Otherwise, the op mapping is dropped. When this is not good enough, transforms can track values instead or provide a custom `findReplacementOp` function. Differential Revision: https://reviews.llvm.org/D147039
2023-03-29 11:02:26 +02:00
":TensorDialect",
[mlir][tensor] Add transform to make tensor.pad loop-independent Add a transform to make `tensor.pad` and `tensor.empty` ops independent of SCF loop IVs. Such ops can then be hoisted. E.g.: ``` scf.for %iv = %lb to %ub step %step { %high = affine.apply affine_map<(d0)[s0] -> (s0 - d0)> (%i)[%ub] %p = tensor.pad %t low[5] high[%high] ... ... } ``` Is transformed to: ``` %high_new = affine.apply affine_map<()[s0, s1] -> (-s0 + s1)> ()[%lb, %ub] %p_hoistable = tensor.pad %t low[5] high[%high_new] %dim = tensor.dim %t, %c0 %size = affine.apply affine_map<(d0)[s0, s1] -> (-d0 + s0 + s1 + 5)>(%iv)[%ub, %dim] %slice = tensor.extract_slice %p_hoistable [0] [%size] [1] ``` Differential Revision: https://reviews.llvm.org/D143910
2023-04-28 10:34:03 +09:00
":TensorTransformOpsIncGen",
":TensorTransforms",
[mlir][tensor] Add pattern to drop redundant insert_slice rank expansion Drop insert_slice rank expansions if they are directly followed by an inverse rank reduction. Differential Revision: https://reviews.llvm.org/D151800
2023-06-01 08:47:00 +02:00
":TensorUtils",
[mlir][transform] Add TrackingListener This change makes it possible to use a greedy pattern rewrite as part of a transform op, even if the transform op does not invalidate the target handle (in particular transform ops without `FunctionalStyleTransformOpTrait`) and the targeted op is not isolated from above. The listener API allows us to track replacements of ops with values, but not ops with ops. Therefore, the TrackingListener is conservative: If an op is replaced with values that all have the same defining op and the defining op is of the same type as the original op, it is safe to assume that the op was replaced with an equivalent op. Otherwise, the op mapping is dropped. When this is not good enough, transforms can track values instead or provide a custom `findReplacementOp` function. Differential Revision: https://reviews.llvm.org/D147039
2023-03-29 11:02:26 +02:00
":TransformDialect",
[mlir][tensor] TrackingListener: Support cast-like InsertSliceOps with dynamic shape When looking for payload op replacements, rank-expanding InsertSliceOps of dynamically-typed tensors are now supported. Differential Revision: https://reviews.llvm.org/D151444
2023-05-25 19:10:05 +02:00
":ValueBoundsOpInterface",
[mlir][transform] Add TrackingListener This change makes it possible to use a greedy pattern rewrite as part of a transform op, even if the transform op does not invalidate the target handle (in particular transform ops without `FunctionalStyleTransformOpTrait`) and the targeted op is not isolated from above. The listener API allows us to track replacements of ops with values, but not ops with ops. Therefore, the TrackingListener is conservative: If an op is replaced with values that all have the same defining op and the defining op is of the same type as the original op, it is safe to assume that the op was replaced with an equivalent op. Otherwise, the op mapping is dropped. When this is not good enough, transforms can track values instead or provide a custom `findReplacementOp` function. Differential Revision: https://reviews.llvm.org/D147039
2023-03-29 11:02:26 +02:00
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "Rewrite",
srcs = glob([
"lib/Rewrite/*.cpp",
"lib/Rewrite/*.h",
]),
hdrs = glob(["include/mlir/Rewrite/*.h"]),
includes = ["include"],
deps = [
":Analysis",
":IR",
":PDLDialect",
":PDLInterpDialect",
":PDLToPDLInterp",
":Pass",
":SideEffectInterfaces",
"//llvm:Support",
],
)
cc_library(
name = "TransformUtils",
srcs = glob([
"lib/Transforms/Utils/*.cpp",
"lib/Transforms/Utils/*.h",
]),
Remove spurious includes and dependencies from Bazel files (NFC) Differential Revision: https://reviews.llvm.org/D119526
2022-02-11 07:14:40 +00:00
hdrs = glob(
[
"include/mlir/Transforms/*.h",
],
exclude = ["include/mlir/Transforms/Passes.h"],
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":ControlFlowInterfaces",
":IR",
[mlir] Refactor LICM into a utility LICM is refactored into a utility that is application on any region. The implementation is moved to Transform/Utils.
2022-04-16 00:36:22 +00:00
":LoopLikeInterface",
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
":MemorySlotInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Rewrite",
":SideEffectInterfaces",
":Support",
":TransformsPassIncGen",
[MLIR] Reformat the Bazel build This patch normalizes formatting of the the root BUILD.bazel file by: 1) adjusting indentation a little bit, 2) alphabetically ordering dependencies. These small deviations were introduced by some yesterday's patches: * https://reviews.llvm.org/D151104 * https://reviews.llvm.org/D151346 * https://reviews.llvm.org/rG16fe2b37365c00b0c6d0ed22c2e6521f2d5de01a * https://reviews.llvm.org/rG4d1cd1d8caab13d6b76ce6fc4ff76a01a7931c34 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D151499
2023-05-26 14:16:47 -07:00
":config",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
gentbl_cc_library(
name = "DerivedAttributeOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/DerivedAttributeOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/DerivedAttributeOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/DerivedAttributeOpInterface.td",
deps = [":DerivedAttributeOpInterfaceTdFiles"],
)
cc_library(
name = "DerivedAttributeOpInterface",
srcs = ["lib/Interfaces/DerivedAttributeOpInterface.cpp"],
hdrs = ["include/mlir/Interfaces/DerivedAttributeOpInterface.h"],
includes = ["include"],
deps = [
":DerivedAttributeOpInterfaceIncGen",
":IR",
"//llvm:Support",
],
)
[mlir][interfaces][NFC] Move DestinationStyleOpInterface to mlir/Interfaces This is the second (and final) step of making "destination style" usable without depending on the Linalg dialect. (The first step was D135129.) This change allows us to provide default bufferization implementations for all destination-style ops. It also allows us to simplify `TilingInterface`. (E.g., `getDestinationOperands` can be removed.) Differential Revision: https://reviews.llvm.org/D136179
2022-10-18 17:23:42 +02:00
gentbl_cc_library(
name = "DestinationStyleOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/DestinationStyleOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/DestinationStyleOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/DestinationStyleOpInterface.td",
deps = [":DestinationStyleOpInterfaceTdFiles"],
)
cc_library(
name = "DestinationStyleOpInterface",
srcs = ["lib/Interfaces/DestinationStyleOpInterface.cpp"],
hdrs = ["include/mlir/Interfaces/DestinationStyleOpInterface.h"],
includes = ["include"],
deps = [
":DestinationStyleOpInterfaceIncGen",
":IR",
"//llvm:Support",
],
)
[bazel] Port 95aff23e2921
2022-06-03 20:43:02 +02:00
gentbl_cc_library(
name = "InferIntRangeInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/InferIntRangeInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/InferIntRangeInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/InferIntRangeInterface.td",
deps = [":InferIntRangeInterfaceTdFiles"],
)
cc_library(
name = "InferIntRangeInterface",
srcs = ["lib/Interfaces/InferIntRangeInterface.cpp"],
hdrs = ["include/mlir/Interfaces/InferIntRangeInterface.h"],
includes = ["include"],
deps = [
":IR",
":InferIntRangeInterfaceIncGen",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "DataLayoutInterfacesTdFiles",
srcs = ["include/mlir/Interfaces/DataLayoutInterfaces.td"],
includes = ["include"],
)
gentbl_cc_library(
name = "DataLayoutInterfacesIncGen",
tbl_outs = [
(
["-gen-attr-interface-decls"],
"include/mlir/Interfaces/DataLayoutAttrInterface.h.inc",
),
(
["-gen-attr-interface-defs"],
"include/mlir/Interfaces/DataLayoutAttrInterface.cpp.inc",
),
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/DataLayoutOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/DataLayoutOpInterface.cpp.inc",
),
(
["-gen-type-interface-decls"],
"include/mlir/Interfaces/DataLayoutTypeInterface.h.inc",
),
(
["-gen-type-interface-defs"],
"include/mlir/Interfaces/DataLayoutTypeInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/DataLayoutInterfaces.td",
deps = [":OpBaseTdFiles"],
)
gentbl_cc_library(
name = "LoopLikeInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/LoopLikeInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/LoopLikeInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/LoopLikeInterface.td",
deps = [":LoopLikeInterfaceTdFiles"],
)
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
gentbl_cc_library(
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
name = "MemorySlotInterfacesIncGen",
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
"include/mlir/Interfaces/MemorySlotOpInterfaces.h.inc",
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
),
(
["-gen-op-interface-defs"],
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
"include/mlir/Interfaces/MemorySlotOpInterfaces.cpp.inc",
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
),
[Bazel] Add missing tablegen. (For MemorySlotTypeInterfaces) Summary: Reviewers: khasanovaa Subscribers:
2023-05-22 13:07:45 +02:00
(
["-gen-type-interface-decls"],
"include/mlir/Interfaces/MemorySlotTypeInterfaces.h.inc",
),
(
["-gen-type-interface-defs"],
"include/mlir/Interfaces/MemorySlotTypeInterfaces.cpp.inc",
),
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
],
tblgen = ":mlir-tblgen",
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
td_file = "include/mlir/Interfaces/MemorySlotInterfaces.td",
deps = [":MemorySlotInterfacesTdFiles"],
[bazel][mlir] update build for mem2reg f88f8fd0bca34dc8e5571cb828b1a159a50fd504 Differential Revision: https://reviews.llvm.org/D149326
2023-04-27 10:43:13 +02:00
)
[mlir][interfaces] Add ShapedDimOpInterface This interface is implemented by memref.dim and tensor.dim. This change makes it possible to remove a build dependency of the Affine dialect on the Tensor dialect (and maybe also the MemRef dialect in the future). Differential Revision: https://reviews.llvm.org/D133595
2022-10-03 13:53:16 +09:00
gentbl_cc_library(
name = "ShapedOpInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/ShapedOpInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/ShapedOpInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/ShapedOpInterfaces.td",
deps = [":ShapedOpInterfacesTdFiles"],
)
[fix/build] bazel rule for ParallelCombiningOpInterface
2022-07-01 10:33:48 +02:00
gentbl_cc_library(
name = "ParallelCombiningOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/ParallelCombiningOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/ParallelCombiningOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/ParallelCombiningOpInterface.td",
deps = [":ParallelCombiningOpInterfaceTdFiles"],
)
[mlir] Add RuntimeVerifiableOpInterface and transform Static op verification cannot detect cases where an op is valid at compile time but may be invalid at runtime. An example of such an op is `memref::ExpandShapeOp`. Invalid at compile time: `memref.expand_shape %m [[0, 1]] : memref<11xf32> into memref<2x5xf32>` Valid at compile time (because we do not know any better): `memref.expand_shape %m [[0, 1]] : memref<?xf32> into memref<?x5xf32>`. This op may or may not be valid at runtime depending on the runtime shape of `%m`. Invalid runtime ops such as the one above are hard to debug because they can crash the program execution at a seemingly unrelated position or (even worse) compute an invalid result without crashing. This revision adds a new op interface `RuntimeVerifiableOpInterface` that can be implemented by ops that provide additional runtime verification. Such runtime verification can be computationally expensive, so it is only generated on an opt-in basis by running `-generate-runtime-verification`. A simple runtime verifier for `memref::ExpandShapeOp` is provided as an example. Differential Revision: https://reviews.llvm.org/D138576
2022-12-21 10:51:10 +01:00
gentbl_cc_library(
name = "RuntimeVerifiableOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/RuntimeVerifiableOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/RuntimeVerifiableOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/RuntimeVerifiableOpInterface.td",
deps = [":RuntimeVerifiableOpInterfaceTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "VectorInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/VectorInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/VectorInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/VectorInterfaces.td",
deps = [":VectorInterfacesTdFiles"],
)
gentbl_cc_library(
name = "ViewLikeInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/ViewLikeInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/ViewLikeInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/ViewLikeInterface.td",
deps = [":ViewLikeInterfaceTdFiles"],
)
gentbl_cc_library(
name = "CopyOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/CopyOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/CopyOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/CopyOpInterface.td",
deps = [":CopyOpInterfaceTdFiles"],
)
gentbl_cc_library(
name = "TransformsPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Transforms",
],
"include/mlir/Transforms/Passes.h.inc",
),
(
[
"-gen-pass-capi-header",
"--prefix=Transforms",
],
"include/mlir/Transforms/Transforms.capi.h.inc",
),
(
[
"-gen-pass-capi-impl",
"--prefix=Transforms",
],
"include/mlir/Transforms/Transforms.capi.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Transforms/Passes.td",
deps = [
":PassBaseTdFiles",
":RewritePassBaseTdFiles",
],
)
cc_library(
name = "Transforms",
srcs = glob([
"lib/Transforms/*.cpp",
"lib/Transforms/*.h",
]),
hdrs = glob(["include/mlir/Transforms/*.h"]),
includes = ["include"],
deps = [
":Analysis",
":ControlFlowInterfaces",
":IR",
":LoopLikeInterface",
[bazel] Port BUILD rules for 92cc30aca78703174e89880aca85e5a5ff41cc98.
2023-05-09 10:21:07 +00:00
":MemorySlotInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":Rewrite",
[mlir] Add RuntimeVerifiableOpInterface and transform Static op verification cannot detect cases where an op is valid at compile time but may be invalid at runtime. An example of such an op is `memref::ExpandShapeOp`. Invalid at compile time: `memref.expand_shape %m [[0, 1]] : memref<11xf32> into memref<2x5xf32>` Valid at compile time (because we do not know any better): `memref.expand_shape %m [[0, 1]] : memref<?xf32> into memref<?x5xf32>`. This op may or may not be valid at runtime depending on the runtime shape of `%m`. Invalid runtime ops such as the one above are hard to debug because they can crash the program execution at a seemingly unrelated position or (even worse) compute an invalid result without crashing. This revision adds a new op interface `RuntimeVerifiableOpInterface` that can be implemented by ops that provide additional runtime verification. Such runtime verification can be computationally expensive, so it is only generated on an opt-in basis by running `-generate-runtime-verification`. A simple runtime verifier for `memref::ExpandShapeOp` is provided as an example. Differential Revision: https://reviews.llvm.org/D138576
2022-12-21 10:51:10 +01:00
":RuntimeVerifiableOpInterface",
[mlir] Remove `Transforms/SideEffectUtils.h` and move the methods into `Interface/SideEffectInterfaces.h`. The methods in `SideEffectUtils.h` (and their implementations in `SideEffectUtils.cpp`) seem to have similar intent to methods already existing in `SideEffectInterfaces.h`. Move the decleration (and implementation) from `SideEffectUtils.h` (and `SideEffectUtils.cpp`) into `SideEffectInterfaces.h` (and `SideEffectInterface.cpp`). Also drop the `SideEffectInterface::hasNoEffect` method in favor of `mlir::isMemoryEffectFree` which actually recurses into the operation instead of just relying on the `hasRecursiveMemoryEffectTrait` exclusively. Differential Revision: https://reviews.llvm.org/D137857
2022-11-11 17:10:38 +00:00
":SideEffectInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":TransformUtils",
":TransformsPassIncGen",
"//llvm:Support",
],
)
cc_library(
name = "CommonFolders",
srcs = [
],
hdrs = ["include/mlir/Dialect/CommonFolders.h"],
includes = ["include"],
deps = [
":IR",
"//llvm:Support",
],
)
cc_library(
name = "SCFToGPU",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
srcs = glob(["lib/Conversion/SCFToGPU/*.cpp"]),
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
hdrs = glob(["include/mlir/Conversion/SCFToGPU/*.h"]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AffineToStandard",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ComplexDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":GPUTransforms",
":IR",
":MemRefDialect",
":Pass",
":SCFDialect",
[mlir] Remove `Transforms/SideEffectUtils.h` and move the methods into `Interface/SideEffectInterfaces.h`. The methods in `SideEffectUtils.h` (and their implementations in `SideEffectUtils.cpp`) seem to have similar intent to methods already existing in `SideEffectInterfaces.h`. Move the decleration (and implementation) from `SideEffectUtils.h` (and `SideEffectUtils.cpp`) into `SideEffectInterfaces.h` (and `SideEffectInterface.cpp`). Also drop the `SideEffectInterface::hasNoEffect` method in favor of `mlir::isMemoryEffectFree` which actually recurses into the operation instead of just relying on the `hasRecursiveMemoryEffectTrait` exclusively. Differential Revision: https://reviews.llvm.org/D137857
2022-11-11 17:10:38 +00:00
":SideEffectInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":TransformUtils",
":Transforms",
"//llvm:Support",
],
)
cc_library(
name = "SCFToSPIRV",
srcs = glob([
"lib/Conversion/SCFToSPIRV/*.cpp",
"lib/Conversion/SCFToSPIRV/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/SCFToSPIRV/*.h",
]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir] NFC: split MemRef to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107094
2021-07-29 16:22:46 -04:00
":MemRefToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":SCFDialect",
":SPIRVConversion",
":SPIRVDialect",
":Support",
":TransformUtils",
":Transforms",
"//llvm:Support",
],
)
cc_library(
name = "SCFToOpenMP",
srcs = [
"lib/Conversion/SCFToOpenMP/SCFToOpenMP.cpp",
],
hdrs = ["include/mlir/Conversion/SCFToOpenMP/SCFToOpenMP.h"],
includes = ["include"],
deps = [
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
":AffineAnalysis",
[mlir] Create a generic reduction detection utility This patch introduces a generic reduction detection utility that works across different dialecs. It is mostly a generalization of the reduction detection algorithm in Affine. The reduction detection logic in Affine, Linalg and SCFToOpenMP have been replaced with this new generic utility. The utility takes some basic components of the potential reduction and returns: 1) the reduced value, and 2) a list with the combiner operations. The logic to match reductions involving multiple combiner operations disabled until we can properly test it. Reviewed By: ftynse, bondhugula, nicolasvasilache, pifon2a Differential Revision: https://reviews.llvm.org/D110303
2021-09-22 17:11:45 +00:00
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir] support reductions in SCF to OpenMP conversion OpenMP reductions need a neutral element, so we match some known reduction kinds (integer add/mul/or/and/xor, float add/mul, integer and float min/max) to define the neutral element and the atomic version when possible to express using atomicrmw (everything except float mul). The SCF-to-OpenMP pass becomes a module pass because it now needs to introduce new symbols for reduction declarations in the module. Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D107549
2021-08-25 11:07:17 +02:00
":LLVMDialect",
[bazel] add missing dependency
2022-03-01 10:43:27 -08:00
":MemRefDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpenMPDialect",
":Pass",
":SCFDialect",
":Support",
":Transforms",
],
)
cc_library(
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
name = "SCFToControlFlow",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = [
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
"lib/Conversion/SCFToControlFlow/SCFToControlFlow.cpp",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
hdrs = ["include/mlir/Conversion/SCFToControlFlow/SCFToControlFlow.h"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":LLVMDialect",
":Pass",
":SCFDialect",
":Support",
":TransformUtils",
":Transforms",
],
)
[Bazel] Fixes for b5d847b1b95750d0af40cfc8c71a8fec50bb8613 and 6412a13539ab2914eed8e1df83c399b9a16e3408
2021-07-07 16:50:23 -07:00
cc_library(
name = "LLVMCommonConversion",
srcs = glob([
"lib/Conversion/LLVMCommon/*.cpp",
]) + ["lib/Conversion/LLVMCommon/MemRefDescriptor.h"],
hdrs = glob(["include/mlir/Conversion/LLVMCommon/*.h"]),
includes = ["include"],
deps = [
":IR",
":LLVMDialect",
":Support",
":Transforms",
"//llvm:Core",
],
)
[mlir] Factor type reconciliation out of Standard-to-LLVM conversion Conversion to the LLVM dialect is being refactored to be more progressive and is now performed as a series of independent passes converting different dialects. These passes may produce `unrealized_conversion_cast` operations that represent pending conversions between built-in and LLVM dialect types. Historically, a more monolithic Standard-to-LLVM conversion pass did not need these casts as all operations were converted in one shot. Previous refactorings have led to the requirement of running the Standard-to-LLVM conversion pass to clean up `unrealized_conversion_cast`s even though the IR had no standard operations in it. The pass must have been also run the last among all to-LLVM passes, in contradiction with the partial conversion logic. Additionally, the way it was set up could produce invalid operations by removing casts between LLVM and built-in types even when the consumer did not accept the uncasted type, or could lead to cryptic conversion errors (recursive application of the rewrite pattern on `unrealized_conversion_cast` as a means to indicate failure to eliminate casts). In fact, the need to eliminate A->B->A `unrealized_conversion_cast`s is not specific to to-LLVM conversions and can be factored out into a separate type reconciliation pass, which is achieved in this commit. While the cast operation itself has a folder pattern, it is insufficient in most conversion passes as the folder only applies to the second cast. Without complex legality setup in the conversion target, the conversion infra will either consider the cast operations valid and not fold them (a separate canonicalization would be necessary to trigger the folding), or consider the first cast invalid upon generation and stop with error. The pattern provided by the reconciliation pass applies to the first cast operation instead. Furthermore, having a separate pass makes it clear when `unrealized_conversion_cast`s could not have been eliminated since it is the only reason why this pass can fail. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D109507
2021-09-09 16:06:10 +02:00
cc_library(
name = "ReconcileUnrealizedCasts",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
srcs = glob(["lib/Conversion/ReconcileUnrealizedCasts/*.cpp"]),
[mlir] Factor type reconciliation out of Standard-to-LLVM conversion Conversion to the LLVM dialect is being refactored to be more progressive and is now performed as a series of independent passes converting different dialects. These passes may produce `unrealized_conversion_cast` operations that represent pending conversions between built-in and LLVM dialect types. Historically, a more monolithic Standard-to-LLVM conversion pass did not need these casts as all operations were converted in one shot. Previous refactorings have led to the requirement of running the Standard-to-LLVM conversion pass to clean up `unrealized_conversion_cast`s even though the IR had no standard operations in it. The pass must have been also run the last among all to-LLVM passes, in contradiction with the partial conversion logic. Additionally, the way it was set up could produce invalid operations by removing casts between LLVM and built-in types even when the consumer did not accept the uncasted type, or could lead to cryptic conversion errors (recursive application of the rewrite pattern on `unrealized_conversion_cast` as a means to indicate failure to eliminate casts). In fact, the need to eliminate A->B->A `unrealized_conversion_cast`s is not specific to to-LLVM conversions and can be factored out into a separate type reconciliation pass, which is achieved in this commit. While the cast operation itself has a folder pattern, it is insufficient in most conversion passes as the folder only applies to the second cast. Without complex legality setup in the conversion target, the conversion infra will either consider the cast operations valid and not fold them (a separate canonicalization would be necessary to trigger the folding), or consider the first cast invalid upon generation and stop with error. The pattern provided by the reconciliation pass applies to the first cast operation instead. Furthermore, having a separate pass makes it clear when `unrealized_conversion_cast`s could not have been eliminated since it is the only reason why this pass can fail. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D109507
2021-09-09 16:06:10 +02:00
hdrs = glob(["include/mlir/Conversion/ReconcileUnrealizedCasts/*.h"]),
includes = ["include"],
deps = [
":ConversionPassIncGen",
":IR",
":Pass",
":TransformUtils",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
name = "FuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = [
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
"lib/Conversion/FuncToLLVM/FuncToLLVM.cpp",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
hdrs = [
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
"include/mlir/Conversion/FuncToLLVM/ConvertFuncToLLVM.h",
"include/mlir/Conversion/FuncToLLVM/ConvertFuncToLLVMPass.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToLLVM",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":ControlFlowToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
":DataLayoutInterfaces",
[Bazel] Update for 0813700de1 Updates Bazel BUILD files for changes in https://github.com/llvm/llvm-project/commit/0813700de1 Differential Revision: https://reviews.llvm.org/D104990
2021-06-27 08:30:54 -07:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
":FuncTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Fixes for b5d847b1b95750d0af40cfc8c71a8fec50bb8613 and 6412a13539ab2914eed8e1df83c399b9a16e3408
2021-07-07 16:50:23 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
":MathDialect",
":MemRefDialect",
":Parser",
":Pass",
":Support",
":TransformUtils",
":Transforms",
"//llvm:Core",
"//llvm:Support",
],
)
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
cc_library(
name = "ControlFlowToLLVM",
srcs = [
"lib/Conversion/ControlFlowToLLVM/ControlFlowToLLVM.cpp",
],
hdrs = [
"include/mlir/Conversion/ControlFlowToLLVM/ControlFlowToLLVM.h",
],
includes = ["include"],
deps = [
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToLLVM",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":ConversionPassIncGen",
":DataLayoutInterfaces",
":DialectUtils",
":IR",
":LLVMCommonConversion",
":LLVMDialect",
":MathDialect",
":MemRefDialect",
":Parser",
":Pass",
":Support",
":TransformUtils",
":Transforms",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "ControlFlowToSPIRV",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
srcs = glob(["lib/Conversion/ControlFlowToSPIRV/*.cpp"]),
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
hdrs = glob(["include/mlir/Conversion/ControlFlowToSPIRV/*.h"]),
includes = ["include"],
deps = [
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":ConversionPassIncGen",
":IR",
":Pass",
":SPIRVCommonConversion",
":SPIRVConversion",
":SPIRVDialect",
":SPIRVUtils",
":Support",
":Transforms",
"//llvm:Support",
],
)
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
cc_library(
name = "MemRefToLLVM",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
srcs = glob(["lib/Conversion/MemRefToLLVM/*.cpp"]),
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
hdrs = glob(["include/mlir/Conversion/MemRefToLLVM/*.h"]),
includes = ["include"],
deps = [
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":ConversionPassIncGen",
":DataLayoutInterfaces",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":IR",
":LLVMCommonConversion",
":LLVMDialect",
":MemRefDialect",
[mlir][MemRef] Move narrow type emulation common methods to MemRefUtils. It also unifies the computation of StridedLayoutAttr. If the stride is static known value, we can just use it. Differential Revision: https://reviews.llvm.org/D155017
2023-07-11 14:13:07 -07:00
":MemRefUtils",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":Pass",
":Support",
":Transforms",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
"//llvm:Support",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
],
)
[mlir] NFC: split MemRef to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107094
2021-07-29 16:22:46 -04:00
cc_library(
name = "MemRefToSPIRV",
srcs = glob([
"lib/Conversion/MemRefToSPIRV/*.cpp",
"lib/Conversion/MemRefToSPIRV/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
[mlir] NFC: split MemRef to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107094
2021-07-29 16:22:46 -04:00
hdrs = glob([
"include/mlir/Conversion/MemRefToSPIRV/*.h",
]),
includes = [
"include",
"lib/Conversion/MemRefToSPIRV",
],
deps = [
[mlir] bazel fix Reviewed By: olegshyshkov Differential Revision: https://reviews.llvm.org/D144124
2023-02-15 10:33:10 -08:00
":ArithDialect",
[mlir] NFC: split MemRef to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107094
2021-07-29 16:22:46 -04:00
":ConversionPassIncGen",
[bazel] Add missing dependency after 713f85d5952ab27d474aba2a960a893b7e7e438d
2022-08-06 11:34:24 +02:00
":FuncDialect",
[mlir] NFC: split MemRef to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107094
2021-07-29 16:22:46 -04:00
":IR",
":MemRefDialect",
":Pass",
":SPIRVConversion",
":SPIRVDialect",
":Support",
":Transforms",
"//llvm:Support",
],
)
[bazel] Port 589764a38264
2022-11-04 19:06:44 +01:00
cc_library(
name = "ArithAttrToLLVMConversion",
srcs = glob(["lib/Conversion/ArithCommon/*.cpp"]),
hdrs = glob(["include/mlir/Conversion/ArithCommon/*.h"]),
includes = ["include"],
deps = [
":ArithDialect",
":LLVMDialect",
],
)
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
cc_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithToLLVM",
srcs = glob(["lib/Conversion/ArithToLLVM/*.cpp"]),
hdrs = glob(["include/mlir/Conversion/ArithToLLVM/*.h"]),
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
includes = ["include"],
deps = [
":Analysis",
[bazel] Port 589764a38264
2022-11-04 19:06:44 +01:00
":ArithAttrToLLVMConversion",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":ConversionPassIncGen",
":IR",
":LLVMCommonConversion",
":LLVMDialect",
":Pass",
":Support",
":Transforms",
],
)
cc_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithToSPIRV",
srcs = glob(["lib/Conversion/ArithToSPIRV/*.cpp"]),
hdrs = glob(["include/mlir/Conversion/ArithToSPIRV/*.h"]),
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":ConversionPassIncGen",
Attempt fix for Bazel build from 3ba66435d9377cb99ee1718807cbf6a7c4b0d6d3
2022-03-02 21:45:25 +00:00
":FuncToSPIRV",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":IR",
":Pass",
":SPIRVCommonConversion",
":SPIRVConversion",
":SPIRVDialect",
":Support",
":Transforms",
"//llvm:Support",
],
)
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
cc_library(
name = "MathToLLVM",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
srcs = glob(["lib/Conversion/MathToLLVM/*.cpp"]),
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
hdrs = glob(["include/mlir/Conversion/MathToLLVM/*.h"]),
includes = ["include"],
deps = [
":Analysis",
[bazel] Port 589764a38264
2022-11-04 19:06:44 +01:00
":ArithAttrToLLVMConversion",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":ConversionPassIncGen",
":DataLayoutInterfaces",
":IR",
":LLVMCommonConversion",
":LLVMDialect",
":MathDialect",
":Pass",
":Support",
":Transforms",
],
)
[mlir][Bazel] Fix bazel build.
2022-08-26 09:04:50 +02:00
cc_library(
name = "MathToFuncs",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
srcs = glob(["lib/Conversion/MathToFuncs/*.cpp"]),
[mlir][Bazel] Fix bazel build.
2022-08-26 09:04:50 +02:00
hdrs = glob(["include/mlir/Conversion/MathToFuncs/*.h"]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][Bazel] Fix bazel build.
2022-08-26 09:04:50 +02:00
":ControlFlowDialect",
":ConversionPassIncGen",
":DialectUtils",
":FuncDialect",
":IR",
":LLVMDialect",
":MathDialect",
":Pass",
[mlir] fix mismerge in bazel
2023-04-13 14:58:11 +02:00
":SCFDialect",
[mlir][Bazel] Fix bazel build.
2022-08-26 09:04:50 +02:00
":Transforms",
":VectorDialect",
":VectorUtils",
[mlir][Bazel] Fix wrong dependency to //llvm:Support
2022-08-26 09:06:40 +02:00
"//llvm:Support",
[mlir][Bazel] Fix bazel build.
2022-08-26 09:04:50 +02:00
],
)
[mlir][bazel] Port for 660f714e26999d266232a1fbb02712bb879bd34e, second attempt.
2023-05-27 08:37:45 +02:00
gentbl_cc_library(
name = "BytecodeOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Bytecode/BytecodeOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Bytecode/BytecodeOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Bytecode/BytecodeOpInterface.td",
deps = [":BytecodeOpInterfaceTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "CallOpInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/CallInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/CallInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/CallInterfaces.td",
deps = [":CallInterfacesTdFiles"],
)
cc_library(
name = "CallOpInterfaces",
srcs = ["lib/Interfaces/CallInterfaces.cpp"],
hdrs = ["include/mlir/Interfaces/CallInterfaces.h"],
includes = ["include"],
deps = [
":CallOpInterfacesIncGen",
":IR",
"//llvm:Support",
],
)
gentbl_cc_library(
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
name = "CastInterfacesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/CastInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/CastInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/CastInterfaces.td",
deps = [":CastInterfacesTdFiles"],
)
cc_library(
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
name = "CastInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = ["lib/Interfaces/CastInterfaces.cpp"],
hdrs = ["include/mlir/Interfaces/CastInterfaces.h"],
includes = ["include"],
deps = [
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfacesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "ControlFlowInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/ControlFlowInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/ControlFlowInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/ControlFlowInterfaces.td",
deps = [":ControlFlowInterfacesTdFiles"],
)
cc_library(
name = "ControlFlowInterfaces",
srcs = ["lib/Interfaces/ControlFlowInterfaces.cpp"],
hdrs = ["include/mlir/Interfaces/ControlFlowInterfaces.h"],
includes = ["include"],
deps = [
":ControlFlowInterfacesIncGen",
":IR",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "InferTypeOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/InferTypeOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/InferTypeOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/InferTypeOpInterface.td",
deps = [":InferTypeOpInterfaceTdFiles"],
)
cc_library(
name = "InferTypeOpInterface",
srcs = ["lib/Interfaces/InferTypeOpInterface.cpp"],
hdrs = ["include/mlir/Interfaces/InferTypeOpInterface.h"],
includes = ["include"],
deps = [
":IR",
":InferTypeOpInterfaceIncGen",
":Support",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "SideEffectInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/SideEffectInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/SideEffectInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/SideEffectInterfaces.td",
deps = [":SideEffectInterfacesTdFiles"],
)
cc_library(
name = "SideEffectInterfaces",
srcs = ["lib/Interfaces/SideEffectInterfaces.cpp"],
hdrs = ["include/mlir/Interfaces/SideEffectInterfaces.h"],
includes = ["include"],
deps = [
":IR",
":SideEffectInterfacesIncGen",
"//llvm:Support",
],
)
[mlir] Add an interface to allow operations to specify how they can be tiled. An interface to allow for tiling of operations is introduced. The tiling of the linalg.pad_tensor operation is modified to use this interface. Differential Revision: https://reviews.llvm.org/D108611
2021-08-30 15:55:30 -07:00
gentbl_cc_library(
name = "TilingInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/TilingInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/TilingInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/TilingInterface.td",
deps = [":TilingInterfaceTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "Analysis",
srcs = glob(
[
"lib/Analysis/*.cpp",
"lib/Analysis/*.h",
"lib/Analysis/*/*.cpp",
"lib/Analysis/*/*.h",
],
),
hdrs = glob(
[
"include/mlir/Analysis/*.h",
"include/mlir/Analysis/*/*.h",
],
),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":CallOpInterfaces",
":ControlFlowInterfaces",
":DataLayoutInterfaces",
":IR",
[bazel] Port 95aff23e2921
2022-06-03 20:43:02 +02:00
":InferIntRangeInterface",
":LoopLikeInterface",
[bazel] Updates for https://github.com/llvm/llvm-project/commit/496f9a7d8d4b61b6f5dfa21e2614043a2b27cf67
2023-01-09 10:27:15 +01:00
":Pass",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfaces",
":Support",
":ViewLikeInterface",
"//llvm:Support",
],
)
cc_library(
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
name = "TranslateLib",
bazel fix for TranslateLib Reviewed By: cota Differential Revision: https://reviews.llvm.org/D136519
2022-10-21 19:50:22 -07:00
srcs = glob([
"lib/Tools/mlir-translate/*.cpp",
]) + [
[mlir] Fix typo s/utilties/utilities/ (including in file name) Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D136887
2022-10-27 16:45:14 -04:00
"include/mlir/Tools/ParseUtilities.h",
bazel fix for TranslateLib Reviewed By: cota Differential Revision: https://reviews.llvm.org/D136519
2022-10-21 19:50:22 -07:00
],
[MLIR] Fix bazel BUILD file after Parser.h moved. See D121024, D121025, D121026. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D121094
2022-03-07 13:04:05 +01:00
hdrs = glob(["include/mlir/Tools/mlir-translate/*.h"]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":IR",
":Parser",
":Support",
"//llvm:Support",
],
)
cc_library(
name = "ToLLVMIRTranslation",
srcs = [
"lib/Target/LLVMIR/DebugTranslation.cpp",
"lib/Target/LLVMIR/DebugTranslation.h",
[mlir][Bazel] Update bazel BUILD after 889a11783ec
2023-02-03 09:42:34 +01:00
"lib/Target/LLVMIR/LoopAnnotationTranslation.cpp",
[bazel] Port b83caa32dc
2023-02-08 13:05:19 +01:00
"lib/Target/LLVMIR/LoopAnnotationTranslation.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"lib/Target/LLVMIR/ModuleTranslation.cpp",
Update Bazel build for 929189a499 Updates Bazel build files to match https://github.com/llvm/llvm-project/commit/929189a499 Differential Revision: https://reviews.llvm.org/D104864
2021-06-24 09:42:14 -07:00
"lib/Target/LLVMIR/TypeToLLVM.cpp",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
hdrs = [
"include/mlir/Target/LLVMIR/Export.h",
"include/mlir/Target/LLVMIR/LLVMTranslationInterface.h",
"include/mlir/Target/LLVMIR/ModuleTranslation.h",
Update Bazel build for 929189a499 Updates Bazel build files to match https://github.com/llvm/llvm-project/commit/929189a499 Differential Revision: https://reviews.llvm.org/D104864
2021-06-24 09:42:14 -07:00
"include/mlir/Target/LLVMIR/TypeToLLVM.h",
[bazel] Add missing includes for 5212058
2023-01-31 11:41:35 +01:00
"lib/Target/LLVMIR/AttrKindDetail.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
[mlir] Add partial support for translating data layout Add support for translating data layout specifications for integer and float types between MLIR and LLVM IR. This is a first step towards removing the string-based LLVM dialect data layout attribute on modules. The latter is still available and will remain so until the first-class MLIR modeling can fully replace it. Depends On D120739 Reviewed By: wsmoses Differential Revision: https://reviews.llvm.org/D120740
2022-03-01 18:21:38 +01:00
":DLTIDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":LLVMConversionIncGen",
":LLVMDialect",
":LLVMIRTransforms",
[mlir] Split intrinsics out of LLVMOps.td Move the operations that correspond to LLVM IR intrinsics in a separate .td file. This makes it easier to maintain the intrinsics and decreases the compile time of LLVMDialect.cpp by ~25%. Depends On D123310 Reviewed By: wsmoses, jacquesguan Differential Revision: https://reviews.llvm.org/D123315
2022-04-13 18:29:47 +02:00
":LLVMIntrinsicConversionIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpenMPDialect",
":Support",
"//llvm:Core",
"//llvm:FrontendOpenMP",
"//llvm:Support",
"//llvm:TransformUtils",
],
)
cc_library(
name = "AMXToLLVMIRTranslation",
srcs = glob(["lib/Target/LLVMIR/Dialect/AMX/*.cpp"]),
hdrs = glob(["include/mlir/Target/LLVMIR/Dialect/AMX/*.h"]),
includes = ["include"],
deps = [
":AMXConversionIncGen",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AMXDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "X86VectorToLLVMIRTranslation",
srcs = glob(["lib/Target/LLVMIR/Dialect/X86Vector/*.cpp"]),
hdrs = glob(["include/mlir/Target/LLVMIR/Dialect/X86Vector/*.h"]),
includes = ["include"],
deps = [
":IR",
":ToLLVMIRTranslation",
":X86VectorConversionIncGen",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":X86VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "ArmNeonToLLVMIRTranslation",
srcs = glob(["lib/Target/LLVMIR/Dialect/ArmNeon/*.cpp"]),
hdrs = glob(["include/mlir/Target/LLVMIR/Dialect/ArmNeon/*.h"]),
includes = ["include"],
deps = [
":ArmNeonConversionIncGen",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ArmNeonDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ArmNeonIncGen",
":IR",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Support",
],
)
[Bazel] Fix for 7a2fdc685f609730af29e5e969843e9eb71a184c
2023-06-14 23:11:53 +00:00
cc_library(
name = "ArmSMEToLLVMIRTranslation",
srcs = glob(["lib/Target/LLVMIR/Dialect/ArmSME/*.cpp"]),
hdrs = glob(["include/mlir/Target/LLVMIR/Dialect/ArmSME/*.h"]),
includes = ["include"],
deps = [
":ArmSMEConversionIncGen",
":ArmSMEDialect",
":IR",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "ArmSVEToLLVMIRTranslation",
srcs = glob(["lib/Target/LLVMIR/Dialect/ArmSVE/*.cpp"]),
hdrs = glob(["include/mlir/Target/LLVMIR/Dialect/ArmSVE/*.h"]),
includes = ["include"],
deps = [
":ArmSVEConversionIncGen",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ArmSVEDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "NVVMToLLVMIRTranslation",
srcs = glob(["lib/Target/LLVMIR/Dialect/NVVM/*.cpp"]),
hdrs = glob(["include/mlir/Target/LLVMIR/Dialect/NVVM/*.h"]),
includes = ["include"],
deps = [
[mlir] One more the bazel build for 3ac17449cf988bfcde804a4cc532420ed1657595
2022-10-28 15:08:41 +02:00
":DialectUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":NVVMConversionIncGen",
":NVVMDialect",
[mlir] One more the bazel build for 3ac17449cf988bfcde804a4cc532420ed1657595
2022-10-28 15:08:41 +02:00
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "ROCDLToLLVMIRTranslation",
srcs = glob(["lib/Target/LLVMIR/Dialect/ROCDL/*.cpp"]),
hdrs = glob(["include/mlir/Target/LLVMIR/Dialect/ROCDL/*.h"]),
includes = ["include"],
deps = [
":IR",
":ROCDLConversionIncGen",
":ROCDLDialect",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Support",
],
)
[mlir][Bazel] Adjust BUILD files to 0e9523efda8a4ad95ecb1d5b5e65e10bcc3711f5
2023-03-21 15:00:10 +01:00
cc_library(
name = "GPUToLLVMIRTranslation",
[bazel] Fix mlir buildifier issues
2023-03-28 13:11:47 -07:00
srcs = ["lib/Target/LLVMIR/Dialect/GPU/GPUToLLVMIRTranslation.cpp"],
hdrs = ["include/mlir/Target/LLVMIR/Dialect/GPU/GPUToLLVMIRTranslation.h"],
[mlir][Bazel] Adjust BUILD files to 0e9523efda8a4ad95ecb1d5b5e65e10bcc3711f5
2023-03-21 15:00:10 +01:00
includes = ["include"],
deps = [
":GPUDialect",
":IR",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "BuiltinToLLVMIRTranslation",
[bazel] Fix mlir buildifier issues
2023-03-28 13:11:47 -07:00
srcs = ["lib/Target/LLVMIR/Dialect/Builtin/BuiltinToLLVMIRTranslation.cpp"],
hdrs = ["include/mlir/Target/LLVMIR/Dialect/Builtin/BuiltinToLLVMIRTranslation.h"],
[mlir][Bazel] Adjust BUILD files to 0e9523efda8a4ad95ecb1d5b5e65e10bcc3711f5
2023-03-21 15:00:10 +01:00
includes = ["include"],
deps = [
":IR",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "LLVMToLLVMIRTranslation",
[mlir][Bazel] Port cf487cce6f64
2023-01-02 13:05:26 +01:00
srcs = ["lib/Target/LLVMIR/Dialect/LLVMIR/LLVMToLLVMIRTranslation.cpp"],
hdrs = ["include/mlir/Target/LLVMIR/Dialect/LLVMIR/LLVMToLLVMIRTranslation.h"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":IR",
":LLVMConversionIncGen",
":LLVMDialect",
[mlir] Split intrinsics out of LLVMOps.td Move the operations that correspond to LLVM IR intrinsics in a separate .td file. This makes it easier to maintain the intrinsics and decreases the compile time of LLVMDialect.cpp by ~25%. Depends On D123310 Reviewed By: wsmoses, jacquesguan Differential Revision: https://reviews.llvm.org/D123315
2022-04-13 18:29:47 +02:00
":LLVMIntrinsicConversionIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Support",
],
)
[mlir][openMP] Add bazel deps Differential Revision: https://reviews.llvm.org/D146455
2023-03-20 13:58:45 -07:00
cc_library(
name = "OpenMPCommon",
srcs = ["lib/Target/LLVMIR/Dialect/OpenMPCommon.cpp"],
hdrs = ["include/mlir/Target/LLVMIR/Dialect/OpenMPCommon.h"],
includes = ["include"],
deps = [
":IR",
":Support",
modify dir of dep in OpenMPCommon Fix dir error of deps in OpenMPCommon bazel build Differential Revision: https://reviews.llvm.org/D146461
2023-03-20 23:33:54 +01:00
"//llvm:Core",
"//llvm:FrontendOpenMP",
[mlir][openMP] Add bazel deps Differential Revision: https://reviews.llvm.org/D146455
2023-03-20 13:58:45 -07:00
],
)
[mlir][Bazel] Port cf487cce6f64
2023-01-02 13:05:26 +01:00
cc_library(
name = "LLVMIRToLLVMTranslation",
srcs = ["lib/Target/LLVMIR/Dialect/LLVMIR/LLVMIRToLLVMTranslation.cpp"],
hdrs = ["include/mlir/Target/LLVMIR/Dialect/LLVMIR/LLVMIRToLLVMTranslation.h"],
includes = ["include"],
deps = [
":FromLLVMIRTranslation",
":LLVMConversionIncGen",
":LLVMDialect",
":LLVMIntrinsicConversionIncGen",
":Support",
"//llvm:Core",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "OpenACCToLLVMIRTranslation",
srcs = glob(["lib/Target/LLVMIR/Dialect/OpenACC/*.cpp"]),
hdrs = glob(["include/mlir/Target/LLVMIR/Dialect/OpenACC/*.h"]),
includes = ["include"],
deps = [
":IR",
":LLVMDialect",
":OpenACCDialect",
":OpenACCToLLVM",
[mlir][openMP] Add bazel deps Differential Revision: https://reviews.llvm.org/D146455
2023-03-20 13:58:45 -07:00
":OpenMPCommon",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:FrontendOpenMP",
"//llvm:Support",
],
)
cc_library(
name = "OpenMPToLLVMIRTranslation",
srcs = glob(["lib/Target/LLVMIR/Dialect/OpenMP/*.cpp"]),
hdrs = glob(["include/mlir/Target/LLVMIR/Dialect/OpenMP/*.h"]),
includes = ["include"],
deps = [
":IR",
[mlir][bazel] Fix build.
2023-07-17 10:47:51 +02:00
":LLVMDialect",
[mlir][openMP] Add bazel deps Differential Revision: https://reviews.llvm.org/D146455
2023-03-20 13:58:45 -07:00
":OpenMPCommon",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpenMPDialect",
":Support",
":ToLLVMIRTranslation",
[bazel][mlir] fix for d3f9388ff
2023-04-26 13:47:00 -04:00
":Transforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
"//llvm:FrontendOpenMP",
"//llvm:Support",
],
)
cc_library(
name = "AllToLLVMIRTranslations",
hdrs = ["include/mlir/Target/LLVMIR/Dialect/All.h"],
includes = ["include"],
deps = [
":AMXToLLVMIRTranslation",
":ArmNeonToLLVMIRTranslation",
[Bazel] Fix for 7a2fdc685f609730af29e5e969843e9eb71a184c
2023-06-14 23:11:53 +00:00
":ArmSMEToLLVMIRTranslation",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ArmSVEToLLVMIRTranslation",
[mlir][Bazel] Adjust BUILD files to 0e9523efda8a4ad95ecb1d5b5e65e10bcc3711f5
2023-03-21 15:00:10 +01:00
":BuiltinToLLVMIRTranslation",
":GPUToLLVMIRTranslation",
[mlir][Bazel] Port cf487cce6f64
2023-01-02 13:05:26 +01:00
":LLVMIRToLLVMTranslation",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMToLLVMIRTranslation",
":NVVMToLLVMIRTranslation",
":OpenACCToLLVMIRTranslation",
":OpenMPToLLVMIRTranslation",
":ROCDLToLLVMIRTranslation",
":X86VectorToLLVMIRTranslation",
],
)
cc_library(
name = "ToLLVMIRTranslationRegistration",
srcs = ["lib/Target/LLVMIR/ConvertToLLVMIR.cpp"],
includes = ["include"],
deps = [
":AllToLLVMIRTranslations",
[mlir] Add partial support for translating data layout Add support for translating data layout specifications for integer and float types between MLIR and LLVM IR. This is a first step towards removing the string-based LLVM dialect data layout attribute on modules. The latter is still available and will remain so until the first-class MLIR modeling can fully replace it. Depends On D120739 Reviewed By: wsmoses Differential Revision: https://reviews.llvm.org/D120740
2022-03-01 18:21:38 +01:00
":DLTIDialect",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":ToLLVMIRTranslation",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":TranslateLib",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "FromLLVMIRTranslation",
srcs = [
[bazel] Port 9d2b84ef6232
2023-03-31 14:20:01 +02:00
"lib/Target/LLVMIR/DataLayoutImporter.cpp",
"lib/Target/LLVMIR/DataLayoutImporter.h",
[mlir][Bazel] Add DebugImporter sources to build target.
2022-11-18 10:18:37 +01:00
"lib/Target/LLVMIR/DebugImporter.cpp",
[bazel] Actually put Importer in the right library Fixes a81136c332
2023-02-08 13:08:26 +01:00
"lib/Target/LLVMIR/LoopAnnotationImporter.cpp",
"lib/Target/LLVMIR/LoopAnnotationImporter.h",
[mlir][Bazel] Add ModuleImport sources to FromLLVMIRTranslation target.
2022-12-19 12:04:47 +01:00
"lib/Target/LLVMIR/ModuleImport.cpp",
Update Bazel build for 929189a499 Updates Bazel build files to match https://github.com/llvm/llvm-project/commit/929189a499 Differential Revision: https://reviews.llvm.org/D104864
2021-06-24 09:42:14 -07:00
"lib/Target/LLVMIR/TypeFromLLVM.cpp",
],
hdrs = [
"include/mlir/Target/LLVMIR/Import.h",
[mlir][Bazel] Port cf487cce6f64
2023-01-02 13:05:26 +01:00
"include/mlir/Target/LLVMIR/LLVMImportInterface.h",
[mlir][Bazel] Add ModuleImport sources to FromLLVMIRTranslation target.
2022-12-19 12:04:47 +01:00
"include/mlir/Target/LLVMIR/ModuleImport.h",
Update Bazel build for 929189a499 Updates Bazel build files to match https://github.com/llvm/llvm-project/commit/929189a499 Differential Revision: https://reviews.llvm.org/D104864
2021-06-24 09:42:14 -07:00
"include/mlir/Target/LLVMIR/TypeFromLLVM.h",
[bazel] Add missing includes for 5212058
2023-01-31 11:41:35 +01:00
"lib/Target/LLVMIR/AttrKindDetail.h",
[mlir][Bazel] Add DebugImporter sources to build target.
2022-11-18 10:18:37 +01:00
"lib/Target/LLVMIR/DebugImporter.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
[mlir] Add partial support for translating data layout Add support for translating data layout specifications for integer and float types between MLIR and LLVM IR. This is a first step towards removing the string-based LLVM dialect data layout attribute on modules. The latter is still available and will remain so until the first-class MLIR modeling can fully replace it. Depends On D120739 Reviewed By: wsmoses Differential Revision: https://reviews.llvm.org/D120740
2022-03-01 18:21:38 +01:00
":DLTIDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":LLVMConversionIncGen",
":LLVMDialect",
[mlir][Bazel] Adjust BUILD.bazel file
2022-05-31 13:35:49 +02:00
":LLVMIntrinsicConversionIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":TranslateLib",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
[IR] Split out IR printing passes into IRPrinter This diff splits out (from LLVMCore) IR printing passes into IRPrinter. This structure is similar to what we already have for IRReader and enables us to avoid circular dependencies between LLVMCore and Analysis (this is a preparation for https://reviews.llvm.org/D137768). The legacy interface is left unchanged, once the legacy pass manager is removed (in the future) we will be able to clean it up further. The bazel build configuration has been updated as well. Test plan: 1/ Tested the following cmake configurations: static/dynamic linking * lld/gold * clang/gcc 2/ bazel build --config=generic_clang @llvm-project//... Differential revision: https://reviews.llvm.org/D138081
2022-11-18 01:47:56 +00:00
"//llvm:IRPrinter",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:IRReader",
"//llvm:Support",
],
)
[mlir][Bazel] Port cf487cce6f64
2023-01-02 13:05:26 +01:00
cc_library(
name = "FromLLVMIRTranslationRegistration",
srcs = [
"lib/Target/LLVMIR/ConvertFromLLVMIR.cpp",
],
deps = [
":AllToLLVMIRTranslations",
":DLTIDialect",
":FromLLVMIRTranslation",
":IR",
":TranslateLib",
"//llvm:Core",
"//llvm:IRReader",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "ExecutionEngine",
srcs = [
"include/mlir/ExecutionEngine/CRunnerUtils.h",
"lib/ExecutionEngine/ExecutionEngine.cpp",
],
hdrs = [
"include/mlir/ExecutionEngine/ExecutionEngine.h",
"include/mlir/ExecutionEngine/MemRefUtils.h",
],
includes = ["include"],
deps = [
":AllToLLVMIRTranslations",
":IR",
":LLVMDialect",
":Support",
":ToLLVMIRTranslation",
Move dependency llvm:AllTargetsAsmParsers from Translation to ExecutionEngine. - Fixes a minor issue in https://reviews.llvm.org/D114338, which seems incorrectly added the llvm:AllTargetsAsmParsers dependency to Translation in bazel build files. Differential Revision: https://reviews.llvm.org/D114471
2021-11-23 13:25:26 -08:00
"//llvm:AllTargetsAsmParsers",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:BitWriter",
"//llvm:Core",
"//llvm:ExecutionEngine",
"//llvm:MC",
"//llvm:OrcJIT",
"//llvm:Support",
"//llvm:Target", # fixdeps: keep
[bazel] Fix build and make all target respect layering_check Tested with `bazelisk-linux-amd64 query //... + @llvm-project//... | xargs bazelisk-linux-amd64 test --config=ci --features=layering_check`
2023-02-10 11:40:14 +00:00
"//llvm:TargetParser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:TransformUtils",
"//llvm:X86CodeGen", # fixdeps: keep
"//llvm:X86Disassembler", # fixdeps: keep
],
)
cc_library(
name = "ExecutionEngineUtils",
srcs = ["lib/ExecutionEngine/OptUtils.cpp"],
hdrs = ["include/mlir/ExecutionEngine/OptUtils.h"],
includes = ["include"],
deps = [
"//llvm:Analysis",
"//llvm:Core",
"//llvm:Coroutines",
"//llvm:IPO",
Fix BUILD dependency for ExecutionEngineUtils Differential Revision: https://reviews.llvm.org/D123570
2022-04-12 04:47:42 +00:00
"//llvm:Passes",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
"//llvm:Target",
"//llvm:common_transforms",
],
)
[mlir] fix mismerge in bazel
2023-04-13 14:58:11 +02:00
cc_library(
name = "PluginsLib",
srcs = [
"lib/Tools/Plugins/DialectPlugin.cpp",
"lib/Tools/Plugins/PassPlugin.cpp",
],
hdrs = [
"include/mlir/Tools/Plugins/DialectPlugin.h",
"include/mlir/Tools/Plugins/PassPlugin.h",
],
includes = ["include"],
deps = [
":IR",
":Pass",
":Support",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "MlirOptLib",
[Bazel] NFC: Move ParseUtilities.h from 'hdrs' to 'srcs'. This is slightly cleaner.
2022-09-28 08:29:00 +02:00
srcs = [
[mlir] Fix typo s/utilties/utilities/ (including in file name) Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D136887
2022-10-27 16:45:14 -04:00
"include/mlir/Tools/ParseUtilities.h",
[Bazel] NFC: Move ParseUtilities.h from 'hdrs' to 'srcs'. This is slightly cleaner.
2022-09-28 08:29:00 +02:00
"lib/Tools/mlir-opt/MlirOptMain.cpp",
[Bazel] Fix after 4b27825ba36d55779561c0a2c3c2f89f52d81303.
2022-09-28 08:17:04 +02:00
],
[Bazel] NFC: Move ParseUtilities.h from 'hdrs' to 'srcs'. This is slightly cleaner.
2022-09-28 08:29:00 +02:00
hdrs = ["include/mlir/Tools/mlir-opt/MlirOptMain.h"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir] Fix compile errors with bytecode support
2022-08-22 17:59:51 +02:00
":BytecodeReader",
[mlir][Bazel] Fix bazel build Add fixes for https://reviews.llvm.org/D131747
2022-08-22 12:06:47 +02:00
":BytecodeWriter",
[bazel] Fix build after 28d04c5
2023-03-06 17:24:26 +01:00
":Debug",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[bazel] Port for 8ac8c922fb3f15706f5cb1db2cc655d30b095766
2023-04-23 23:54:29 +02:00
":IRDLDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Parser",
":Pass",
[mlir] fix mismerge in bazel
2023-04-13 14:58:11 +02:00
":PluginsLib",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
"//llvm:Support",
],
)
cc_library(
name = "AllTranslations",
hdrs = ["include/mlir/InitAllTranslations.h"],
deps = [
[mlir][Bazel] Port cf487cce6f64
2023-01-02 13:05:26 +01:00
":FromLLVMIRTranslationRegistration",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SPIRVTranslateRegistration",
[Bazel] Update for Cpp emitter (2f0750dd2e) Update the Bazel build for https://github.com/llvm/llvm-project/commit/2f0750dd2e. Differential Revision: https://reviews.llvm.org/D109171
2021-09-02 10:04:48 -07:00
":TargetCpp",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ToLLVMIRTranslationRegistration",
],
)
cc_library(
name = "MlirTranslateMain",
srcs = ["tools/mlir-translate/mlir-translate.cpp"],
deps = [
":AllPassesAndDialects",
":AllTranslations",
":Support",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":TranslateLib",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
cc_binary(
name = "mlir-translate",
deps = [":MlirTranslateMain"],
)
cc_library(
name = "AllPassesAndDialects",
hdrs = [
"include/mlir/InitAllDialects.h",
"include/mlir/InitAllPasses.h",
],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AMDGPUDialect",
[mlir][linalg] Add lowering of named ops on complex numbers This lets linalg.dot and friends lower to a complex muladd using ops from the complex dialect. Differential Revision: https://reviews.llvm.org/D125461
2022-05-12 13:35:27 +02:00
":AMDGPUToROCDL",
[mlir][Bazel] Update BUILD.bazel file for cc4703745ffa398b66f985b483cb8b61eb2ed425 Differential Revision: https://reviews.llvm.org/D149823
2023-05-04 10:09:53 +02:00
":AMDGPUTransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AMXDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AMXTransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AffinePassIncGen",
":AffineToStandard",
[mlir] bazel for D137997
2022-11-21 10:20:19 +01:00
":AffineTransformOps",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AffineTransforms",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithToLLVM",
":ArithToSPIRV",
":ArithTransforms",
[mlir][arith] Add ValueBoundsOpInterface impls These ops are useful for unit testing. (They do not fold/canonicalize with affine.apply etc.) Differential Revision: https://reviews.llvm.org/D145696
2023-04-06 10:46:48 +09:00
":ArithValueBoundsOpInterfaceImpl",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ArmNeonDialect",
[Bazel] Another fix for 7a2fdc685f609730af29e5e969843e9eb71a184c
2023-06-14 23:28:50 +00:00
":ArmSMEDialect",
[bazel][mlir] Port 65305aeab99ad8ea09dd85e28a41c657152a08fb
2023-06-16 13:20:32 +02:00
":ArmSMETransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ArmSVEDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ArmSVETransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AsyncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AsyncPassIncGen",
":AsyncToLLVM",
":AsyncTransforms",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationDialect",
[mlir][bufferization] Add OneShotBufferize transform op This commit allows for One-Shot Bufferize to be used through the transform dialect. No op handle is currently returned for the bufferized IR. Differential Revision: https://reviews.llvm.org/D125098
2022-06-09 13:00:08 +02:00
":BufferizationTransformOps",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":BufferizationTransforms",
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ComplexDialect",
":ComplexToLLVM",
[mlir][complex] Add pow/sqrt/tanh ops and lowering to libm Lowering through libm gives us a baseline version, even though it's not going to be particularly fast. This is similar to what we do for some math dialect ops. Differential Revision: https://reviews.llvm.org/D125550
2022-05-13 17:12:11 +02:00
":ComplexToLibm",
[mlir][spirv][complex] Convert complex ops to SPIR-V ops This commit adds conversion from complex construction and extraction ops to SPIR-V. Other arithemtic ops can be done via ComplexToStandard patterns. Reviewed By: kuhar Differential Revision: https://reviews.llvm.org/D147193
2023-03-30 08:49:57 -07:00
":ComplexToSPIRV",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPasses",
":DLTIDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":EmitCDialect",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
":FuncToSPIRV",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
":FuncTransforms",
":FuncTransformsPassIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":GPUPassIncGen",
":GPUToGPURuntimeTransforms",
":GPUToNVVMTransforms",
":GPUToROCDLTransforms",
":GPUToSPIRV",
":GPUToVulkanTransforms",
[mlir][transform] Create GPU transform dialect This revision adds GPU transform dialect. It also introduce a prefix such as "transform.gpu" for all ops related to this dialect. MLIR already had two GPU transform op in linalg. This revision moves these ops into GPUTransformOps. The Ops are as follows: `transform.structured.map_nested_foreach_thread_to_gpu_blocks` -> `transform.gpu.map_foreach_to_blocks` This op selects the outermost (toplevel) foreach_thread and parallelize across GPU blocks. It can also generate `gpu_launch`. `transform.structured.map_nested_foreach_thread_to_gpu_threads` -> `transform.gpu.map_nested_foreach_to_threads` This op parallelizes nested foreach_thread that are inside `gpu_launch` across GPU threads. It doesn't add new functionality, but there are some minor refactoring of the code. Reviewed By: ftynse Differential Revision: https://reviews.llvm.org/D134800
2022-10-03 09:56:42 +02:00
":GPUTransformOps",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUTransforms",
":IR",
Fix bazel overlay after "[mlir] Introduce IRDL dialect" Reviewed By: dblaikie Differential Revision: https://reviews.llvm.org/D147583
2023-04-05 01:07:25 +00:00
":IRDLDialect",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
":IndexDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
":LLVMIRTransforms",
":LLVMPassIncGen",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LinalgPassIncGen",
":LinalgToStandard",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
":LinalgTransformOps",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LinalgTransforms",
[mlir] Introduce ml_program dialect. Differential Revision: https://reviews.llvm.org/D120203
2022-04-13 20:16:04 -07:00
":MLProgramDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MathDialect",
[mlir][Bazel] Fix bazel build.
2022-08-26 09:04:50 +02:00
":MathToFuncs",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MathToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MathToLibm",
[mlir] NFC: split Math to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107093
2021-07-29 16:21:43 -04:00
":MathToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MathTransforms",
":MemRefDialect",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MemRefToLLVM",
[mlir] NFC: split MemRef to SPIR-V conversion into their own files Reviewed By: hanchung Differential Revision: https://reviews.llvm.org/D107094
2021-07-29 16:22:46 -04:00
":MemRefToSPIRV",
[bazel] Port 3f050f6ac4626c9bf99bfdf5e4d7162ba7ad5cdc
2022-09-29 00:12:46 +02:00
":MemRefTransformOps",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MemRefTransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":NVGPUDialect",
[mlir][nvgpu] shared memory access optimization pass This change adds a transformation and pass to the NvGPU dialect that attempts to optimize reads/writes from a memref representing GPU shared memory in order to avoid bank conflicts. Given a value representing a shared memory memref, it traverses all reads/writes within the parent op and, subject to suitable conditions, rewrites all last dimension index values such that element locations in the final (col) dimension are given by `newColIdx = col % vecSize + perm[row](col/vecSize,row)` where `perm` is a permutation function indexed by `row` and `vecSize` is the vector access size in elements (currently assumes 128bit vectorized accesses, but this can be made a parameter). This specific transformation can help optimize typical distributed & vectorized accesses common to loading matrix multiplication operands to/from shared memory. Differential Revision: https://reviews.llvm.org/D127457
2022-06-07 10:51:27 -06:00
":NVGPUPassIncGen",
[mlir] Update bazel file after adding nvgpu to nvvm conversion
2022-04-15 04:07:46 +00:00
":NVGPUToNVVM",
Revert "Revert "[mlir][Transform] Add support for mma.sync m16n8k16 f16 rewrite." and "[mlir][Transform] Introduce nvgpu transform extensions"" This reverts commit 6506692fe619ef8a1f7c6ea829d9a9eceb31622d. Differential Revision: https://reviews.llvm.org/D153845
2023-06-27 09:02:31 +00:00
":NVGPUTransformOps",
[mlir][Bazel] Add missing dependencies after 564713c47175d9f61fe8e18e750fb7f7b486d533
2023-07-03 13:16:28 +02:00
":NVGPUTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":NVVMDialect",
":OpenACCDialect",
":OpenMPDialect",
":OpenMPToLLVM",
":PDLDialect",
":PDLInterpDialect",
":PDLToPDLInterp",
":QuantOps",
":ROCDLDialect",
[mlir] Factor type reconciliation out of Standard-to-LLVM conversion Conversion to the LLVM dialect is being refactored to be more progressive and is now performed as a series of independent passes converting different dialects. These passes may produce `unrealized_conversion_cast` operations that represent pending conversions between built-in and LLVM dialect types. Historically, a more monolithic Standard-to-LLVM conversion pass did not need these casts as all operations were converted in one shot. Previous refactorings have led to the requirement of running the Standard-to-LLVM conversion pass to clean up `unrealized_conversion_cast`s even though the IR had no standard operations in it. The pass must have been also run the last among all to-LLVM passes, in contradiction with the partial conversion logic. Additionally, the way it was set up could produce invalid operations by removing casts between LLVM and built-in types even when the consumer did not accept the uncasted type, or could lead to cryptic conversion errors (recursive application of the rewrite pattern on `unrealized_conversion_cast` as a means to indicate failure to eliminate casts). In fact, the need to eliminate A->B->A `unrealized_conversion_cast`s is not specific to to-LLVM conversions and can be factored out into a separate type reconciliation pass, which is achieved in this commit. While the cast operation itself has a folder pattern, it is insufficient in most conversion passes as the folder only applies to the second cast. Without complex legality setup in the conversion target, the conversion infra will either consider the cast operations valid and not fold them (a separate canonicalization would be necessary to trigger the folding), or consider the first cast invalid upon generation and stop with error. The pattern provided by the reconciliation pass applies to the first cast operation instead. Furthermore, having a separate pass makes it clear when `unrealized_conversion_cast`s could not have been eliminated since it is the only reason why this pass can fail. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D109507
2021-09-09 16:06:10 +02:00
":ReconcileUnrealizedCasts",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SCFDialect",
":SCFPassIncGen",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":SCFToControlFlow",
":SCFToGPU",
[mlir] Introduce Transform ops for loops Introduce transform ops for "for" loops, in particular for peeling, software pipelining and unrolling, along with a couple of "IR navigation" ops. These ops are intended to be generalized to different kinds of loops when possible and therefore use the "loop" prefix. They currently live in the SCF dialect as there is no clear place to put transform ops that may span across several dialects, this decision is postponed until the ops actually need to handle non-SCF loops. Additionally refactor some common utilities for transform ops into trait or interface methods, and change the loop pipelining to be a returning pattern. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D127300
2022-06-09 11:10:32 +02:00
":SCFTransformOps",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SCFTransforms",
":SDBM",
":SPIRVDialect",
":SPIRVPassIncGen",
":SPIRVToLLVM",
":SPIRVTransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ShapeDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ShapeToStandard",
":ShapeTransforms",
":ShapeTransformsPassIncGen",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":SparseTensorDialect",
[mlir][sparse] Adding standard pipeline for tests. Addresses https://bugs.llvm.org/show_bug.cgi?id=52409 aka https://github.com/llvm/llvm-project/issues/51751 Reviewed By: aartbik, mehdi_amini Differential Revision: https://reviews.llvm.org/D117919
2022-01-26 16:44:32 -08:00
":SparseTensorPipelines",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SparseTensorTransforms",
":TensorDialect",
[mlir] Move `linalg.tensor_expand/collapse_shape` to TensorDialect. RFC: https://llvm.discourse.group/t/rfc-reshape-ops-restructuring/3310 linalg.fill gets a canonicalizer, because `FoldFillWithTensorReshape` cannot be moved to tensorops (it uses linalg::FillOp inside). Before it was listed as a canonicalization pattern for the reshape operations, now it became a canonicalization for FillOp. Differential Revision: https://reviews.llvm.org/D115502
2021-12-10 12:03:47 +01:00
":TensorInferTypeOpInterfaceImpl",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
":TensorTilingInterfaceImpl",
[mlir][transform] Add TrackingListener This change makes it possible to use a greedy pattern rewrite as part of a transform op, even if the transform op does not invalidate the target handle (in particular transform ops without `FunctionalStyleTransformOpTrait`) and the targeted op is not isolated from above. The listener API allows us to track replacements of ops with values, but not ops with ops. Therefore, the TrackingListener is conservative: If an op is replaced with values that all have the same defining op and the defining op is of the same type as the original op, it is safe to assume that the op was replaced with an equivalent op. Otherwise, the op mapping is dropped. When this is not good enough, transforms can track values instead or provide a custom `findReplacementOp` function. Differential Revision: https://reviews.llvm.org/D147039
2023-03-29 11:02:26 +02:00
":TensorTransformOps",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TensorTransforms",
":TosaDialect",
":TosaToLinalg",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
":TransformDialect",
[mlir] Use-after-free checker for the Transform dialect The Transform dialect uses the side effect modeling mechanism to record the effects of the transform ops on the mapping between Transform IR values and Payload IR ops. Introduce a checker pass that warns if a Transform IR value is used after it has been freed (consumed). This pass is mostly intended as a debugging aid in addition to the verification/assertion mechanisms in the transform interpreter. It reports all potential use-after-free situations. The implementation makes a series of simplifying assumptions to be simple and conservative. A more advanced implementation would rely on the data flow-like analysis associated with a side-effect resource rather than a value, which is currently not supported by the analysis infrastructure. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D126381
2022-05-25 15:57:22 +02:00
":TransformDialectTransforms",
[mlir] move PDL-related transform ops into an extension The initial bring-up of the Transform dialect relied on PDL to provide the default handle type (`!pdl.operation`) and the matching capability. Both are now provided natively by the Transform dialect removing the reason to have a hard dependency on the PDL dialect and its interpreter. Move PDL-related transform operations into a separate extension. This requires us to introduce a dialect state extension mechanism into the Transform dialect so it no longer needs to know about PDL constraint functions that may be injected by extensions similarly to operations and types. This mechanism will be reused to connect pattern application drivers and the Transform dialect. This completes the restructuring of the Transform dialect to remove overrilance on PDL. Note to downstreams: flow that are using `!pdl.operation` with Transform dialect operations will now require `transform::PDLExtension` to be applied to the transform dialect in order to provide the transform handle type interface for `!pdl.operation`. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D151104
2023-05-22 14:36:58 +00:00
":TransformPDLExtension",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
":TransformsPassIncGen",
[bazel] Fix dependency path
2023-07-20 17:05:28 +02:00
":UBDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":VectorToLLVM",
":VectorToSCF",
":VectorToSPIRV",
[mlir][Vector] Vector transform skeleton Differential Revision: https://reviews.llvm.org/D134722
2022-06-21 06:52:01 -07:00
":VectorTransformOps",
[mlir][bufferize] Add vector-bufferize pass and remove obsolete patterns from Linalg Bufferize Differential Revision: https://reviews.llvm.org/D119444
2022-02-15 21:16:50 +09:00
":VectorTransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":X86VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":X86VectorTransforms",
],
)
cc_binary(
name = "mlir-lsp-server",
srcs = ["tools/mlir-lsp-server/mlir-lsp-server.cpp"],
includes = ["include"],
deps = [
":AllPassesAndDialects",
[mlir:vscode] Fix bazel build after commit f7b8a70
2022-09-04 11:37:06 +02:00
":BytecodeWriter",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":MlirLspServerLib",
[mlir:vscode] Fix bazel build after commit f7b8a70
2022-09-04 11:37:06 +02:00
":Parser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
cc_binary(
name = "mlir-opt",
srcs = ["tools/mlir-opt/mlir-opt.cpp"],
local_defines = ["MLIR_INCLUDE_TESTS"],
deps = [
[bazel][NFC] Add Dialect/Func/Extensions library and deps Added in D120368
2023-06-09 17:04:41 -07:00
":AllExtensions",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AllPassesAndDialects",
":Analysis",
":IR",
":MlirOptLib",
":OpenMPDialect",
":Pass",
":QuantOps",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":SCFToGPU",
[mlir][Bazel] Add missing dependencies.
2023-03-20 09:28:25 +01:00
":SerializeToCubin",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":Transforms",
"//llvm:AllTargetsCodeGens",
"//llvm:Support",
"//mlir/test:TestAffine",
"//mlir/test:TestAnalysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"//mlir/test:TestArith",
[mlir][bufferization] Make `TensorCopyInsertionPass` a test pass TensorCopyInsertion should not have been exposed as a pass. This was a flaw in the original design. It is a preparation step for bufferization and certain transforms (that would otherwise be legal) are illegal between TensorCopyInsertion and actual rewrite to MemRef ops. Therefore, even if broken down as two separate steps internally, they should be exposed as a single pass. This change affects the sparse compiler, which uses `TensorCopyInsertionPass`. A new `SparsificationAndBufferizationPass` is added to replace all passes in the sparse tensor pipeline from `TensorCopyInsertionPass` until the actual bufferization (rewrite to memref/non-tensor). It is generally unsafe to run arbitrary passes in-between, in particular passes that hoist tensor ops out of loops or change SSA use-def chains along tensor ops. Differential Revision: https://reviews.llvm.org/D138915
2022-12-02 11:46:14 +01:00
"//mlir/test:TestBufferization",
[mlir][cf] Print message in cf.assert to LLVM lowering The assert message was previously ignored. The lowered IR now calls `puts` it in case of a failed assertion. Differential Revision: https://reviews.llvm.org/D138647
2022-12-15 17:37:49 +01:00
"//mlir/test:TestControlFlow",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//mlir/test:TestDLTI",
"//mlir/test:TestDialect",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
"//mlir/test:TestFunc",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
"//mlir/test:TestFuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//mlir/test:TestGPU",
"//mlir/test:TestIR",
[mlir][Test] Add a test pass to act as a sink towards LLVM conversion This allows writing simple e2e tests where we can check for the proper materialization of specific LLVM IR (e.g. `llvm.intr.fmuladd`). Differential Revision: https://reviews.llvm.org/D138776
2022-11-25 10:22:32 -08:00
"//mlir/test:TestLLVM",
[bazel] Add a missing dependency after 786cbb09ed
2022-12-02 16:56:05 +01:00
"//mlir/test:TestLinalg",
[bazel] Port 81a79ee446
2023-02-10 19:02:01 +01:00
"//mlir/test:TestLoopLikeInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//mlir/test:TestMath",
[mlir][Bazel] Add TestMemRef target. Needed after D118285 Differential Revision: https://reviews.llvm.org/D118330
2022-01-27 09:44:06 +01:00
"//mlir/test:TestMemRef",
[mlir][Bazel] Update bazel build after 14d79afeae63d78de9483f750fafaba13c7ae2dc Differential Revision: https://reviews.llvm.org/D130965
2022-08-02 08:53:31 +02:00
"//mlir/test:TestNVGPU",
Fix bazel build for mlir-opt Tested locally with `bazel build --repo_env=CC=clang --repo_env=BAZEL_CXXOPTS="-std=c++17" @llvm-project//mlir:mlir-opt`. Reviewed By: jyknight Differential Revision: https://reviews.llvm.org/D146992
2023-03-27 11:44:15 -07:00
"//mlir/test:TestOneToNTypeConversion",
[mlir][bazel] Add suport for PDLL tests. Differential Revision: https://reviews.llvm.org/D124515
2022-04-27 12:11:03 +02:00
"//mlir/test:TestPDLL",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//mlir/test:TestPass",
"//mlir/test:TestReducer",
"//mlir/test:TestRewrite",
"//mlir/test:TestSCF",
"//mlir/test:TestSPIRV",
"//mlir/test:TestShapeDialect",
[mlir][tensor] Add a pattern to split tensor.pad ops This commit adds a pattern to wrap a tensor.pad op with an scf.if op to separate the cases where we don't need padding (all pad sizes are actually zeros) and where we indeed need padding. This pattern is meant to handle padding inside tiled loops. Under such cases the padding sizes typically depend on the loop induction variables. Splitting them would allow treating perfect tiles and edge tiles separately. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D117018
2022-02-16 11:26:47 -05:00
"//mlir/test:TestTensor",
[bazel] Port ba8424a251d7
2022-09-19 21:20:12 +02:00
"//mlir/test:TestTestDynDialect",
[mlir][TilingInterface] Add pattern to tile using TilingInterface and implement TilingInterface for Linalg ops. This patch adds support for tiling operations that implement the TilingInterface. - It separates the loop constructs that are used to iterate over tile from the implementation of the tiling itself. For example, the use of destructive updates is more related to use of scf.for for iterating over tiles that are tensors. - To test the transformation, TilingInterface is implemented for LinalgOps. The separation of the looping constructs used from the implementation of tile code generation greatly simplifies the latter. - The implementation of TilingInterface for LinalgOp is kept as an external model for now till this approach can be fully flushed out to replace the existing tiling + fusion approaches in Linalg. Differential Revision: https://reviews.llvm.org/D127133
2022-06-13 19:56:32 +00:00
"//mlir/test:TestTilingInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//mlir/test:TestTosaDialect",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
"//mlir/test:TestTransformDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//mlir/test:TestTransforms",
"//mlir/test:TestTypeDialect",
"//mlir/test:TestVector",
[mlir][spirv][vector] Add pattern to convert reduction to SPIR-V dot prod This converts a specific form of `vector.reduction` to SPIR-V integer dot product ops. Add a new test pass to excercise this outside of the main vector to spirv conversion pass. Reviewed By: antiagainst Differential Revision: https://reviews.llvm.org/D145760
2023-03-10 13:54:16 -05:00
"//mlir/test:TestVectorToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
cc_library(
name = "MlirJitRunner",
srcs = ["lib/ExecutionEngine/JitRunner.cpp"],
[mlir] fix bazel file Reviewed By: Peiming Differential Revision: https://reviews.llvm.org/D135104
2022-10-03 13:54:58 -07:00
hdrs = [
"include/mlir/ExecutionEngine/JitRunner.h",
[mlir] Fix typo s/utilties/utilities/ (including in file name) Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D136887
2022-10-27 16:45:14 -04:00
"include/mlir/Tools/ParseUtilities.h",
[mlir] fix bazel file Reviewed By: Peiming Differential Revision: https://reviews.llvm.org/D135104
2022-10-03 13:54:58 -07:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":AllPassesAndDialects",
":ExecutionEngine",
":ExecutionEngineUtils",
":IR",
":LLVMDialect",
":LLVMToLLVMIRTranslation",
":OpenACCToLLVMIRTranslation",
":OpenMPToLLVMIRTranslation",
":Parser",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":SCFToControlFlow",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
"//llvm:Core",
"//llvm:OrcJIT",
"//llvm:Support",
],
)
cc_library(
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
name = "mlir_async_runtime_api",
hdrs = ["include/mlir/ExecutionEngine/AsyncRuntime.h"],
includes = ["include"],
)
cc_library(
name = "_mlir_async_runtime",
srcs = ["lib/ExecutionEngine/AsyncRuntime.cpp"],
copts = ["-Dmlir_async_runtime_EXPORTS"],
deps = [
":mlir_async_runtime_api",
"//llvm:Support",
],
)
[MLIR] NFC: expand comment in bazel BUILD file ... and remove unnecessry `exports_files()`.
2022-09-07 21:57:02 +02:00
# Indirection to avoid 'libmlir_async_runtime.so' filename clash.
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
alias(
name = "mlir_async_runtime",
actual = "_mlir_async_runtime",
)
cc_binary(
name = "libmlir_async_runtime.so",
linkshared = True,
linkstatic = False,
deps = [":mlir_async_runtime"],
)
[mlir][sparse] refactoring SparseTensorUtils: (1 of 4) file-splitting Previously, the SparseTensorUtils.cpp library contained a C++ core implementation, but hid it in an anonymous namespace and only exposed a C-API for accessing it. Now we are factoring out that C++ core into a standalone C++ library so that it can be used directly by downstream clients (per request of one such client). This refactoring has been decomposed into a stack of differentials in order to simplify the code review process, however the full stack of changes should be considered together. * (this): Part 1: split one file into several * D133830: Part 2: Reorder chunks within files * D133831: Part 3: General code cleanup * D133833: Part 4: Update documentation This part aims to make no changes other than the 1:N file splitting, and things which are forced to accompany that change. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D133462
2022-09-29 11:40:40 -07:00
cc_library(
name = "_mlir_float16_utils",
srcs = ["lib/ExecutionEngine/Float16bits.cpp"],
hdrs = ["include/mlir/ExecutionEngine/Float16bits.h"],
copts = ["-Dmlir_float16_utils_EXPORTS"],
includes = ["include"],
)
# Indirection to avoid 'libmlir_float16_utils.so' filename clash.
alias(
name = "mlir_float16_utils",
actual = "_mlir_float16_utils",
)
cc_binary(
name = "libmlir_float16_utils.so",
linkshared = True,
linkstatic = False,
deps = [":mlir_float16_utils"],
)
# Unlike mlir_float16_utils, mlir_c_runner_utils, etc, we do *not* make
[mlir][sparse] Renaming mlir_sparse_tensor_utils library to SparseTensorRuntime The "mlir_xxx_utils" naming scheme is reserved/intended for shared libraries, whereas this library must be static due to issues of linking DLLs on Windows. So we rename the library to avoid any potential confusion. In addition we also rename the ExecutionEngine/SparseTensorUtils.{h,cpp} files to match the new library name. Reviewed By: aartbik, stella.stamenova Differential Revision: https://reviews.llvm.org/D135613
2022-10-10 13:41:58 -07:00
# this a shared library: because on the CMake side, doing so causes issues
# when building on Windows. In particular, various functions take/return
# `std::vector` but that class is not designed with dllimport/dllexport
# pragma, therefore it cannot be safely/correctly used across DLL boundaries.
# Consequently, we avoid using the "mlir_xxx_utils" naming scheme,
# since that is reserved/intended for shared libraries only.
[mlir][sparse] refactoring SparseTensorUtils: (1 of 4) file-splitting Previously, the SparseTensorUtils.cpp library contained a C++ core implementation, but hid it in an anonymous namespace and only exposed a C-API for accessing it. Now we are factoring out that C++ core into a standalone C++ library so that it can be used directly by downstream clients (per request of one such client). This refactoring has been decomposed into a stack of differentials in order to simplify the code review process, however the full stack of changes should be considered together. * (this): Part 1: split one file into several * D133830: Part 2: Reorder chunks within files * D133831: Part 3: General code cleanup * D133833: Part 4: Update documentation This part aims to make no changes other than the 1:N file splitting, and things which are forced to accompany that change. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D133462
2022-09-29 11:40:40 -07:00
cc_library(
[mlir][sparse] Renaming mlir_sparse_tensor_utils library to SparseTensorRuntime The "mlir_xxx_utils" naming scheme is reserved/intended for shared libraries, whereas this library must be static due to issues of linking DLLs on Windows. So we rename the library to avoid any potential confusion. In addition we also rename the ExecutionEngine/SparseTensorUtils.{h,cpp} files to match the new library name. Reviewed By: aartbik, stella.stamenova Differential Revision: https://reviews.llvm.org/D135613
2022-10-10 13:41:58 -07:00
name = "SparseTensorRuntime",
[mlir][sparse] refactoring SparseTensorUtils: (1 of 4) file-splitting Previously, the SparseTensorUtils.cpp library contained a C++ core implementation, but hid it in an anonymous namespace and only exposed a C-API for accessing it. Now we are factoring out that C++ core into a standalone C++ library so that it can be used directly by downstream clients (per request of one such client). This refactoring has been decomposed into a stack of differentials in order to simplify the code review process, however the full stack of changes should be considered together. * (this): Part 1: split one file into several * D133830: Part 2: Reorder chunks within files * D133831: Part 3: General code cleanup * D133833: Part 4: Update documentation This part aims to make no changes other than the 1:N file splitting, and things which are forced to accompany that change. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D133462
2022-09-29 11:40:40 -07:00
srcs = [
"lib/ExecutionEngine/SparseTensor/File.cpp",
"lib/ExecutionEngine/SparseTensor/NNZ.cpp",
[mlir][sparse] Making way for SparseTensorRuntime to support non-permutations Systematically updates the SparseTensorRuntime to properly distinguish tensor-dimensions from storage-levels (and their associated ranks, shapes, sizes, indices, etc). With a few exceptions which are noted in the code, this ensures the runtime has all the **semantic** changes necessary to support non-permutations. (Whereas **operationally**, since we're still using `std::vector<uing64_t>` to represent the mappings, there's no way to pass in any interesting non-permutations. Changing the representation to `std::function` will be done in a separate differential.) Depends On D137680 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D137681
2022-11-09 13:33:01 -08:00
"lib/ExecutionEngine/SparseTensor/PermutationRef.cpp",
[mlir][sparse] refactoring SparseTensorUtils: (1 of 4) file-splitting Previously, the SparseTensorUtils.cpp library contained a C++ core implementation, but hid it in an anonymous namespace and only exposed a C-API for accessing it. Now we are factoring out that C++ core into a standalone C++ library so that it can be used directly by downstream clients (per request of one such client). This refactoring has been decomposed into a stack of differentials in order to simplify the code review process, however the full stack of changes should be considered together. * (this): Part 1: split one file into several * D133830: Part 2: Reorder chunks within files * D133831: Part 3: General code cleanup * D133833: Part 4: Update documentation This part aims to make no changes other than the 1:N file splitting, and things which are forced to accompany that change. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D133462
2022-09-29 11:40:40 -07:00
"lib/ExecutionEngine/SparseTensor/Storage.cpp",
],
hdrs = [
[mlir][sparse] Adding safe comparison functions to MLIRSparseTensorRuntime. Different platforms use different signedness for `StridedMemRefType::sizes` and `std::vector::size_type`, and this has been causing a lot of portability issues re [-Wsign-compare] warnings. These new functions ensure that we need never worry about those signedness warnings ever again. Also merging CheckedMul.h into ArithmeticUtils.h Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D138149
2022-11-16 12:59:54 -08:00
"include/mlir/ExecutionEngine/SparseTensor/ArithmeticUtils.h",
[mlir][sparse] Making way for SparseTensorRuntime to support non-permutations Systematically updates the SparseTensorRuntime to properly distinguish tensor-dimensions from storage-levels (and their associated ranks, shapes, sizes, indices, etc). With a few exceptions which are noted in the code, this ensures the runtime has all the **semantic** changes necessary to support non-permutations. (Whereas **operationally**, since we're still using `std::vector<uing64_t>` to represent the mappings, there's no way to pass in any interesting non-permutations. Changing the representation to `std::function` will be done in a separate differential.) Depends On D137680 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D137681
2022-11-09 13:33:01 -08:00
"include/mlir/ExecutionEngine/SparseTensor/Attributes.h",
[mlir][sparse] refactoring SparseTensorUtils: (1 of 4) file-splitting Previously, the SparseTensorUtils.cpp library contained a C++ core implementation, but hid it in an anonymous namespace and only exposed a C-API for accessing it. Now we are factoring out that C++ core into a standalone C++ library so that it can be used directly by downstream clients (per request of one such client). This refactoring has been decomposed into a stack of differentials in order to simplify the code review process, however the full stack of changes should be considered together. * (this): Part 1: split one file into several * D133830: Part 2: Reorder chunks within files * D133831: Part 3: General code cleanup * D133833: Part 4: Update documentation This part aims to make no changes other than the 1:N file splitting, and things which are forced to accompany that change. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D133462
2022-09-29 11:40:40 -07:00
"include/mlir/ExecutionEngine/SparseTensor/COO.h",
"include/mlir/ExecutionEngine/SparseTensor/ErrorHandling.h",
"include/mlir/ExecutionEngine/SparseTensor/File.h",
[mlir][sparse] Making way for SparseTensorRuntime to support non-permutations Systematically updates the SparseTensorRuntime to properly distinguish tensor-dimensions from storage-levels (and their associated ranks, shapes, sizes, indices, etc). With a few exceptions which are noted in the code, this ensures the runtime has all the **semantic** changes necessary to support non-permutations. (Whereas **operationally**, since we're still using `std::vector<uing64_t>` to represent the mappings, there's no way to pass in any interesting non-permutations. Changing the representation to `std::function` will be done in a separate differential.) Depends On D137680 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D137681
2022-11-09 13:33:01 -08:00
"include/mlir/ExecutionEngine/SparseTensor/PermutationRef.h",
[mlir][sparse] refactoring SparseTensorUtils: (1 of 4) file-splitting Previously, the SparseTensorUtils.cpp library contained a C++ core implementation, but hid it in an anonymous namespace and only exposed a C-API for accessing it. Now we are factoring out that C++ core into a standalone C++ library so that it can be used directly by downstream clients (per request of one such client). This refactoring has been decomposed into a stack of differentials in order to simplify the code review process, however the full stack of changes should be considered together. * (this): Part 1: split one file into several * D133830: Part 2: Reorder chunks within files * D133831: Part 3: General code cleanup * D133833: Part 4: Update documentation This part aims to make no changes other than the 1:N file splitting, and things which are forced to accompany that change. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D133462
2022-09-29 11:40:40 -07:00
"include/mlir/ExecutionEngine/SparseTensor/Storage.h",
],
includes = ["include"],
[mlir][sparse] Factoring out SparseTensorEnums library This differential splits the SparseTensorEnums library out from the SparseTensorRuntime library. The actual moving of files will be handled in the next differential. Depends On D135996 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136002
2022-10-17 18:13:05 -07:00
deps = [
":SparseTensorEnums",
":mlir_float16_utils",
],
[mlir][sparse] refactoring SparseTensorUtils: (1 of 4) file-splitting Previously, the SparseTensorUtils.cpp library contained a C++ core implementation, but hid it in an anonymous namespace and only exposed a C-API for accessing it. Now we are factoring out that C++ core into a standalone C++ library so that it can be used directly by downstream clients (per request of one such client). This refactoring has been decomposed into a stack of differentials in order to simplify the code review process, however the full stack of changes should be considered together. * (this): Part 1: split one file into several * D133830: Part 2: Reorder chunks within files * D133831: Part 3: General code cleanup * D133833: Part 4: Update documentation This part aims to make no changes other than the 1:N file splitting, and things which are forced to accompany that change. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D133462
2022-09-29 11:40:40 -07:00
)
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
cc_library(
name = "_mlir_c_runner_utils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = [
"lib/ExecutionEngine/CRunnerUtils.cpp",
[mlir][sparse] Renaming mlir_sparse_tensor_utils library to SparseTensorRuntime The "mlir_xxx_utils" naming scheme is reserved/intended for shared libraries, whereas this library must be static due to issues of linking DLLs on Windows. So we rename the library to avoid any potential confusion. In addition we also rename the ExecutionEngine/SparseTensorUtils.{h,cpp} files to match the new library name. Reviewed By: aartbik, stella.stamenova Differential Revision: https://reviews.llvm.org/D135613
2022-10-10 13:41:58 -07:00
"lib/ExecutionEngine/SparseTensorRuntime.cpp",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
[mlir][sparse] Factoring magic numbers into a header Addresses https://bugs.llvm.org/show_bug.cgi?id=52303 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D112962
2021-11-05 15:15:39 -07:00
hdrs = [
"include/mlir/ExecutionEngine/CRunnerUtils.h",
[mlir] Add msan memory unpoisoning macros to mlir ExecutionEngine Adding annotations on as-needed bases, currently only for memrefCopy, but in general all C API functions that take pointers to memory allocated/initialized inside the jit-compiled code must be annotated, to be able to run with msan. Reviewed By: mehdi_amini Differential Revision: https://reviews.llvm.org/D123557
2022-04-11 17:28:51 -07:00
"include/mlir/ExecutionEngine/Msan.h",
[mlir][sparse] Renaming mlir_sparse_tensor_utils library to SparseTensorRuntime The "mlir_xxx_utils" naming scheme is reserved/intended for shared libraries, whereas this library must be static due to issues of linking DLLs on Windows. So we rename the library to avoid any potential confusion. In addition we also rename the ExecutionEngine/SparseTensorUtils.{h,cpp} files to match the new library name. Reviewed By: aartbik, stella.stamenova Differential Revision: https://reviews.llvm.org/D135613
2022-10-10 13:41:58 -07:00
"include/mlir/ExecutionEngine/SparseTensorRuntime.h",
[mlir][sparse] Factoring magic numbers into a header Addresses https://bugs.llvm.org/show_bug.cgi?id=52303 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D112962
2021-11-05 15:15:39 -07:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
[mlir][sparse] Factoring out SparseTensorEnums library This differential splits the SparseTensorEnums library out from the SparseTensorRuntime library. The actual moving of files will be handled in the next differential. Depends On D135996 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136002
2022-10-17 18:13:05 -07:00
deps = [
":SparseTensorEnums",
":SparseTensorRuntime",
":mlir_float16_utils",
[Bazel][mlir] Avoid ODR violation introduced in 7ab749c. This change also prepares for 9119325 to land again. Adds `mlir_c_runner_utils_hdrs` and `mlir_runner_utils_hdrs` targets which do not depend on `//llvm::Support`. These can be used by other 'runner.so' targets if they are loaded along with the 'runner_utils.so' without calling `__mlir_execution_engine_init()` twice.
2023-06-22 07:50:38 +02:00
"//llvm:Support",
[mlir][sparse] Factoring out SparseTensorEnums library This differential splits the SparseTensorEnums library out from the SparseTensorRuntime library. The actual moving of files will be handled in the next differential. Depends On D135996 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136002
2022-10-17 18:13:05 -07:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
[MLIR] NFC: expand comment in bazel BUILD file ... and remove unnecessry `exports_files()`.
2022-09-07 21:57:02 +02:00
# Indirection to avoid 'libmlir_c_runner_utils.so' filename clash.
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
alias(
name = "mlir_c_runner_utils",
actual = "_mlir_c_runner_utils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
[Bazel][mlir] Avoid ODR violation introduced in 7ab749c. This change also prepares for 9119325 to land again. Adds `mlir_c_runner_utils_hdrs` and `mlir_runner_utils_hdrs` targets which do not depend on `//llvm::Support`. These can be used by other 'runner.so' targets if they are loaded along with the 'runner_utils.so' without calling `__mlir_execution_engine_init()` twice.
2023-06-22 07:50:38 +02:00
cc_headers_only(
name = "mlir_c_runner_utils_hdrs",
src = ":mlir_c_runner_utils",
)
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
cc_binary(
name = "libmlir_c_runner_utils.so",
linkshared = True,
linkstatic = False,
deps = [":mlir_c_runner_utils"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
cc_library(
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
name = "_mlir_runner_utils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = ["lib/ExecutionEngine/RunnerUtils.cpp"],
hdrs = ["include/mlir/ExecutionEngine/RunnerUtils.h"],
includes = ["include"],
[Bazel][mlir] Fix after bba2b656110209a3d9863b92c060082479b06ab1
2023-06-20 23:00:38 +02:00
deps = [
":mlir_c_runner_utils",
Revert "[Bazel][mlir] Fix ODR violation introduced in 7ab749c." This reverts commit e83c8c36005f0068841e628612e9e5bce7e2ac9e. Depending only on the support header files is not sufficient.
2023-06-21 14:29:44 +02:00
"//llvm:Support",
[Bazel][mlir] Fix after bba2b656110209a3d9863b92c060082479b06ab1
2023-06-20 23:00:38 +02:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
[MLIR] NFC: expand comment in bazel BUILD file ... and remove unnecessry `exports_files()`.
2022-09-07 21:57:02 +02:00
# Indirection to avoid 'libmlir_runner_utils.so' filename clash.
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
alias(
name = "mlir_runner_utils",
actual = "_mlir_runner_utils",
)
[Bazel][mlir] Avoid ODR violation introduced in 7ab749c. This change also prepares for 9119325 to land again. Adds `mlir_c_runner_utils_hdrs` and `mlir_runner_utils_hdrs` targets which do not depend on `//llvm::Support`. These can be used by other 'runner.so' targets if they are loaded along with the 'runner_utils.so' without calling `__mlir_execution_engine_init()` twice.
2023-06-22 07:50:38 +02:00
cc_headers_only(
name = "mlir_runner_utils_hdrs",
src = ":mlir_runner_utils",
)
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
cc_binary(
name = "libmlir_runner_utils.so",
linkshared = True,
linkstatic = False,
deps = [":mlir_runner_utils"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_binary(
name = "mlir-cpu-runner",
srcs = ["tools/mlir-cpu-runner/mlir-cpu-runner.cpp"],
deps = [
":AllToLLVMIRTranslations",
[mlir][Bazel] Adjust BUILD files to 0e9523efda8a4ad95ecb1d5b5e65e10bcc3711f5
2023-03-21 15:00:10 +01:00
":BuiltinToLLVMIRTranslation",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ExecutionEngineUtils",
":IR",
":LLVMDialect",
":LLVMToLLVMIRTranslation",
":MlirJitRunner",
":OpenACCToLLVMIRTranslation",
":OpenMPToLLVMIRTranslation",
":ToLLVMIRTranslation",
"//llvm:AsmParser",
"//llvm:Support",
"//llvm:X86AsmParser",
],
)
# This target provides the headers from LLVM's Support target without any of
# the symbols. In particular, it does not contain the static registration code
# which may be executed by at most one shared library loaded by ORCJit. Direct
# dependencies need to avoid requiring symbols from LLVMSupport by adding
# copts = ["-DLLVM_DISABLE_ABI_BREAKING_CHECKS_ENFORCING=1"].
#
# Bazel links the dependencies' object files instead of the archives, which
# means that symbols are linked in even if none are used. The LLVM cmake build
# on the other hand links archives (or shared libraries, depending on
# BUILD_SHARED_LIBS), skipping them if none of the symbols are used.
# See also https://reviews.llvm.org/D95613.
cc_headers_only(
name = "LLVMSupportHeaders",
src = "//llvm:Support",
)
cc_library(
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
name = "_mlir_cuda_runtime",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = ["lib/ExecutionEngine/CudaRuntimeWrappers.cpp"],
# Prevent needing EnableABIBreakingChecks symbol from LLVMSupport.
copts = ["-DLLVM_DISABLE_ABI_BREAKING_CHECKS_ENFORCING=1"],
[mlir][sparse][gpu] add sm8.0+ tensor core 2:4 sparsity support Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D151775
2023-06-06 23:12:45 +00:00
# Here:
# MLIR_ENABLE_CUDA_CUSPARSE : enables cuSPARSE
# MLIR_ENABLE_CUDA_CUSPARSELT : enables cuSPARSElt
local_defines = ["MLIR_ENABLE_CUDA_CUSPARSE"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
deps = [
":LLVMSupportHeaders",
[Bazel][mlir] Avoid ODR violation introduced in 7ab749c. This change also prepares for 9119325 to land again. Adds `mlir_c_runner_utils_hdrs` and `mlir_runner_utils_hdrs` targets which do not depend on `//llvm::Support`. These can be used by other 'runner.so' targets if they are loaded along with the 'runner_utils.so' without calling `__mlir_execution_engine_init()` twice.
2023-06-22 07:50:38 +02:00
":mlir_c_runner_utils_hdrs",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"@cuda//:cuda_headers",
[mlir][gpu][sparse] add gpu ops for sparse matrix computations This revision extends the GPU dialect with ops that can be lowered to host-oriented sparse matrix library calls (in this case cuSparse focused although the ops could be generalized to support more GPUs in principle). This will allow the "sparse compiler pipeline" to accelerate sparse operations (see follow up revisions with examples of this). For some background; https://discourse.llvm.org/t/sparse-compiler-and-gpu-code-generation/69786/2 Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D150152
2023-05-11 13:56:34 -07:00
"@cuda//:cusparse_static",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"@cuda//:libcuda",
],
)
[MLIR] NFC: expand comment in bazel BUILD file ... and remove unnecessry `exports_files()`.
2022-09-07 21:57:02 +02:00
# Indirection to avoid 'libmlir_cuda_runtime.so' filename clash.
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
alias(
name = "mlir_cuda_runtime",
actual = "_mlir_cuda_runtime",
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
)
cc_binary(
name = "libmlir_cuda_runtime.so",
linkshared = True,
linkstatic = False,
tags = [
"manual", # External dependency
"nobuildkite", # TODO(gcmn): Add support for this target
],
deps = [":mlir_cuda_runtime"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "VulkanRuntime",
srcs = ["tools/mlir-vulkan-runner/VulkanRuntime.cpp"],
hdrs = ["tools/mlir-vulkan-runner/VulkanRuntime.h"],
tags = [
"manual", # External dependency
],
deps = [
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":Pass",
":SPIRVDialect",
":SideEffectInterfaces",
":Support",
"//llvm:Support",
"@vulkan_headers",
"@vulkan_sdk//:sdk",
],
)
cc_binary(
[MLIR] Switch lit tests to %mlir_lib_dir and %mlir_src_dir replacements. The old replacements will be removed soon: - `%linalg_test_lib_dir` - `%cuda_wrapper_library_dir` - `%spirv_wrapper_library_dir` - `%vulkan_wrapper_library_dir` - `%mlir_runner_utils_dir` - `%mlir_integration_test_dir` Reviewed By: herhut Differential Revision: https://reviews.llvm.org/D133270
2022-09-04 10:18:31 +02:00
name = "libvulkan-runtime-wrappers.so",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = ["tools/mlir-vulkan-runner/vulkan-runtime-wrappers.cpp"],
linkshared = True,
[Bazel][mlir] Avoid ODR violation introduced in 7ab749c. This change also prepares for 9119325 to land again. Adds `mlir_c_runner_utils_hdrs` and `mlir_runner_utils_hdrs` targets which do not depend on `//llvm::Support`. These can be used by other 'runner.so' targets if they are loaded along with the 'runner_utils.so' without calling `__mlir_execution_engine_init()` twice.
2023-06-22 07:50:38 +02:00
linkstatic = False,
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
tags = [
"manual", # External dependency
],
[Bazel][mlir] Avoid ODR violation introduced in 7ab749c. This change also prepares for 9119325 to land again. Adds `mlir_c_runner_utils_hdrs` and `mlir_runner_utils_hdrs` targets which do not depend on `//llvm::Support`. These can be used by other 'runner.so' targets if they are loaded along with the 'runner_utils.so' without calling `__mlir_execution_engine_init()` twice.
2023-06-22 07:50:38 +02:00
deps = [":VulkanRuntime"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
cc_binary(
name = "mlir-vulkan-runner",
srcs = ["tools/mlir-vulkan-runner/mlir-vulkan-runner.cpp"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][Bazel] Adjust BUILD files to 0e9523efda8a4ad95ecb1d5b5e65e10bcc3711f5
2023-03-21 15:00:10 +01:00
":BuiltinToLLVMIRTranslation",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ExecutionEngineUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
":FuncToSPIRV",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":GPUToSPIRV",
":GPUToVulkanTransforms",
":GPUTransforms",
[Bazel] Fixes for b5d847b1b95750d0af40cfc8c71a8fec50bb8613 and 6412a13539ab2914eed8e1df83c399b9a16e3408
2021-07-07 16:50:23 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
[mlir] replace 'emit_c_wrappers' func->llvm conversion option with a pass The 'emit_c_wrappers' option in the FuncToLLVM conversion requests C interface wrappers to be emitted for every builtin function in the module. While this has been useful to bootstrap the interface, it is problematic in the longer term as it may unintentionally affect the functions that should retain their existing interface, e.g., libm functions obtained by lowering math operations (see D126964 for an example). Since D77314, we have a finer-grain control over interface generation via an attribute that avoids the problem entirely. Remove the 'emit_c_wrappers' option. Introduce the '-llvm-request-c-wrappers' pass that can be run in any pipeline that needs blanket emission of functions to annotate all builtin functions with the attribute before performing the usual lowering that accounts for the attribute. Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D127952
2022-06-16 12:53:49 +02:00
":LLVMIRTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMToLLVMIRTranslation",
":MemRefDialect",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MemRefToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MemRefTransforms",
":MlirJitRunner",
":Pass",
[mlir] Factor type reconciliation out of Standard-to-LLVM conversion Conversion to the LLVM dialect is being refactored to be more progressive and is now performed as a series of independent passes converting different dialects. These passes may produce `unrealized_conversion_cast` operations that represent pending conversions between built-in and LLVM dialect types. Historically, a more monolithic Standard-to-LLVM conversion pass did not need these casts as all operations were converted in one shot. Previous refactorings have led to the requirement of running the Standard-to-LLVM conversion pass to clean up `unrealized_conversion_cast`s even though the IR had no standard operations in it. The pass must have been also run the last among all to-LLVM passes, in contradiction with the partial conversion logic. Additionally, the way it was set up could produce invalid operations by removing casts between LLVM and built-in types even when the consumer did not accept the uncasted type, or could lead to cryptic conversion errors (recursive application of the rewrite pattern on `unrealized_conversion_cast` as a means to indicate failure to eliminate casts). In fact, the need to eliminate A->B->A `unrealized_conversion_cast`s is not specific to to-LLVM conversions and can be factored out into a separate type reconciliation pass, which is achieved in this commit. While the cast operation itself has a folder pattern, it is insufficient in most conversion passes as the folder only applies to the second cast. Without complex legality setup in the conversion target, the conversion infra will either consider the cast operations valid and not fold them (a separate canonicalization would be necessary to trigger the folding), or consider the first cast invalid upon generation and stop with error. The pattern provided by the reconciliation pass applies to the first cast operation instead. Furthermore, having a separate pass makes it clear when `unrealized_conversion_cast`s could not have been eliminated since it is the only reason why this pass can fail. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D109507
2021-09-09 16:06:10 +02:00
":ReconcileUnrealizedCasts",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SPIRVDialect",
":SPIRVTransforms",
":ToLLVMIRTranslation",
[mlir][spirv] Fix extended umul expansion for WebGPU Fix an off-by-one error in extended umul extension for WebGPU. Revert to the long multiplication algorithm originally added to wide integer emulation, which was deleted in D139776. It is much easier to see why it is correct. Add runtime tests based on the mlir-vulkan-runner. These run both with and without umul extension. Issue: https://github.com/llvm/llvm-project/issues/59563 Reviewed By: antiagainst Differential Revision: https://reviews.llvm.org/D141085
2023-01-05 18:37:49 -05:00
":VectorDialect",
":VectorToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
cc_binary(
name = "mlir-spirv-cpu-runner",
srcs = ["tools/mlir-spirv-cpu-runner/mlir-spirv-cpu-runner.cpp"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][Bazel] Adjust BUILD files to 0e9523efda8a4ad95ecb1d5b5e65e10bcc3711f5
2023-03-21 15:00:10 +01:00
":BuiltinToLLVMIRTranslation",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ExecutionEngineUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":GPUToSPIRV",
":GPUTransforms",
":LLVMDialect",
":LLVMToLLVMIRTranslation",
":MemRefDialect",
":MlirJitRunner",
":Pass",
":SPIRVConversion",
":SPIRVDialect",
":SPIRVToLLVM",
":SPIRVTransforms",
":ToLLVMIRTranslation",
"//llvm:Core",
"//llvm:Linker",
"//llvm:Support",
],
)
cc_library(
name = "TableGen",
srcs = glob(["lib/TableGen/*.cpp"]),
hdrs = glob(["include/mlir/TableGen/*.h"]),
includes = ["include"],
deps = [
":Support",
"//llvm:Support",
"//llvm:TableGen",
[bazel] add TargetParser
2022-12-20 20:09:00 +01:00
"//llvm:TargetParser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
cc_library(
name = "MlirTableGenMain",
[bazel] Port 1c5a50e32815a49a41d79ff529ca8611ee49c5c8
2022-08-06 16:08:48 +02:00
srcs = glob(["lib/Tools/mlir-tblgen/*.cpp"]),
hdrs = glob(["include/mlir/Tools/mlir-tblgen/*.h"]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":TableGen",
"//llvm:Support",
"//llvm:TableGen",
[bazel] add TargetParser
2022-12-20 20:09:00 +01:00
"//llvm:TargetParser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:config",
],
)
cc_binary(
name = "mlir-tblgen",
srcs = glob([
"tools/mlir-tblgen/*.h",
"tools/mlir-tblgen/*.cpp",
]),
deps = [
":MlirTableGenMain",
":Support",
":TableGen",
Try to fix CodeGenTypes issues in mlir
2023-05-03 00:42:24 +09:00
"//llvm:CodeGenTypes",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
"//llvm:TableGen",
[bazel] add TargetParser
2022-12-20 20:09:00 +01:00
"//llvm:TargetParser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:config",
],
)
cc_binary(
name = "mlir-linalg-ods-yaml-gen",
srcs = [
"tools/mlir-linalg-ods-gen/mlir-linalg-ods-yaml-gen.cpp",
],
deps = [
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
":AsmParser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":Support",
"//llvm:Support",
"//llvm:TableGen",
"//llvm:config",
],
)
## OpenACC dialect
# TODO(gcmn): This is sticking td files in a cc_library
gentbl_cc_library(
name = "AccCommonGen",
includes = ["/llvm/include"],
strip_include_prefix = "include",
tbl_outs = [
(
[mlir] Replace StrEnumAttr -> EnumAttr in core dialects Removes uses of `StrEnumAttr` in core dialects Reviewed By: mehdi_amini, rriddle Differential Revision: https://reviews.llvm.org/D117514
2022-01-18 16:53:55 +00:00
[
"-gen-directive-decl",
"-directives-dialect=OpenACC",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/Dialect/OpenACC/AccCommon.td",
),
],
tblgen = ":mlir-tblgen",
td_file = "//llvm:include/llvm/Frontend/OpenACC/ACC.td",
deps = ["//llvm:acc_td_files"],
)
td_library(
name = "OpenAccOpsTdFiles",
srcs = [
"include/mlir/Dialect/OpenACC/AccCommon.td",
bazel patch for OpenACC changes Reviewed By: razvanlupusoru, anlunx Differential Revision: https://reviews.llvm.org/D148678
2023-04-18 18:18:34 -07:00
"include/mlir/Dialect/OpenACC/OpenACCBase.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/Dialect/OpenACC/OpenACCOps.td",
bazel patch for OpenACC changes Reviewed By: razvanlupusoru, anlunx Differential Revision: https://reviews.llvm.org/D148678
2023-04-18 18:18:34 -07:00
"include/mlir/Dialect/OpenACC/OpenACCOpsTypes.td",
"include/mlir/Dialect/OpenACC/OpenACCTypeInterfaces.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
gentbl_cc_library(
name = "OpenACCOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=acc",
],
"include/mlir/Dialect/OpenACC/OpenACCOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=acc",
],
"include/mlir/Dialect/OpenACC/OpenACCOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-decls"],
"include/mlir/Dialect/OpenACC/OpenACCOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/OpenACC/OpenACCOps.cpp.inc",
),
(
["-gen-enum-decls"],
"include/mlir/Dialect/OpenACC/OpenACCOpsEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/OpenACC/OpenACCOpsEnums.cpp.inc",
),
[mlir] Replace StrEnumAttr -> EnumAttr in core dialects Removes uses of `StrEnumAttr` in core dialects Reviewed By: mehdi_amini, rriddle Differential Revision: https://reviews.llvm.org/D117514
2022-01-18 16:53:55 +00:00
(
[
"-gen-attrdef-decls",
"-attrdefs-dialect=acc",
],
"include/mlir/Dialect/OpenACC/OpenACCOpsAttributes.h.inc",
),
(
[
"-gen-attrdef-defs",
"-attrdefs-dialect=acc",
],
"include/mlir/Dialect/OpenACC/OpenACCOpsAttributes.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-doc"],
"g3doc/Dialects/OpenACC/OpenACCOps.md",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/OpenACC/OpenACCOps.td",
bazel patch for OpenACC changes Reviewed By: razvanlupusoru, anlunx Differential Revision: https://reviews.llvm.org/D148678
2023-04-18 18:18:34 -07:00
deps = [
":BuiltinDialectTdFiles",
[bazel] Fix build after 0c4d7d14e94d
2023-05-16 18:27:12 +02:00
":ControlFlowInterfacesTdFiles",
bazel patch for OpenACC changes Reviewed By: razvanlupusoru, anlunx Differential Revision: https://reviews.llvm.org/D148678
2023-04-18 18:18:34 -07:00
":OpenAccOpsTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "OpenACCTypesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"--gen-typedef-decls",
"-typedefs-dialect=acc",
],
"include/mlir/Dialect/OpenACC/OpenACCOpsTypes.h.inc",
),
(
[
"--gen-typedef-defs",
"-typedefs-dialect=acc",
],
"include/mlir/Dialect/OpenACC/OpenACCOpsTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/OpenACC/OpenACCOpsTypes.td",
deps = [":OpenAccOpsTdFiles"],
)
gentbl_cc_library(
name = "OpenACCTypeInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-type-interface-decls"],
"include/mlir/Dialect/OpenACC/OpenACCTypeInterfaces.h.inc",
),
(
["-gen-type-interface-defs"],
"include/mlir/Dialect/OpenACC/OpenACCTypeInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/OpenACC/OpenACCTypeInterfaces.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [":OpenAccOpsTdFiles"],
)
cc_library(
name = "OpenACCDialect",
srcs = glob(
[
"lib/Dialect/OpenACC/IR/*.cpp",
"lib/Dialect/OpenACC/IR/*.h",
],
),
hdrs = glob([
"include/mlir/Dialect/OpenACC/*.h",
]),
includes = ["include"],
deps = [
[bazel] Fix build Differential Revision: https://reviews.llvm.org/D150734
2023-05-16 15:40:04 -07:00
":ControlFlowInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
bazel patch for OpenACC changes Reviewed By: razvanlupusoru, anlunx Differential Revision: https://reviews.llvm.org/D148678
2023-04-18 18:18:34 -07:00
":LLVMDialect",
":MemRefDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpenACCOpsIncGen",
bazel patch for OpenACC changes Reviewed By: razvanlupusoru, anlunx Differential Revision: https://reviews.llvm.org/D148678
2023-04-18 18:18:34 -07:00
":OpenACCTypeInterfacesIncGen",
":OpenACCTypesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
"//llvm:Support",
],
)
## OpenMP dialect
# TODO(gcmn): This is sticking td files in a cc_library
gentbl_cc_library(
name = "OmpCommonTdGen",
includes = ["/llvm/include"],
strip_include_prefix = "include",
tbl_outs = [
(
[mlir] Replace StrEnumAttr -> EnumAttr in core dialects Removes uses of `StrEnumAttr` in core dialects Reviewed By: mehdi_amini, rriddle Differential Revision: https://reviews.llvm.org/D117514
2022-01-18 16:53:55 +00:00
[
"-gen-directive-decl",
"-directives-dialect=OpenMP",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/Dialect/OpenMP/OmpCommon.td",
),
],
tblgen = ":mlir-tblgen",
td_file = "//llvm:include/llvm/Frontend/OpenMP/OMP.td",
deps = [
":OpBaseTdFiles",
"//llvm:omp_td_files",
],
)
td_library(
name = "OpenMPOpsTdFiles",
srcs = [
"include/mlir/Dialect/OpenMP/OmpCommon.td",
"include/mlir/Dialect/OpenMP/OpenMPOps.td",
[Bazel] Update config for 3b01cf9286 Updates the Bazel config for changes from https://github.com/llvm/llvm-project/commit/3b01cf9286 by adding configuration for the new OpenMPOpsInterfaces tablegn target. Differential Revision: https://reviews.llvm.org/D111347
2021-10-07 12:45:53 -07:00
"include/mlir/Dialect/OpenMP/OpenMPOpsInterfaces.td",
[mlir][bzl] Update for 1a92dbcfa88a857bf735c897125d9a2f9be53e9e and cab44c515c63db876a7f626fdc36be4a9157ebad
2022-07-07 17:36:28 -07:00
"include/mlir/Dialect/OpenMP/OpenMPTypeInterfaces.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
deps = [
":LLVMOpsTdFiles",
":OpBaseTdFiles",
],
)
gentbl_cc_library(
name = "OpenMPOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/OpenMP/OpenMPOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/OpenMP/OpenMPOps.cpp.inc",
),
(
["-gen-enum-decls"],
"include/mlir/Dialect/OpenMP/OpenMPOpsEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/OpenMP/OpenMPOpsEnums.cpp.inc",
),
(
[
"-gen-dialect-decls",
"-dialect=omp",
],
"include/mlir/Dialect/OpenMP/OpenMPOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=omp",
],
"include/mlir/Dialect/OpenMP/OpenMPOpsDialect.cpp.inc",
),
[mlir] Replace StrEnumAttr -> EnumAttr in core dialects Removes uses of `StrEnumAttr` in core dialects Reviewed By: mehdi_amini, rriddle Differential Revision: https://reviews.llvm.org/D117514
2022-01-18 16:53:55 +00:00
(
[
"-gen-attrdef-decls",
"-attrdefs-dialect=omp",
],
"include/mlir/Dialect/OpenMP/OpenMPOpsAttributes.h.inc",
),
(
[
"-gen-attrdef-defs",
"-attrdefs-dialect=omp",
],
"include/mlir/Dialect/OpenMP/OpenMPOpsAttributes.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-doc"],
"g3doc/Dialects/OpenMP/OpenMPOps.md",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/OpenMP/OpenMPOps.td",
deps = [":OpenMPOpsTdFiles"],
)
[mlir][bzl] Update for 1a92dbcfa88a857bf735c897125d9a2f9be53e9e and cab44c515c63db876a7f626fdc36be4a9157ebad
2022-07-07 17:36:28 -07:00
gentbl_cc_library(
name = "OpenMPTypeInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-type-interface-decls"],
"include/mlir/Dialect/OpenMP/OpenMPTypeInterfaces.h.inc",
),
(
["-gen-type-interface-defs"],
"include/mlir/Dialect/OpenMP/OpenMPTypeInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/OpenMP/OpenMPTypeInterfaces.td",
deps = [":OpenMPOpsTdFiles"],
)
[Bazel] Update config for 3b01cf9286 Updates the Bazel config for changes from https://github.com/llvm/llvm-project/commit/3b01cf9286 by adding configuration for the new OpenMPOpsInterfaces tablegn target. Differential Revision: https://reviews.llvm.org/D111347
2021-10-07 12:45:53 -07:00
gentbl_cc_library(
name = "OpenMPInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Dialect/OpenMP/OpenMPOpsInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Dialect/OpenMP/OpenMPOpsInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/OpenMP/OpenMPOpsInterfaces.td",
deps = [":OpenMPOpsTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "OpenMPDialect",
srcs = glob(
[
"lib/Dialect/OpenMP/IR/*.cpp",
"lib/Dialect/OpenMP/IR/*.h",
],
),
[mlir] [bazel] Make OpenMPInterfaces.h a textual header. This header can't be built standalone. Making it textual will prevent blaze from attempting to do so.
2023-03-29 09:40:39 +02:00
hdrs = glob(
[
"include/mlir/Dialect/OpenMP/*.h",
],
exclude = ["include/mlir/Dialect/OpenMP/OpenMPInterfaces.h"],
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
[mlir] [bazel] Make OpenMPInterfaces.h a textual header. This header can't be built standalone. Making it textual will prevent blaze from attempting to do so.
2023-03-29 09:40:39 +02:00
textual_hdrs = [
"include/mlir/Dialect/OpenMP/OpenMPInterfaces.h",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [
":ControlFlowInterfaces",
[bazel] Port 231a6959025b3101942701b214c64762b55222c6
2023-05-22 17:11:25 +02:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":LLVMDialect",
[Bazel] Update config for 3b01cf9286 Updates the Bazel config for changes from https://github.com/llvm/llvm-project/commit/3b01cf9286 by adding configuration for the new OpenMPOpsInterfaces tablegn target. Differential Revision: https://reviews.llvm.org/D111347
2021-10-07 12:45:53 -07:00
":OpenMPInterfacesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpenMPOpsIncGen",
[mlir][bzl] Update for 1a92dbcfa88a857bf735c897125d9a2f9be53e9e and cab44c515c63db876a7f626fdc36be4a9157ebad
2022-07-07 17:36:28 -07:00
":OpenMPTypeInterfacesIncGen",
bazel: adapt for https://github.com/llvm/llvm-project/commit/a4699a43e42615281c96599d20977cabf10bfb9c
2023-01-23 15:38:46 +00:00
"//llvm:FrontendOpenMP",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
cc_library(
name = "OpenACCToSCF",
srcs = glob([
"lib/Conversion/OpenACCToSCF/*.cpp",
"lib/Conversion/OpenACCToSCF/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/OpenACCToSCF/*.h",
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":OpenACCDialect",
bazel patch for OpenACC changes Reviewed By: razvanlupusoru, anlunx Differential Revision: https://reviews.llvm.org/D148678
2023-04-18 18:18:34 -07:00
":OpenACCOpsIncGen",
":OpenACCTypesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":SCFDialect",
":Transforms",
],
)
cc_library(
name = "OpenACCToLLVM",
srcs = glob([
"lib/Conversion/OpenACCToLLVM/*.cpp",
"lib/Conversion/OpenACCToLLVM/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/OpenACCToLLVM/*.h",
]),
includes = ["include"],
deps = [
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
":OpenACCDialect",
":Pass",
":Transforms",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "OpenMPToLLVM",
srcs = glob([
"lib/Conversion/OpenMPToLLVM/*.cpp",
"lib/Conversion/OpenMPToLLVM/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/OpenMPToLLVM/*.h",
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithToLLVM",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
":ControlFlowToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":MemRefToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpenMPDialect",
":Pass",
":Transforms",
"//llvm:Core",
"//llvm:Support",
],
)
td_library(
name = "QuantizationOpsTdFiles",
srcs = [
"include/mlir/Dialect/Quant/QuantOps.td",
"include/mlir/Dialect/Quant/QuantOpsBase.td",
],
includes = ["include"],
deps = [
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "QuantOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Quant/QuantOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Quant/QuantOps.cpp.inc",
),
(
["-gen-dialect-decls"],
"include/mlir/Dialect/Quant/QuantOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
"include/mlir/Dialect/Quant/QuantOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-doc"],
"g3doc/Dialects/QuantOps/QuantOps.md",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Quant/QuantOps.td",
deps = [":QuantizationOpsTdFiles"],
)
[mlir] Convert quantized dialect bytecode to generated. Serves as rather self-contained documentation for using the generator from https://reviews.llvm.org/D144820. Differential Revision: https://reviews.llvm.org/D152118
2023-06-06 11:16:07 -07:00
gentbl_cc_library(
name = "QuantDialectBytecodeGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-bytecode",
"-bytecode-dialect=Quant",
],
"include/mlir/Dialect/Quant/QuantDialectBytecode.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Quant/QuantDialectBytecode.td",
deps = [
":BytecodeTdFiles",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "QuantOps",
srcs = [
[bazel] Port dd38f899803465dd2765d1601b3989df3bd53863 and fix 7732c97f52e72a0737aaafd19d1f3be9f26d1a20.
2022-10-17 16:54:37 -07:00
"lib/Dialect/Quant/IR/QuantDialectBytecode.cpp",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
"lib/Dialect/Quant/IR/QuantDialectBytecode.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"lib/Dialect/Quant/IR/QuantOps.cpp",
"lib/Dialect/Quant/IR/QuantTypes.cpp",
"lib/Dialect/Quant/IR/TypeDetail.h",
"lib/Dialect/Quant/IR/TypeParser.cpp",
"lib/Dialect/Quant/Utils/FakeQuantSupport.cpp",
"lib/Dialect/Quant/Utils/UniformSupport.cpp",
],
hdrs = [
"include/mlir/Dialect/Quant/FakeQuantSupport.h",
"include/mlir/Dialect/Quant/QuantOps.h",
"include/mlir/Dialect/Quant/QuantTypes.h",
"include/mlir/Dialect/Quant/UniformSupport.h",
],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][Bazel] Adapt to move of FuncOp.
2022-03-17 09:24:53 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[bazel] Port 92a836da0759
2022-04-28 22:51:27 +02:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
[mlir] Convert quantized dialect bytecode to generated. Serves as rather self-contained documentation for using the generator from https://reviews.llvm.org/D144820. Differential Revision: https://reviews.llvm.org/D152118
2023-06-06 11:16:07 -07:00
":QuantDialectBytecodeGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":QuantOpsIncGen",
":SideEffectInterfaces",
[bazel] Port dd38f899803465dd2765d1601b3989df3bd53863 and fix 7732c97f52e72a0737aaafd19d1f3be9f26d1a20.
2022-10-17 16:54:37 -07:00
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TransformUtils",
"//llvm:Support",
],
)
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
td_library(
name = "IndexOpsTdFiles",
srcs = [
"include/mlir/Dialect/Index/IR/IndexDialect.td",
"include/mlir/Dialect/Index/IR/IndexEnums.td",
"include/mlir/Dialect/Index/IR/IndexOps.td",
],
includes = ["include"],
deps = [
":CastInterfacesTdFiles",
Revert "Revert "[MLIR] Update bazel build file after 455305624884cf9237143e2ba0635fcc5ba5206"" This reverts commit 9ef7ae507387f133cf34ca0bc0edd3f009a3ff71.
2023-01-20 21:56:25 +01:00
":InferIntRangeInterfaceTdFiles",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
":InferTypeOpInterfaceTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "IndexOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Index/IR/IndexOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Index/IR/IndexOps.cpp.inc",
),
(
[
"-gen-dialect-decls",
"-dialect=index",
],
"include/mlir/Dialect/Index/IR/IndexOpsDialect.h.inc",
),
(
[
"-gen-dialect-defs",
"-dialect=index",
],
"include/mlir/Dialect/Index/IR/IndexOpsDialect.cpp.inc",
),
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/Index/IR/IndexOpsAttrDefs.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/Index/IR/IndexOpsAttrDefs.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Index/IR/IndexOps.td",
deps = [":IndexOpsTdFiles"],
)
gentbl_cc_library(
name = "IndexEnumsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-enum-decls"],
"include/mlir/Dialect/Index/IR/IndexEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/Index/IR/IndexEnums.cpp.inc",
),
(
[
"-gen-attrdef-decls",
"-attrdefs-dialect=index",
],
"include/mlir/Dialect/Index/IR/IndexAttrs.h.inc",
),
(
[
"-gen-attrdef-defs",
"-attrdefs-dialect=index",
],
"include/mlir/Dialect/Index/IR/IndexAttrs.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Index/IR/IndexEnums.td",
deps = [":IndexOpsTdFiles"],
)
cc_library(
name = "IndexToLLVM",
srcs = glob([
"lib/Conversion/IndexToLLVM/*.cpp",
"lib/Conversion/IndexToLLVM/*.h",
]),
hdrs = glob([
"include/mlir/Conversion/IndexToLLVM/*.h",
]),
includes = ["include"],
deps = [
":Analysis",
":ConversionPassIncGen",
":IR",
[mlir][Bazel] Port 3bef1e0f4c1d7fd901b58abe0c5c0e67da764ba2
2022-10-24 09:10:30 +02:00
":IndexDialect",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
":LLVMCommonConversion",
":LLVMDialect",
":Pass",
":Support",
":Transforms",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "IndexDialect",
srcs = glob(["lib/Dialect/Index/IR/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/Index/IR/*.h"]),
includes = ["include"],
deps = [
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfaces",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
":IR",
":IndexEnumsIncGen",
":IndexOpsIncGen",
Revert "Revert "[MLIR] Update bazel build file after 455305624884cf9237143e2ba0635fcc5ba5206"" This reverts commit 9ef7ae507387f133cf34ca0bc0edd3f009a3ff71.
2023-01-20 21:56:25 +01:00
":InferIntRangeCommon",
":InferIntRangeInterface",
Fix build.bazel files for adding Index dialect and LLVMTypes. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D136481
2022-10-21 12:23:46 -07:00
":InferTypeOpInterface",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "LinalgOpsTdFiles",
srcs = [
"include/mlir/Dialect/Linalg/IR/LinalgBase.td",
[mlir] Add IteratorType enum to StructuredOpsUtils. Summary: Use the new enum in TilingIterface and verify that `iterator_type` attribute in LinalgOp interface is compatible with the enum values. Later IteratorType enum will be used in LinalgInterface to replace the current `iterator_type` attribute array of string. Existing enums in Linalg are moved into a separate td file and tablegen build target. This is necessary, have one I32EnumAttr in a shared space that generated enum class definition and EnumAttrs is dialect-specific location. Otherwise there might be a conflict that I32EnumAttr generates enum definitions in multiple places. Differential Revision: https://reviews.llvm.org/D134634
2022-09-26 10:33:08 +00:00
"include/mlir/Dialect/Linalg/IR/LinalgEnums.td",
[mlir][bazel] Fix build.
2023-07-19 11:10:53 +02:00
"include/mlir/Dialect/Linalg/IR/LinalgInterfaces.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"include/mlir/Dialect/Linalg/IR/LinalgOps.td",
],
includes = ["include"],
deps = [
":ControlFlowInterfacesTdFiles",
[mlir][interfaces][NFC] Move DestinationStyleOpInterface to mlir/Interfaces This is the second (and final) step of making "destination style" usable without depending on the Linalg dialect. (The first step was D135129.) This change allows us to provide default bufferization implementations for all destination-style ops. It also allows us to simplify `TilingInterface`. (E.g., `getDestinationOperands` can be removed.) Differential Revision: https://reviews.llvm.org/D136179
2022-10-18 17:23:42 +02:00
":DestinationStyleOpInterfaceTdFiles",
[mlir] Add IteratorType enum to StructuredOpsUtils. Summary: Use the new enum in TilingIterface and verify that `iterator_type` attribute in LinalgOp interface is compatible with the enum values. Later IteratorType enum will be used in LinalgInterface to replace the current `iterator_type` attribute array of string. Existing enums in Linalg are moved into a separate td file and tablegen build target. This is necessary, have one I32EnumAttr in a shared space that generated enum class definition and EnumAttrs is dialect-specific location. Otherwise there might be a conflict that I32EnumAttr generates enum definitions in multiple places. Differential Revision: https://reviews.llvm.org/D134634
2022-09-26 10:33:08 +00:00
":DialectUtilsTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":InferTypeOpInterfaceTdFiles",
":LoopLikeInterfaceTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
[mlir] Add an interface to allow operations to specify how they can be tiled. An interface to allow for tiling of operations is introduced. The tiling of the linalg.pad_tensor operation is modified to use this interface. Differential Revision: https://reviews.llvm.org/D108611
2021-08-30 15:55:30 -07:00
":TilingInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ViewLikeInterfaceTdFiles",
],
)
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
td_library(
name = "LinalgTransformOpsTdFiles",
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
srcs = glob([
"include/mlir/Dialect/Linalg/TransformOps/*.td",
]),
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
includes = ["include"],
deps = [
[mlir] Introduce device mapper attribute for `thread_dim_map` and `mapped to dims` `scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [0, 1]} } { thread_dim_mapping = [0, 1]} ``` It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation. The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]} } { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]} ``` Reviewed By: ftynse, nicolasvasilache Differential Revision: https://reviews.llvm.org/D137413
2022-11-10 17:55:49 +01:00
":SCFTdFiles",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
":TransformDialectTdFiles",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "LinalgOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Linalg/IR/LinalgOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Linalg/IR/LinalgOps.cpp.inc",
),
(
[
"-gen-dialect-decls",
"-dialect=linalg",
],
"include/mlir/Dialect/Linalg/IR/LinalgOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=linalg",
],
"include/mlir/Dialect/Linalg/IR/LinalgOpsDialect.cpp.inc",
),
[mlir][OpDSL] Add type function attributes. Previously, OpDSL operation used hardcoded type conversion operations (cast or cast_unsigned). Supporting signed and unsigned casts thus meant implementing two different operations. Type function attributes allow us to define a single operation that has a cast type function attribute which at operation instantiation time may be set to cast or cast_unsigned. We may for example, defina a matmul operation with a cast argument: ``` @linalg_structured_op def matmul(A=TensorDef(T1, S.M, S.K), B=TensorDef(T2, S.K, S.N), C=TensorDef(U, S.M, S.N, output=True), cast=TypeFnAttrDef(default=TypeFn.cast)): C[D.m, D.n] += cast(U, A[D.m, D.k]) * cast(U, B[D.k, D.n]) ``` When instantiating the operation the attribute may be set to the desired cast function: ``` linalg.matmul(lhs, rhs, outs=[out], cast=TypeFn.cast_unsigned) ``` The revsion introduces a enum in the Linalg dialect that maps one-by-one to the type functions defined by OpDSL. Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D119718
2022-02-25 08:12:34 +00:00
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/Linalg/IR/LinalgOpsAttrDefs.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/Linalg/IR/LinalgOpsAttrDefs.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Linalg/IR/LinalgOps.td",
deps = [":LinalgOpsTdFiles"],
)
[mlir] Add IteratorType enum to StructuredOpsUtils. Summary: Use the new enum in TilingIterface and verify that `iterator_type` attribute in LinalgOp interface is compatible with the enum values. Later IteratorType enum will be used in LinalgInterface to replace the current `iterator_type` attribute array of string. Existing enums in Linalg are moved into a separate td file and tablegen build target. This is necessary, have one I32EnumAttr in a shared space that generated enum class definition and EnumAttrs is dialect-specific location. Otherwise there might be a conflict that I32EnumAttr generates enum definitions in multiple places. Differential Revision: https://reviews.llvm.org/D134634
2022-09-26 10:33:08 +00:00
gentbl_cc_library(
name = "LinalgEnumsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-enum-decls"],
"include/mlir/Dialect/Linalg/IR/LinalgOpsEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/Linalg/IR/LinalgOpsEnums.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Linalg/IR/LinalgEnums.td",
deps = [":LinalgOpsTdFiles"],
)
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
gentbl_cc_library(
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
name = "LinalgMatchOpsIncGen",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
"include/mlir/Dialect/Linalg/TransformOps/LinalgMatchOps.h.inc",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
),
(
["-gen-op-defs"],
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
"include/mlir/Dialect/Linalg/TransformOps/LinalgMatchOps.cpp.inc",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
),
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Linalg/TransformOps/LinalgMatchOps.td",
deps = [
":LinalgTransformEnumsIncGen",
":LinalgTransformOpsIncGen",
":LinalgTransformOpsTdFiles",
":SCFDeviceMappingInterfacesIncGen",
],
)
gentbl_cc_library(
name = "LinalgTransformEnumsIncGen",
strip_include_prefix = "include",
tbl_outs = [
[mlir][Linalg] Add a Transform dialect NavigationOp op to match a list of ops or an interface. This operation is a NavigationOp that simplifies the writing of transform IR. Since there is no way of refering to an interface by name, the current implementation uses an EnumAttr and depends on the interfaces it supports. In the future, it would be worthwhile to remove this dependence and generalize. Differential Revision: https://reviews.llvm.org/D130267
2022-07-21 06:44:43 -07:00
(
["-gen-enum-decls"],
"include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOpsEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOpsEnums.cpp.inc",
),
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
],
tblgen = ":mlir-tblgen",
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
td_file = "include/mlir/Dialect/Linalg/TransformOps/LinalgTransformEnums.td",
deps = [
":LinalgTransformOpsTdFiles",
":SCFDeviceMappingInterfacesIncGen",
],
)
gentbl_cc_library(
name = "LinalgTransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
td_file = "include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td",
deps = [
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
":LinalgTransformEnumsIncGen",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
":LinalgTransformOpsTdFiles",
[mlir] Introduce device mapper attribute for `thread_dim_map` and `mapped to dims` `scf.foreach_thread` defines mapping its loops to processors via an integer array, see an example below. A lowering can use this mapping. However, expressing mapping as an integer array is very confusing, especially when there are multiple levels of parallelism. In addition, the op does not verify the integer array. This change introduces device mapping attribute to make mapping descriptive and verifiable. Then it makes GPU transform dialect use it. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [0, 1]} } { thread_dim_mapping = [0, 1]} ``` It first introduces a `DeviceMappingInterface` which is an attribute interface. `scf.foreach_thread` defines its mapping via this interface. A lowering must define its attributes and implement this interface as well. This way gives us a clear validation. The change also introduces two new attributes (`#gpu.thread<x/y/z>` and `#gpu.block<x,y,z>` ). After this change, the above code prints as below, as seen here, this way clarifies the loop mappings. The change also implements consuming of these two new attribute by the transform dialect. Transform dialect binds the outermost loops to the thread blocks and innermost loops to threads. ``` scf.foreach_thread (%i, %j) in (%c1, %c2) { scf.foreach_thread (%i2, %j2) in (%c1, %c2) {...} { thread_dim_mapping = [#gpu.thread<x>, #gpu.thread<y>]} } { thread_dim_mapping = [#gpu.block<x>, #gpu.block<y>]} ``` Reviewed By: ftynse, nicolasvasilache Differential Revision: https://reviews.llvm.org/D137413
2022-11-10 17:55:49 +01:00
":SCFDeviceMappingInterfacesIncGen",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
genlinalg(
name = "LinalgNamedStructuredOpsYamlIncGen",
src = "include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml",
linalg_outs = [
(
"-o-impl=$@",
"include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yamlgen.cpp.inc",
),
(
"-o-ods-decl=$@",
"include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yamlgen.td",
),
],
linalggen = ":mlir-linalg-ods-yaml-gen",
)
td_library(
name = "LinalgStructuredOpsTdFiles",
srcs = [
"include/mlir/Dialect/Linalg/IR/LinalgInterfaces.td",
"include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yamlgen.td",
"include/mlir/Dialect/Linalg/IR/LinalgStructuredOps.td",
],
includes = ["include"],
deps = [
":CopyOpInterfaceTdFiles",
[mlir][interfaces][NFC] Move DestinationStyleOpInterface to mlir/Interfaces This is the second (and final) step of making "destination style" usable without depending on the Linalg dialect. (The first step was D135129.) This change allows us to provide default bufferization implementations for all destination-style ops. It also allows us to simplify `TilingInterface`. (E.g., `getDestinationOperands` can be removed.) Differential Revision: https://reviews.llvm.org/D136179
2022-10-18 17:23:42 +02:00
":DestinationStyleOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LinalgOpsTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "LinalgStructuredOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Linalg/IR/LinalgStructuredOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Linalg/IR/LinalgStructuredOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Linalg/IR/LinalgStructuredOps.td",
deps = [":LinalgStructuredOpsTdFiles"],
)
gentbl_cc_library(
name = "LinalgInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Dialect/Linalg/IR/LinalgInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Dialect/Linalg/IR/LinalgInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Linalg/IR/LinalgInterfaces.td",
deps = [":LinalgStructuredOpsTdFiles"],
)
[mlir][linalg][bufferize] Add BufferizableOpInterface This commit adds a new op interface: BufferizableOpInterface. In the future, ops that implement this interface can be bufferized using Comprehensive Bufferize. Note: The interface methods of this interface correspond to the "op interface" in ComprehensiveBufferize.cpp. Differential Revision: https://reviews.llvm.org/D112974
2021-11-02 16:05:27 +09:00
td_library(
name = "BufferizableOpInterfaceTdFiles",
srcs = [
[mlir][bufferization] Move one-shot bufferization to Bufferization dialect This commit is the first step towards unifying core bufferization and One-Shot Bufferize. This commit does not move over the implementations of BufferizableOpInterface yet. This will be done in separate commits. This change does also not move the unit tests yet. The tests will be moved together with op interface implementations and split into separate files. Differential Revision: https://reviews.llvm.org/D117641
2022-01-20 18:14:59 +09:00
"include/mlir/Dialect/Bufferization/IR/BufferizableOpInterface.td",
[mlir][linalg][bufferize] Add BufferizableOpInterface This commit adds a new op interface: BufferizableOpInterface. In the future, ops that implement this interface can be bufferized using Comprehensive Bufferize. Note: The interface methods of this interface correspond to the "op interface" in ComprehensiveBufferize.cpp. Differential Revision: https://reviews.llvm.org/D112974
2021-11-02 16:05:27 +09:00
],
includes = ["include"],
deps = [
":OpBaseTdFiles",
],
)
gentbl_cc_library(
name = "BufferizableOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
[mlir][bufferization] Move one-shot bufferization to Bufferization dialect This commit is the first step towards unifying core bufferization and One-Shot Bufferize. This commit does not move over the implementations of BufferizableOpInterface yet. This will be done in separate commits. This change does also not move the unit tests yet. The tests will be moved together with op interface implementations and split into separate files. Differential Revision: https://reviews.llvm.org/D117641
2022-01-20 18:14:59 +09:00
"include/mlir/Dialect/Bufferization/IR/BufferizableOpInterface.h.inc",
[mlir][linalg][bufferize] Add BufferizableOpInterface This commit adds a new op interface: BufferizableOpInterface. In the future, ops that implement this interface can be bufferized using Comprehensive Bufferize. Note: The interface methods of this interface correspond to the "op interface" in ComprehensiveBufferize.cpp. Differential Revision: https://reviews.llvm.org/D112974
2021-11-02 16:05:27 +09:00
),
(
["-gen-op-interface-defs"],
[mlir][bufferization] Move one-shot bufferization to Bufferization dialect This commit is the first step towards unifying core bufferization and One-Shot Bufferize. This commit does not move over the implementations of BufferizableOpInterface yet. This will be done in separate commits. This change does also not move the unit tests yet. The tests will be moved together with op interface implementations and split into separate files. Differential Revision: https://reviews.llvm.org/D117641
2022-01-20 18:14:59 +09:00
"include/mlir/Dialect/Bufferization/IR/BufferizableOpInterface.cpp.inc",
[mlir][linalg][bufferize] Add BufferizableOpInterface This commit adds a new op interface: BufferizableOpInterface. In the future, ops that implement this interface can be bufferized using Comprehensive Bufferize. Note: The interface methods of this interface correspond to the "op interface" in ComprehensiveBufferize.cpp. Differential Revision: https://reviews.llvm.org/D112974
2021-11-02 16:05:27 +09:00
),
],
tblgen = ":mlir-tblgen",
[mlir][bufferization] Move one-shot bufferization to Bufferization dialect This commit is the first step towards unifying core bufferization and One-Shot Bufferize. This commit does not move over the implementations of BufferizableOpInterface yet. This will be done in separate commits. This change does also not move the unit tests yet. The tests will be moved together with op interface implementations and split into separate files. Differential Revision: https://reviews.llvm.org/D117641
2022-01-20 18:14:59 +09:00
td_file = "include/mlir/Dialect/Bufferization/IR/BufferizableOpInterface.td",
[mlir][linalg][bufferize] Add BufferizableOpInterface This commit adds a new op interface: BufferizableOpInterface. In the future, ops that implement this interface can be bufferized using Comprehensive Bufferize. Note: The interface methods of this interface correspond to the "op interface" in ComprehensiveBufferize.cpp. Differential Revision: https://reviews.llvm.org/D112974
2021-11-02 16:05:27 +09:00
deps = [
":BufferizableOpInterfaceTdFiles",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "LinalgDocTdFiles",
srcs = ["include/mlir/Dialect/Linalg/IR/LinalgDoc.td"],
includes = ["include"],
deps = [
":LinalgOpsTdFiles",
":LinalgStructuredOpsTdFiles",
],
)
gentbl_cc_library(
name = "LinalgDocIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-doc"],
"g3doc/Dialects/Linalg/LinalgOps.md",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Linalg/IR/LinalgDoc.td",
deps = [":LinalgDocTdFiles"],
)
cc_library(
name = "LinalgToStandard",
srcs = glob([
"lib/Conversion/LinalgToStandard/*.cpp",
"lib/Conversion/LinalgToStandard/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/LinalgToStandard/*.h",
]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir] replace 'emit_c_wrappers' func->llvm conversion option with a pass The 'emit_c_wrappers' option in the FuncToLLVM conversion requests C interface wrappers to be emitted for every builtin function in the module. While this has been useful to bootstrap the interface, it is problematic in the longer term as it may unintentionally affect the functions that should retain their existing interface, e.g., libm functions obtained by lowering math operations (see D126964 for an example). Since D77314, we have a finer-grain control over interface generation via an attribute that avoids the problem entirely. Remove the 'emit_c_wrappers' option. Introduce the '-llvm-request-c-wrappers' pass that can be run in any pipeline that needs blanket emission of functions to annotate all builtin functions with the attribute before performing the usual lowering that accounts for the attribute. Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D127952
2022-06-16 12:53:49 +02:00
":LLVMDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LinalgTransforms",
":MemRefDialect",
":Pass",
":SCFDialect",
":Support",
":Transforms",
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
name = "LinalgDialect",
srcs = glob(["lib/Dialect/Linalg/IR/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/Linalg/IR/*.h"]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
":AsmParser",
[mlir][bufferize][NFC] Implement BufferizableOpInterface on bufferization ops directly No longer go through an external model. Also put BufferizableOpInterface into the same build target as the BufferizationDialect. This allows for some code reuse between BufferizationOps canonicalizers and BufferizableOpInterface implementations. Differential Revision: https://reviews.llvm.org/D117987
2022-01-25 00:51:41 +09:00
":BufferizationDialect",
[mlir][linalg] Add lowering of named ops on complex numbers This lets linalg.dot and friends lower to a complex muladd using ops from the complex dialect. Differential Revision: https://reviews.llvm.org/D125461
2022-05-12 13:35:27 +02:00
":ComplexDialect",
[mlir][Bazel] Add missing dependencies. When building with layering_check enabled, there needs to be a dependency for each header include.
2022-04-25 09:16:16 +02:00
":ControlFlowInterfaces",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":CopyOpInterface",
[mlir][interfaces][NFC] Move DestinationStyleOpInterface to mlir/Interfaces This is the second (and final) step of making "destination style" usable without depending on the Linalg dialect. (The first step was D135129.) This change allows us to provide default bufferization implementations for all destination-style ops. It also allows us to simplify `TilingInterface`. (E.g., `getDestinationOperands` can be removed.) Differential Revision: https://reviews.llvm.org/D136179
2022-10-18 17:23:42 +02:00
":DestinationStyleOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":InferTypeOpInterface",
[mlir] Add IteratorType enum to StructuredOpsUtils. Summary: Use the new enum in TilingIterface and verify that `iterator_type` attribute in LinalgOp interface is compatible with the enum values. Later IteratorType enum will be used in LinalgInterface to replace the current `iterator_type` attribute array of string. Existing enums in Linalg are moved into a separate td file and tablegen build target. This is necessary, have one I32EnumAttr in a shared space that generated enum class definition and EnumAttrs is dialect-specific location. Otherwise there might be a conflict that I32EnumAttr generates enum definitions in multiple places. Differential Revision: https://reviews.llvm.org/D134634
2022-09-26 10:33:08 +00:00
":LinalgEnumsIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LinalgInterfacesIncGen",
":LinalgNamedStructuredOpsYamlIncGen",
":LinalgOpsIncGen",
":LinalgStructuredOpsIncGen",
[mlir][bazel] Added missing MathDialect dep to LinalgOps target LinalgOps needs MathDialect as a dependency for it to build correctly. Differential Revision: https://reviews.llvm.org/D105656
2021-07-08 13:25:46 -07:00
":MathDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MemRefDialect",
":Parser",
[mlir] Add an interface to allow operations to specify how they can be tiled. An interface to allow for tiling of operations is introduced. The tiling of the linalg.pad_tensor operation is modified to use this interface. Differential Revision: https://reviews.llvm.org/D108611
2021-08-30 15:55:30 -07:00
":SCFDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfaces",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":SparseTensorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":TensorDialect",
[mlir] Add an interface to allow operations to specify how they can be tiled. An interface to allow for tiling of operations is introduced. The tiling of the linalg.pad_tensor operation is modified to use this interface. Differential Revision: https://reviews.llvm.org/D108611
2021-08-30 15:55:30 -07:00
":TilingInterface",
[mlir] fix mismerge in bazel
2023-04-13 14:58:11 +02:00
":ValueBoundsOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ViewLikeInterface",
"//llvm:Support",
],
)
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
cc_library(
name = "LinalgTransformOps",
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
srcs = glob([
"lib/Dialect/Linalg/TransformOps/*.cpp",
]),
[mlir] reorgnize Linalg TransformOps files. NFC Mirror the separation between LinalgTransformOps and LinalgMatchOps in headers. Create a separate pair of files for the extension. Depends on D148017 Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D148075
2023-04-12 08:10:24 +00:00
hdrs = glob([
"include/mlir/Dialect/Linalg/TransformOps/*.h",
]),
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
includes = ["include"],
deps = [
[mlir] Transform op for multitile size generation Introduce a structured transform op that emits IR computing the multi-tile sizes with requested parameters (target size and divisor) for the given structured op. The sizes may fold to arithmetic constant operations when the shape is constant. These operations may then be used to call the existing tiling transformation with a single non-zero dynamic size (i.e. perform strip-mining) for each of the dimensions separately, thus achieving multi-size tiling with optional loop interchange. A separate test exercises the entire script. Depends On D129217 Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D129287
2022-07-07 15:56:06 +02:00
":AffineDialect",
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
":AsmParser",
[mlir][linalg] LinalgOp-anchored empty tensor elimination This revision adds a pre-bufferization transform that can reduce the number of allocation. It is similar to `bufferization.eliminate_empty_tensors`, but specific to LinalgOp. The transform looks for `tensor.empty` ops where the SSA use-def chain ends in an "ins" operand of a `LinalgOp`. If the same `LinalgOp` has an unused "outs" operand (and some other conditions are met), this "outs" operand can be used instead of the `tensor.empty` and the "ins" operand can be turned into an "outs" operand. Differential Revision: https://reviews.llvm.org/D153952
2023-07-03 09:15:22 +02:00
":BufferizationTransforms",
[bazel] Add missing dependencies for 790f237012
2023-01-20 11:05:29 +01:00
":DialectUtils",
[mlir][linalg][TransformOps] Connect hoistRedundantVectorTransfers Connect the hoistRedundantVectorTransfers functionality to the transform dialect. Authored-by: Quentin Colombet <quentin.colombet@gmail.com> Differential Revision: https://reviews.llvm.org/D144260
2023-02-20 01:40:18 -08:00
":FuncDialect",
[bazel] Port 233de4e808b3
2022-09-19 16:38:20 +02:00
":GPUDialect",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
":IR",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
":LinalgMatchOpsIncGen",
":LinalgTransformEnumsIncGen",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
":LinalgTransformOpsIncGen",
":LinalgTransforms",
[mlir][Linalg] Add a transform.structured.lower_pack op This revision introduces `transform.structured.lower_pack` which allows rewriting a `tensor.pack` to `tensor.pad` + `tensor.expand_shape` + `linalg.transpose`. The implementation is currently limited to static pack ops that do not have outer_dims permutations. Differential Revision: https://reviews.llvm.org/D142881
2023-01-27 08:24:14 -08:00
":LinalgUtils",
[mlir] reorgnize Linalg TransformOps files. NFC Mirror the separation between LinalgTransformOps and LinalgMatchOps in headers. Create a separate pair of files for the extension. Depends on D148017 Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D148075
2023-04-12 08:10:24 +00:00
":SCFDialect",
[mlir][Bazel] Fix for reviews.llvm.org/D135559
2022-10-10 16:27:53 +02:00
":SCFTransforms",
[bazel] Add missing dependency after 55cf0de35e
2023-01-26 23:25:15 +01:00
":Support",
[bazel] Add missing dependencies for 790f237012
2023-01-20 11:05:29 +01:00
":TensorDialect",
[mlir][Linalg] Add a transform.structured.lower_pack op This revision introduces `transform.structured.lower_pack` which allows rewriting a `tensor.pack` to `tensor.pad` + `tensor.expand_shape` + `linalg.transpose`. The implementation is currently limited to static pack ops that do not have outer_dims permutations. Differential Revision: https://reviews.llvm.org/D142881
2023-01-27 08:24:14 -08:00
":TensorUtils",
[bazel] Fix the build after 18b92c66fe59a44f50bc211a418eaf48fe1cf7c1
2022-07-19 17:34:39 +02:00
":TilingInterface",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
":TransformDialect",
[mlir][transform] Add PackedOrDynamicIndexList helper This customer parser/printer is similar to DynamicIndexList, but has special syntax for the case where one handle represents the entire list. Example: ``` // Regular index list [10, 20, %val] // Packed handle (no square parentheses) %val ``` Differential Revision: https://reviews.llvm.org/D138825
2022-12-19 07:57:46 +01:00
":TransformDialectUtils",
[mlir] add VectorizeOp to structured transform ops Vectorization is a key transformation to achieve high performance on most architectures. In the transform dialect, vectorization is implemented as a parameterizable transform op. It currently applies to a scope of payload IR delimited by some isolated-from-above op, mainly because several enabling transformations (such as affine simplification) are needed to perform vectorization and these transformation would apply to ops other than the "main" computational payload op. A separate "navigation" transform op that obtains the isolated-from-above ancestor of an op is introduced in the core transform dialect. Even though it is currently only useful for vectorization, isolated-from-above ops are a common anchor for transformations (usually implemented as passes) that is likely to be reused in the future. Depends On D126374 Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D126542
2022-05-30 15:13:23 +02:00
":TransformUtils",
[mlir] reorgnize Linalg TransformOps files. NFC Mirror the separation between LinalgTransformOps and LinalgMatchOps in headers. Create a separate pair of files for the extension. Depends on D148017 Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D148075
2023-04-12 08:10:24 +00:00
":VectorDialect",
[mlir][Vector] NFC - Reorganize vector patterns Vector dialect patterns have grown enormously in the past year to a point where they are now impenetrable. Start reorganizing them towards finer-grained control. Differential Revision: https://reviews.llvm.org/D146736
2023-03-23 08:32:48 -07:00
":VectorTransforms",
[mlir][linalg][transform] Add TileOp to transform dialect This commit adds a tiling op to the transform dialect as an external op. Differential Revision: https://reviews.llvm.org/D124661
2022-04-29 21:34:41 +09:00
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "LinalgPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Linalg",
],
"include/mlir/Dialect/Linalg/Passes.h.inc",
),
(
[
"-gen-pass-capi-header",
"--prefix=Linalg",
],
"include/mlir/Dialect/Linalg/Passes.capi.h.inc",
),
(
[
"-gen-pass-capi-impl",
"--prefix=Linalg",
],
"include/mlir/Dialect/Linalg/Passes.capi.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Linalg/Passes.td",
deps = [":PassBaseTdFiles"],
)
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
cc_library(
name = "LinalgUtils",
srcs = glob([
"lib/Dialect/Linalg/Utils/*.cpp",
"lib/Dialect/Linalg/Utils/*.h",
]),
hdrs = glob([
"include/mlir/Dialect/Linalg/Utils/*.h",
]),
deps = [
":AffineAnalysis",
":AffineDialect",
":AffineUtils",
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":DialectUtils",
[mlir][linalg] move isElementwise() to Linalg/Utils (NFC) Differential Revision: https://reviews.llvm.org/D128398
2022-06-22 15:59:53 -07:00
":FuncDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":IR",
":LinalgDialect",
":MemRefDialect",
":Pass",
":SCFDialect",
":TensorDialect",
":TensorUtils",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "LinalgTransforms",
[mlir][linalg][bufferize] Separate pass from ComprehensiveBufferize This commit separates the bufferization from the bufferization pass in Linalg. This allows other dialects to use ComprehensiveBufferize more easily. This commit mainly moves files to a new directory and adds a new build target. Differential Revision: https://reviews.llvm.org/D112989
2021-11-05 13:55:47 +09:00
srcs = glob([
"lib/Dialect/Linalg/Transforms/*.cpp",
"lib/Dialect/Linalg/Transforms/*.h",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = [
"include/mlir/Dialect/Linalg/Passes.h",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
] + glob([
"include/mlir/Dialect/Linalg/Transforms/*.h",
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[bazel] Port 755dc07d69ed Still a fair bit of spaghetti in the unittest rules
2022-01-19 00:55:56 +01:00
":AffineAnalysis",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir] fix mismerge in bazel
2023-04-13 14:58:11 +02:00
":AffineTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AffineUtils",
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithTransforms",
":ArithUtils",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationDialect",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":BufferizationTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ComplexDialect",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":ControlFlowDialect",
[mlir][interfaces][NFC] Move DestinationStyleOpInterface to mlir/Interfaces This is the second (and final) step of making "destination style" usable without depending on the Linalg dialect. (The first step was D135129.) This change allows us to provide default bufferization implementations for all destination-style ops. It also allows us to simplify `TilingInterface`. (E.g., `getDestinationOperands` can be removed.) Differential Revision: https://reviews.llvm.org/D136179
2022-10-18 17:23:42 +02:00
":DestinationStyleOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[bazel] Build fixes for 23aa5a744666
2022-03-02 00:07:56 +01:00
":FuncTransforms",
[mlir] Port bazel for 115711c19cd287c098a872c63a00478ca635f642
2023-02-27 16:50:39 +01:00
":GPUDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LinalgPassIncGen",
":LinalgStructuredOpsIncGen",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":LinalgUtils",
[mlir][Linalg] Refactor HoistPadding and add support for hoisting in the absence of packing loops. This revision cleans up the implementation of hoist padding and extends it to also work in the absence of packing loops. This allows better composition when hoisting the padded result of a DPS operation. A systematic usage of RewriterBase is applied to the implementation. Depends on: D144856 Differential Revision: https://reviews.llvm.org/D144855
2023-02-27 01:28:56 -08:00
":LoopLikeInterface",
[mlir][Vector] Initial masking support in Linalg vectorizer This patch introduces the initial bits to support vector masking using the `vector.mask` operation. Vectorization changes should be NFC for non-masked cases. We can't test masked cases directly until we extend the Transform dialect to support masking. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D137690
2022-11-24 02:16:46 +00:00
":MaskableOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MathDialect",
":MemRefDialect",
[mlir][linalg] Add a new pattern to handle folding unit reduction dims. The output operands will be added to input operands if the generic op (on tensors) becomes an elementwise operation. The outputs of the generic op is still the same. They will be cleaned up by ReplaceWithEmptyTensorIfUnused pattern. This is https://reviews.llvm.org/D138251, plus a cmake dep fix. Reviewed By: mravishankar Differential Revision: https://reviews.llvm.org/D138843
2022-11-23 10:46:46 -08:00
":MemRefTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":SCFDialect",
":SCFTransforms",
[mlir] Move SCF utils implementations to SCF/Utils. BEGIN_PUBLIC [mlir] Move SCF utils implementations to SCF/Utils. END_PUBLIC
2022-01-27 09:10:59 +01:00
":SCFUtils",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":SparseTensorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Support",
":TensorDialect",
[mlir][linalg] Add control to pad-slice swap pattern The pad-slice swap pattern generates `scf.if` and `tensor.generate` to guard against zero-sized slices if it cannot prove the slice is always non-zero. This is safe but quite conservative. It can be unnecessary for cases where we know by problem definition such cases does not exist, even if with dynamic shaped ops or unknown tile/slice sizes, e.g., convolution padding size = 1 with kernel dim size = 3. So this commit introduces a control to the pattern to specify whether to generate the if constructs to handle such cases better, given that once the if constructs is materialized, it's very hard to analyze and simplify. Reviewed By: mravishankar Differential Revision: https://reviews.llvm.org/D117017
2022-02-16 10:28:51 -05:00
":TensorTilingInterfaceImpl",
[mlir][tensor] Move BufferizableOpInterface impl to tensor dialect This is in preparation of unifying the existing bufferization with One-Shot bufferization. A subsequent commit will replace `tensor-bufferize`'s implementation with the BufferizableOpInterface-based implementation and move over missing test cases. Differential Revision: https://reviews.llvm.org/D117984
2022-01-24 23:16:29 +09:00
":TensorTransforms",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
":TensorUtils",
[mlir][TilingInterface] Add pattern to tile using TilingInterface and implement TilingInterface for Linalg ops. This patch adds support for tiling operations that implement the TilingInterface. - It separates the loop constructs that are used to iterate over tile from the implementation of the tiling itself. For example, the use of destructive updates is more related to use of scf.for for iterating over tiles that are tensors. - To test the transformation, TilingInterface is implemented for LinalgOps. The separation of the looping constructs used from the implementation of tile code generation greatly simplifies the latter. - The implementation of TilingInterface for LinalgOp is kept as an external model for now till this approach can be fully flushed out to replace the existing tiling + fusion approaches in Linalg. Differential Revision: https://reviews.llvm.org/D127133
2022-06-13 19:56:32 +00:00
":TilingInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TransformUtils",
Remove spurious includes and dependencies from Bazel files (NFC) Differential Revision: https://reviews.llvm.org/D119526
2022-02-11 07:14:40 +00:00
":Transforms",
[mlir] fix mismerge in bazel
2023-04-13 14:58:11 +02:00
":ValueBoundsOpInterface",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":VectorToSCF",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":VectorTransforms",
":VectorUtils",
[mlir][X86Vector] Add specialized vector.transpose lowering patterns for AVX2 This revision adds an implementation of 2-D vector.transpose for 4x8 and 8x8 for AVX2 and surfaces it to the Linalg level of control. Reviewed By: dcaballe Differential Revision: https://reviews.llvm.org/D113347
2021-11-10 13:19:41 +00:00
":X86VectorTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
[mlir][Interfaces] Add ValueBoundsOpInterface Ops can implement this interface to specify lower/upper bounds for their result values and block arguments. Bounds can be specified for: * Index-type values * Dimension sizes of shapes values The bounds are added to a constraint set. Users can query this constraint set to compute bounds wrt. to a user-specified set of values. Only EQ bounds are supported at the moment. This revision also contains interface implementations for various tensor dialect ops, which illustrates how to implement this interface. Differential Revision: https://reviews.llvm.org/D145681
2023-04-06 02:41:15 +02:00
td_library(
name = "ValueBoundsOpInterfaceTdFiles",
srcs = [
"include/mlir/Interfaces/ValueBoundsOpInterface.td",
],
includes = ["include"],
deps = [
":OpBaseTdFiles",
],
)
gentbl_cc_library(
name = "ValueBoundsOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Interfaces/ValueBoundsOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Interfaces/ValueBoundsOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Interfaces/ValueBoundsOpInterface.td",
deps = [
":ValueBoundsOpInterfaceTdFiles",
],
)
cc_library(
name = "ValueBoundsOpInterface",
srcs = ["lib/Interfaces/ValueBoundsOpInterface.cpp"],
hdrs = ["include/mlir/Interfaces/ValueBoundsOpInterface.h"],
includes = ["include"],
deps = [
":Analysis",
":DestinationStyleOpInterface",
":IR",
":Support",
":ValueBoundsOpInterfaceIncGen",
"//llvm:Support",
],
)
[mlir][arith] Add ValueBoundsOpInterface impls These ops are useful for unit testing. (They do not fold/canonicalize with affine.apply etc.) Differential Revision: https://reviews.llvm.org/D145696
2023-04-06 10:46:48 +09:00
cc_library(
name = "ArithValueBoundsOpInterfaceImpl",
srcs = ["lib/Dialect/Arith/IR/ValueBoundsOpInterfaceImpl.cpp"],
hdrs = ["include/mlir/Dialect/Arith/IR/ValueBoundsOpInterfaceImpl.h"],
includes = ["include"],
deps = [
":ArithDialect",
":IR",
":ValueBoundsOpInterface",
],
)
[mlir] Add an interface to allow operations to specify how they can be tiled. An interface to allow for tiling of operations is introduced. The tiling of the linalg.pad_tensor operation is modified to use this interface. Differential Revision: https://reviews.llvm.org/D108611
2021-08-30 15:55:30 -07:00
cc_library(
name = "TilingInterface",
srcs = ["lib/Interfaces/TilingInterface.cpp"],
hdrs = ["include/mlir/Interfaces/TilingInterface.h"],
includes = ["include"],
deps = [
[mlir] Add IteratorType enum to StructuredOpsUtils. Summary: Use the new enum in TilingIterface and verify that `iterator_type` attribute in LinalgOp interface is compatible with the enum values. Later IteratorType enum will be used in LinalgInterface to replace the current `iterator_type` attribute array of string. Existing enums in Linalg are moved into a separate td file and tablegen build target. This is necessary, have one I32EnumAttr in a shared space that generated enum class definition and EnumAttrs is dialect-specific location. Otherwise there might be a conflict that I32EnumAttr generates enum definitions in multiple places. Differential Revision: https://reviews.llvm.org/D134634
2022-09-26 10:33:08 +00:00
":DialectUtils",
[mlir] Add an interface to allow operations to specify how they can be tiled. An interface to allow for tiling of operations is introduced. The tiling of the linalg.pad_tensor operation is modified to use this interface. Differential Revision: https://reviews.llvm.org/D108611
2021-08-30 15:55:30 -07:00
":IR",
":Support",
":TilingInterfaceIncGen",
":ViewLikeInterface",
"//llvm:Support",
],
)
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
##---------------------------------------------------------------------------##
# Vector dialect.
##---------------------------------------------------------------------------##
td_library(
[mlir][Vector] Introduce the MaskingOpInterface This MaskingOpInterface provides masking cababilitites to those operations that implement it. For only is only implemented by the `vector.mask` operation and it's used to break the dependency between the Vector dialect (where the `vector.mask` op lives) and operations implementing the MaskableOpInterface. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D136734
2022-10-26 00:58:56 +00:00
name = "MaskingOpInterfaceTdFiles",
srcs = ["include/mlir/Dialect/Vector/Interfaces/MaskingOpInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
td_library(
name = "MaskableOpInterfaceTdFiles",
srcs = ["include/mlir/Dialect/Vector/Interfaces/MaskableOpInterface.td"],
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "VectorOpsTdFiles",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
srcs = ["include/mlir/Dialect/Vector/IR/VectorOps.td"],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
[mlir][vector] Add operations used for Vector distribution Add vector op warp_execute_on_lane_0 that will be used to do incremental vector distribution in order to target warp level vector programming for architectures with GPU-like SIMT programming model. The idea behing the op is discussed further on discourse: https://discourse.llvm.org/t/vector-vector-distribution-large-vector-to-small-vector/1983/23 Differential Revision: https://reviews.llvm.org/D123703
2022-04-13 18:38:11 +00:00
":ControlFlowInterfacesTdFiles",
[mlir][vector][bufferize] Implement DestinationStyleOpInterface on TransferWriteOp This simplifies the BufferizableOpInterface implementation of vector.transfer_write. Differential Revision: https://reviews.llvm.org/D136348
2022-10-27 10:59:52 +02:00
":DestinationStyleOpInterfaceTdFiles",
[mlir][Vector] Add InferTypeOpInterface to vector dialect and remove the now redundant builders
2022-02-18 01:15:14 +01:00
":InferTypeOpInterfaceTdFiles",
[mlir][Vector] Introduce the MaskingOpInterface This MaskingOpInterface provides masking cababilitites to those operations that implement it. For only is only implemented by the `vector.mask` operation and it's used to break the dependency between the Vector dialect (where the `vector.mask` op lives) and operations implementing the MaskableOpInterface. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D136734
2022-10-26 00:58:56 +00:00
":MaskableOpInterfaceTdFiles",
":MaskingOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
":VectorInterfacesTdFiles",
":ViewLikeInterfaceTdFiles",
],
)
[mlir][Vector] Vector transform skeleton Differential Revision: https://reviews.llvm.org/D134722
2022-06-21 06:52:01 -07:00
td_library(
name = "VectorTransformOpsTdFiles",
srcs = [
"include/mlir/Dialect/Vector/TransformOps/VectorTransformOps.td",
],
includes = ["include"],
deps = [
":PDLDialectTdFiles",
":SCFTdFiles",
":TransformDialectTdFiles",
Fix bazel build overlay.
2023-01-17 13:17:56 +01:00
":VectorTransformsTdFiles",
[mlir][Vector] Vector transform skeleton Differential Revision: https://reviews.llvm.org/D134722
2022-06-21 06:52:01 -07:00
],
)
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
gentbl_cc_library(
[mlir][Vector] Introduce the MaskingOpInterface This MaskingOpInterface provides masking cababilitites to those operations that implement it. For only is only implemented by the `vector.mask` operation and it's used to break the dependency between the Vector dialect (where the `vector.mask` op lives) and operations implementing the MaskableOpInterface. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D136734
2022-10-26 00:58:56 +00:00
name = "MaskableOpInterfaceIncGen",
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
[mlir][Vector] Introduce the MaskingOpInterface This MaskingOpInterface provides masking cababilitites to those operations that implement it. For only is only implemented by the `vector.mask` operation and it's used to break the dependency between the Vector dialect (where the `vector.mask` op lives) and operations implementing the MaskableOpInterface. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D136734
2022-10-26 00:58:56 +00:00
"include/mlir/Dialect/Vector/Interfaces/MaskableOpInterface.h.inc",
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
),
(
["-gen-op-interface-defs"],
[mlir][Vector] Introduce the MaskingOpInterface This MaskingOpInterface provides masking cababilitites to those operations that implement it. For only is only implemented by the `vector.mask` operation and it's used to break the dependency between the Vector dialect (where the `vector.mask` op lives) and operations implementing the MaskableOpInterface. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D136734
2022-10-26 00:58:56 +00:00
"include/mlir/Dialect/Vector/Interfaces/MaskableOpInterface.cpp.inc",
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
),
],
tblgen = ":mlir-tblgen",
[mlir][Vector] Introduce the MaskingOpInterface This MaskingOpInterface provides masking cababilitites to those operations that implement it. For only is only implemented by the `vector.mask` operation and it's used to break the dependency between the Vector dialect (where the `vector.mask` op lives) and operations implementing the MaskableOpInterface. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D136734
2022-10-26 00:58:56 +00:00
td_file = "include/mlir/Dialect/Vector/Interfaces/MaskableOpInterface.td",
deps = [":MaskableOpInterfaceTdFiles"],
)
gentbl_cc_library(
name = "MaskingOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Dialect/Vector/Interfaces/MaskingOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Dialect/Vector/Interfaces/MaskingOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Vector/Interfaces/MaskingOpInterface.td",
deps = [":MaskingOpInterfaceTdFiles"],
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "VectorOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
"include/mlir/Dialect/Vector/IR/VectorOps.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-op-defs"],
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
"include/mlir/Dialect/Vector/IR/VectorOps.cpp.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
[
"-gen-dialect-decls",
"-dialect=vector",
],
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
"include/mlir/Dialect/Vector/IR/VectorOpsDialect.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=vector",
],
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
"include/mlir/Dialect/Vector/IR/VectorOpsDialect.cpp.inc",
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-enum-decls"],
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
"include/mlir/Dialect/Vector/IR/VectorOpsEnums.h.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
(
["-gen-enum-defs"],
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
"include/mlir/Dialect/Vector/IR/VectorOpsEnums.cpp.inc",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
),
[mlir] Change CombiningKind in Vector dialect to EnumAttr. CombiningKind was implemented before EnumAttr, so it reimplements the same behaviour with the custom code. Except for a few places, EnumAttr is a drop-in replacement. Reviewed By: nicolasvasilache, pifon2a Differential Revision: https://reviews.llvm.org/D133343
2022-09-07 13:33:02 +02:00
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/Vector/IR/VectorOpsAttrDefs.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/Vector/IR/VectorOpsAttrDefs.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-doc"],
"g3doc/Dialects/Vector/VectorOps.md",
),
],
tblgen = ":mlir-tblgen",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
td_file = "include/mlir/Dialect/Vector/IR/VectorOps.td",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
deps = [":VectorOpsTdFiles"],
)
[mlir][Vector] Vector transform skeleton Differential Revision: https://reviews.llvm.org/D134722
2022-06-21 06:52:01 -07:00
gentbl_cc_library(
name = "VectorTransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Vector/TransformOps/VectorTransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Vector/TransformOps/VectorTransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Vector/TransformOps/VectorTransformOps.td",
deps = [
":VectorTransformOpsTdFiles",
],
)
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
cc_library(
[mlir][Vector] Introduce the MaskingOpInterface This MaskingOpInterface provides masking cababilitites to those operations that implement it. For only is only implemented by the `vector.mask` operation and it's used to break the dependency between the Vector dialect (where the `vector.mask` op lives) and operations implementing the MaskableOpInterface. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D136734
2022-10-26 00:58:56 +00:00
name = "MaskableOpInterface",
srcs = ["lib/Dialect/Vector/Interfaces/MaskableOpInterface.cpp"],
hdrs = ["include/mlir/Dialect/Vector/Interfaces/MaskableOpInterface.h"],
includes = ["include"],
deps = [
":IR",
":MaskableOpInterfaceIncGen",
":MaskingOpInterface",
":Support",
"//llvm:Support",
],
)
cc_library(
name = "MaskingOpInterface",
srcs = ["lib/Dialect/Vector/Interfaces/MaskingOpInterface.cpp"],
hdrs = ["include/mlir/Dialect/Vector/Interfaces/MaskingOpInterface.h"],
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
includes = ["include"],
deps = [
":IR",
[mlir][Vector] Introduce the MaskingOpInterface This MaskingOpInterface provides masking cababilitites to those operations that implement it. For only is only implemented by the `vector.mask` operation and it's used to break the dependency between the Vector dialect (where the `vector.mask` op lives) and operations implementing the MaskableOpInterface. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D136734
2022-10-26 00:58:56 +00:00
":MaskingOpInterfaceIncGen",
[mlir][Vector] Introduce 'vector.mask' operation and MaskableOpInterface This patch introduces the `vector.mask` operation and the MaskableOpInterface as described in https://discourse.llvm.org/t/rfc-vector-masking-representation-in-mlir/64964. The `vector.mask` operation is used to predicate the execution of operations implementing the MaskableOpInterface. This interface will be implemented by maskable operations and provides information about its masking constraints and semantics. For now, only vector transfer and reduction ops implement the MaskableOpInterface for illustration and testing purposes. Reviewed By: nicolasvasilache, rriddle Differential Revision: https://reviews.llvm.org/D134939
2022-10-10 21:21:03 +00:00
":Support",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "VectorToLLVM",
srcs = glob([
"lib/Conversion/VectorToLLVM/*.cpp",
"lib/Conversion/VectorToLLVM/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/VectorToLLVM/*.h",
]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AMXDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AMXTransforms",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ArmNeonDialect",
[bazel][mlir] Port 65305aeab99ad8ea09dd85e28a41c657152a08fb
2023-06-16 13:20:32 +02:00
":ArmSMEDialect",
":ArmSMETransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":ArmSVEDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ArmSVETransforms",
":ConversionPassIncGen",
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Update configuration to HEAD Bazel configuration updates for https://github.com/llvm/llvm-project/commit/75e5f0aac97d, https://github.com/llvm/llvm-project/commit/c282d55a3857, and https://github.com/llvm/llvm-project/commit/26e59cc19f86 Tested: `bazel test ... @llvm-project//...` Differential Revision: https://reviews.llvm.org/D105830
2021-07-12 11:24:13 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
[mlir][Bazel] Add dependency needed for e9b82a5c
2023-02-15 07:36:00 +01:00
":MaskableOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MemRefDialect",
":Pass",
":Support",
":ToLLVMIRTranslation",
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":VectorTransforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":X86VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":X86VectorTransforms",
"//llvm:Core",
"//llvm:Support",
],
)
[mlir][Bazel] Add VectorToArmSME target.
2023-07-18 16:02:15 +02:00
cc_library(
name = "VectorToArmSME",
srcs = glob([
"lib/Conversion/VectorToArmSME/*.cpp",
"lib/Conversion/VectorToArmSME/*.h",
]),
hdrs = glob([
"include/mlir/Conversion/VectorToArmSME/*.h",
]),
includes = ["include"],
deps = [
":ArmSMEDialect",
[bazel][mlir] Port ca9a3354d04b15366088d7831b40f891e3d77b95
2023-07-25 11:38:35 +02:00
":ArmSMEUtils",
[mlir][Bazel] Add VectorToArmSME target.
2023-07-18 16:02:15 +02:00
":ConversionPassIncGen",
":IR",
":Pass",
":Transforms",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "VectorToGPU",
srcs = glob([
"lib/Conversion/VectorToGPU/*.cpp",
"lib/Conversion/VectorToGPU/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/VectorToGPU/*.h",
]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":GPUDialect",
":IR",
":LLVMDialect",
":MemRefDialect",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":NVGPUDialect",
Add NVGPUUtil bazel lib
2023-01-17 15:22:30 -08:00
":NVGPUUtils",
Adjust BUILD files for [MLIR][GPU] Add NvGpu mma.sync path to the VectorToGPU pass
2022-05-20 18:03:20 +02:00
":NVVMDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
":SCFDialect",
":Support",
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir][vector] Add missing build dependencies for VectorToGPU Post-commit fix for https://reviews.llvm.org/D145638.
2023-03-09 15:13:28 -05:00
":VectorTransforms",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":VectorUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
"//llvm:Support",
],
)
cc_library(
name = "VectorToSCF",
srcs = glob([
"lib/Conversion/VectorToSCF/*.cpp",
"lib/Conversion/VectorToSCF/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/VectorToSCF/*.h",
]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":AffineUtils",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":FuncToLLVM",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":LLVMDialect",
":MemRefDialect",
":Pass",
":SCFDialect",
":Support",
Fix bazel build
2023-01-31 17:05:49 -08:00
":TensorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":VectorTransforms",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
"//llvm:Support",
],
)
td_library(
name = "TosaDialectTdFiles",
srcs = glob(["include/mlir/Dialect/Tosa/IR/*.td"]),
deps = [
[mlir][tosa] Update Bazel files for TOSA pass changes There were some missing bazel dependencies for the Tosa dialect. Added these deps. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D105326
2021-07-01 17:44:52 -07:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LoopLikeInterfaceTdFiles",
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
[mlir][tosa] Enable `apply_scale` unrolling Make `tosa.apply_scale` implement `VectorUnrollOpInterface` so that we can unroll it. Reviewed By: rsuderman Differential Revision: https://reviews.llvm.org/D143944
2023-02-13 16:51:37 -05:00
":VectorInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
gentbl_cc_library(
name = "TosaDialectIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Tosa/IR/TosaOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Tosa/IR/TosaOps.cpp.inc",
),
(
["-gen-dialect-decls"],
"include/mlir/Dialect/Tosa/IR/TosaOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
["-gen-dialect-defs"],
"include/mlir/Dialect/Tosa/IR/TosaOpsDialect.cpp.inc",
),
[mlir][tosa] Replace StructAttrs with AttrDefs Depends on D127352 Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D127370
2022-06-09 21:34:45 +00:00
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/Tosa/IR/TosaAttributes.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/Tosa/IR/TosaAttributes.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
(
["-gen-op-doc"],
"g3doc/Dialects/Tosa/TosaOps.md",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Tosa/IR/TosaOps.td",
deps = [":TosaDialectTdFiles"],
)
gentbl_cc_library(
name = "TosaInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Dialect/Tosa/IR/TosaInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Dialect/Tosa/IR/TosaInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Tosa/IR/TosaInterfaces.td",
deps = [":TosaDialectTdFiles"],
)
gentbl_cc_library(
name = "TosaPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=TosaOpt",
],
"include/mlir/Dialect/Tosa/Transforms/Passes.h.inc",
),
Update bazel files.
2022-11-15 07:48:51 +01:00
(
[
"-gen-enum-decls",
"-name=TosaOpt",
],
"include/mlir/Dialect/Tosa/Transforms/PassesEnums.h.inc",
),
(
[
"-gen-enum-defs",
"-name=TosaOpt",
],
"include/mlir/Dialect/Tosa/Transforms/PassesEnums.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Tosa/Transforms/Passes.td",
Update bazel files.
2022-11-15 07:48:51 +01:00
deps = [
":OpBaseTdFiles",
":PassBaseTdFiles",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
cc_library(
name = "TosaDialect",
srcs = glob([
"lib/Dialect/Tosa/IR/*.cpp",
"lib/Dialect/Tosa/IR/*.h",
"lib/Dialect/Tosa/Utils/*.cpp",
"lib/Dialect/Tosa/Transforms/*.cpp",
]),
hdrs = glob([
"include/mlir/Dialect/Tosa/IR/*.h",
"include/mlir/Dialect/Tosa/Utils/*.h",
"include/mlir/Dialect/Tosa/Transforms/*.h",
]),
includes = ["include"],
deps = [
[Bazel] Fix build for c0a6318d96 This adds explicit deps to satisfy layering_check Differential Revision: https://reviews.llvm.org/D105356
2021-07-02 09:09:59 -07:00
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Dialect",
[tosa][mlir] Support dynamic batch dimension for ops where the batch dim is explicit Dynamic batch for rescale, gather, max_pool, avg_pool, conv2D and depthwise_conv2D. Split helper functions into a separate header file. Reviewed By: rsuderman Differential Revision: https://reviews.llvm.org/D117031
2022-01-12 14:10:27 -08:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir][tosa] Update Bazel files for TOSA pass changes There were some missing bazel dependencies for the Tosa dialect. Added these deps. Reviewed By: GMNGeoffrey Differential Revision: https://reviews.llvm.org/D105326
2021-07-01 17:44:52 -07:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LoopLikeInterface",
":Pass",
":QuantOps",
Fix bazel build after https://reviews.llvm.org/D150578.
2023-07-05 14:32:42 +02:00
":Support",
Update Bazel BUILD files up to be9a87fe9b Differential Revision: https://reviews.llvm.org/D104791
2021-06-23 09:25:36 -07:00
":TensorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TosaDialectIncGen",
":TosaInterfacesIncGen",
":TosaPassIncGen",
":TransformUtils",
[bazel] build fix Differential Revision: https://reviews.llvm.org/D143973
2023-02-13 19:27:55 -08:00
":VectorInterfaces",
[mlir][Bazel] Adjust BUILD.bazel file. The dependency is needed after 1b00b94ffc2d60 Differential Revision: https://reviews.llvm.org/D107426
2021-08-04 08:49:30 +02:00
"//llvm:Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
[tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard Apply scale should be optionally disabled when lowering via TosaToStandard. In most cases it should persist until the lowering to specific backend. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D122948
2022-04-04 12:22:09 -07:00
cc_library(
name = "TosaToArith",
srcs = glob([
"lib/Conversion/TosaToArith/*.cpp",
"lib/Conversion/TosaToArith/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
[tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard Apply scale should be optionally disabled when lowering via TosaToStandard. In most cases it should persist until the lowering to specific backend. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D122948
2022-04-04 12:22:09 -07:00
hdrs = glob([
"include/mlir/Conversion/TosaToArith/*.h",
]),
includes = [
"include",
"lib/Conversion/TosaToArith",
],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard Apply scale should be optionally disabled when lowering via TosaToStandard. In most cases it should persist until the lowering to specific backend. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D122948
2022-04-04 12:22:09 -07:00
":ConversionPassIncGen",
":FuncDialect",
":IR",
":Pass",
":TosaDialect",
":Transforms",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "TosaToLinalg",
srcs = glob([
"lib/Conversion/TosaToLinalg/*.cpp",
"lib/Conversion/TosaToLinalg/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/TosaToLinalg/*.h",
]),
includes = [
"include",
"lib/Conversion/TosaToLinalg",
],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
add missing dependency for TosaToLinalg
2023-05-31 01:37:14 +00:00
":ArithUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[Bazel] Fixes for b5d847b1b95750d0af40cfc8c71a8fec50bb8613 and 6412a13539ab2914eed8e1df83c399b9a16e3408
2021-07-07 16:50:23 -07:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[mlir][NFC] Rename Bazel target aliases and consolidate targets This patch completes outstanding TODOs of removing aliases bazel target names. This patch also renames and cosolidates some bazel targets to be more in line with their CMake counterparts, e.g. combining `:LinalgOps` and `:LinalgInterfaces` into `:LinalgDialect`. Differential Revision: https://reviews.llvm.org/D127459
2022-06-09 23:34:57 +00:00
":LinalgDialect",
add missing dependency for TosaToLinalg
2023-05-31 01:37:14 +00:00
":LinalgUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MathDialect",
":Pass",
[mlir][tosa] Add lowering for tosa.clz using scf::whileOp Reviewed By: rsuderman Differential Revision: https://reviews.llvm.org/D109540
2021-09-09 15:57:22 -07:00
":SCFDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TensorDialect",
[mlir] Move linalg::PadTensorOp to tensor::PadOp. RFC: https://llvm.discourse.group/t/rfc-move-linalg-padtensorop-to-tensor-padop/5785 Differential Revision: https://reviews.llvm.org/D117892
2022-01-21 19:29:08 +01:00
":TensorUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TosaDialect",
":Transforms",
[bazel] Add missing dependency for b2d76a063dd7fb681c98a10d8e7f54fd6d25dd27
2023-07-22 13:21:51 +02:00
"//llvm:Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
cc_library(
name = "TosaToSCF",
srcs = glob([
"lib/Conversion/TosaToSCF/*.cpp",
"lib/Conversion/TosaToSCF/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/TosaToSCF/*.h",
]),
includes = [
"include",
"lib/Conversion/TosaToSCF",
],
deps = [
":ConversionPassIncGen",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":Pass",
":SCFDialect",
":TensorDialect",
":TosaDialect",
":Transforms",
],
)
cc_library(
[tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard Apply scale should be optionally disabled when lowering via TosaToStandard. In most cases it should persist until the lowering to specific backend. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D122948
2022-04-04 12:22:09 -07:00
name = "TosaToTensor",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
srcs = glob([
[tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard Apply scale should be optionally disabled when lowering via TosaToStandard. In most cases it should persist until the lowering to specific backend. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D122948
2022-04-04 12:22:09 -07:00
"lib/Conversion/TosaToTensor/*.cpp",
"lib/Conversion/TosaToTensor/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
[tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard Apply scale should be optionally disabled when lowering via TosaToStandard. In most cases it should persist until the lowering to specific backend. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D122948
2022-04-04 12:22:09 -07:00
"include/mlir/Conversion/TosaToTensor/*.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
]),
includes = [
"include",
[tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard Apply scale should be optionally disabled when lowering via TosaToStandard. In most cases it should persist until the lowering to specific backend. Reviewed By: jpienaar Differential Revision: https://reviews.llvm.org/D122948
2022-04-04 12:22:09 -07:00
"lib/Conversion/TosaToTensor",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[Bazel][mlir][tosa] Fix for 86c4972f5f6759b0ba85c568d934f9b8db8875a2
2023-06-15 19:50:08 +00:00
":ArithUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":Pass",
":TensorDialect",
[Bazel][mlir][tosa] Fix for 86c4972f5f6759b0ba85c568d934f9b8db8875a2
2023-06-15 19:50:08 +00:00
":TensorUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TosaDialect",
":Transforms",
],
)
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
td_library(
name = "TransformDialectTdFiles",
srcs = glob(["include/mlir/Dialect/Transform/IR/*.td"]),
deps = [
[mlir] add types to the transform dialect Introduce a type system for the transform dialect. A transform IR type captures the expectations of the transform IR on the payload IR operations that are being transformed, such as being of a certain kind or implementing an interface that enables the transformation. This provides stricter checking and better readability of the transform IR than using the catch-all "handle" type. This change implements the basic support for a type system amendable to dialect extensions and adds a drop-in replacement for the unrestricted "handle" type. The actual switch of transform dialect ops to that type will happen in a separate commit. See https://discourse.llvm.org/t/rfc-type-system-for-the-transform-dialect/65702 Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D135164
2022-10-04 11:49:21 +00:00
":CastInterfacesTdFiles",
[mlir] Use-after-free checker for the Transform dialect The Transform dialect uses the side effect modeling mechanism to record the effects of the transform ops on the mapping between Transform IR values and Payload IR ops. Introduce a checker pass that warns if a Transform IR value is used after it has been freed (consumed). This pass is mostly intended as a debugging aid in addition to the verification/assertion mechanisms in the transform interpreter. It reports all potential use-after-free situations. The implementation makes a series of simplifying assumptions to be simple and conservative. A more advanced implementation would rely on the data flow-like analysis associated with a side-effect resource rather than a value, which is currently not supported by the analysis infrastructure. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D126381
2022-05-25 15:57:22 +02:00
":ControlFlowInterfacesTdFiles",
[mlir] switch transform dialect ops to use TransformTypeInterface Use the recently introduced TransformTypeInterface instead of hardcoding the PDLOperationType. This will allow the operations to use more specific transform types to express pre/post-conditions in the future. It requires the syntax and Python op construction API to be updated. Dialect extensions will be switched separately. Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D135584
2022-10-05 14:23:19 +00:00
":InferTypeOpInterfaceTdFiles",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
":OpBaseTdFiles",
[mlir] use side effects in the Transform dialect Currently, the sequence of Transform dialect operations only supports a single use of each operand (verified by the `transform.sequence` operation). This was originally motivated by the need to guard against accessing a payload IR operation associated with a transform IR value after this operation has likely been rewritten by a transformation. However, not all Transform dialect operations rewrite payload IR, in particular the "navigation" operation such as `transform.pdl_match` do not. Introduce memory effects to the Transform dialect operations to describe their effect on the payload IR and the mapping between payload IR opreations and transform IR values. Use these effects to replace the single-use rule, allowing repeated reads and disallowing use-after-free, where operations with the "free" effect are considered to "consume" the transform IR value and rewrite the corresponding payload IR operations). As an additional improvement, this enables code motion transformation on the transform IR itself. Reviewed By: Mogball Differential Revision: https://reviews.llvm.org/D124181
2022-04-22 10:26:53 +02:00
":SideEffectInterfacesTdFiles",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
],
)
[mlir][transform] failure propagation mode in sequence Introduce two different failure propagation mode in the Transform dialect's Sequence operation. These modes specify whether silenceable errors produced by nested ops are immediately propagated, thus stopping the sequence, or suppressed. The latter is useful in end-to-end transform application scenarios where the user cannot correct the transformation, but it is robust enough to silenceable failures. It can be combined with the "alternatives" operation. There is intentionally no default value to avoid favoring one mode over the other. Downstreams can update their tests using: S='s/sequence \(%.*\) {/sequence \1 failures(propagate) {/' T='s/sequence {/sequence failures(propagate) {/' git grep -l transform.sequence | xargs sed -i -e "$S" git grep -l transform.sequence | xargs sed -i -e "$T" Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D131774
2022-08-12 13:53:46 +00:00
gentbl_cc_library(
name = "TransformDialectEnumsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-enum-decls",
],
"include/mlir/Dialect/Transform/IR/TransformDialectEnums.h.inc",
),
(
[
"-gen-enum-defs",
],
"include/mlir/Dialect/Transform/IR/TransformDialectEnums.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Transform/IR/TransformAttrs.td",
deps = [":TransformDialectTdFiles"],
)
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
gentbl_cc_library(
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
name = "TransformDialectMatchInterfacesIncGen",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-op-interface-decls",
],
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
"include/mlir/Dialect/Transform/IR/MatchInterfaces.h.inc",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
),
(
[
"-gen-op-interface-defs",
],
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
"include/mlir/Dialect/Transform/IR/MatchInterfaces.cpp.inc",
[mlir] add types to the transform dialect Introduce a type system for the transform dialect. A transform IR type captures the expectations of the transform IR on the payload IR operations that are being transformed, such as being of a certain kind or implementing an interface that enables the transformation. This provides stricter checking and better readability of the transform IR than using the catch-all "handle" type. This change implements the basic support for a type system amendable to dialect extensions and adds a drop-in replacement for the unrestricted "handle" type. The actual switch of transform dialect ops to that type will happen in a separate commit. See https://discourse.llvm.org/t/rfc-type-system-for-the-transform-dialect/65702 Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D135164
2022-10-04 11:49:21 +00:00
),
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
],
tblgen = ":mlir-tblgen",
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
td_file = "include/mlir/Dialect/Transform/IR/MatchInterfaces.td",
deps = [
":TransformDialectInterfacesIncGen",
":TransformDialectTdFiles",
],
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
)
[mlir] Add transform.foreach_match Add a new transform op combinator that implements an "if-then-else" style of mechanism for applying transformations. Its main purpose is to serve as a higher-level driver when applying multiple transform scripts to potentially overlapping pieces of the payload IR. This is similar to how the various rewrite drivers operate in C++, but at a higher level and with more declarative expressions. This is not intended to replace existing pattern-based rewrites, but to to drive more complex transformations that are exposed in the transform dialect and are too complex to be expressed as simple declarative rewrites. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D148013
2023-04-03 12:59:49 +00:00
gentbl_cc_library(
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
name = "TransformDialectInterfacesIncGen",
[mlir] Add transform.foreach_match Add a new transform op combinator that implements an "if-then-else" style of mechanism for applying transformations. Its main purpose is to serve as a higher-level driver when applying multiple transform scripts to potentially overlapping pieces of the payload IR. This is similar to how the various rewrite drivers operate in C++, but at a higher level and with more declarative expressions. This is not intended to replace existing pattern-based rewrites, but to to drive more complex transformations that are exposed in the transform dialect and are too complex to be expressed as simple declarative rewrites. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D148013
2023-04-03 12:59:49 +00:00
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-op-interface-decls",
],
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
"include/mlir/Dialect/Transform/IR/TransformInterfaces.h.inc",
[mlir] Add transform.foreach_match Add a new transform op combinator that implements an "if-then-else" style of mechanism for applying transformations. Its main purpose is to serve as a higher-level driver when applying multiple transform scripts to potentially overlapping pieces of the payload IR. This is similar to how the various rewrite drivers operate in C++, but at a higher level and with more declarative expressions. This is not intended to replace existing pattern-based rewrites, but to to drive more complex transformations that are exposed in the transform dialect and are too complex to be expressed as simple declarative rewrites. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D148013
2023-04-03 12:59:49 +00:00
),
(
[
"-gen-op-interface-defs",
],
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
"include/mlir/Dialect/Transform/IR/TransformInterfaces.cpp.inc",
),
(
[
"-gen-type-interface-decls",
],
"include/mlir/Dialect/Transform/IR/TransformTypeInterfaces.h.inc",
),
(
[
"-gen-type-interface-defs",
],
"include/mlir/Dialect/Transform/IR/TransformTypeInterfaces.cpp.inc",
[mlir] Add transform.foreach_match Add a new transform op combinator that implements an "if-then-else" style of mechanism for applying transformations. Its main purpose is to serve as a higher-level driver when applying multiple transform scripts to potentially overlapping pieces of the payload IR. This is similar to how the various rewrite drivers operate in C++, but at a higher level and with more declarative expressions. This is not intended to replace existing pattern-based rewrites, but to to drive more complex transformations that are exposed in the transform dialect and are too complex to be expressed as simple declarative rewrites. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D148013
2023-04-03 12:59:49 +00:00
),
],
tblgen = ":mlir-tblgen",
[mlir] add structured (Linalg) transform op matchers Add a set of transform operations into the "structured" extension of the Transform dialect that allow one to select transformation targets more specifically than the currently available matching. In particular, add the mechanism for identifying the producers of operands (input and init in destination-passing style) and users of results, as well as mechanisms for reasoning about the shape of the iteration space. Additionally, add several transform operations to manipulate parameters that could be useful to implement more advanced selectors. Specifically, new operations let one produce and compare parameter values to implement shape-driven transformations. New operations are placed in separate files to decrease compilation time. Some relayering of the extension is necessary to avoid repeated generation of enums. Depends on D148013 Depends on D148014 Depends on D148015 Reviewed By: chelini Differential Revision: https://reviews.llvm.org/D148017
2023-04-11 14:06:06 +00:00
td_file = "include/mlir/Dialect/Transform/IR/TransformInterfaces.td",
deps = [":TransformDialectTdFiles"],
[mlir] Add transform.foreach_match Add a new transform op combinator that implements an "if-then-else" style of mechanism for applying transformations. Its main purpose is to serve as a higher-level driver when applying multiple transform scripts to potentially overlapping pieces of the payload IR. This is similar to how the various rewrite drivers operate in C++, but at a higher level and with more declarative expressions. This is not intended to replace existing pattern-based rewrites, but to to drive more complex transformations that are exposed in the transform dialect and are too complex to be expressed as simple declarative rewrites. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D148013
2023-04-03 12:59:49 +00:00
)
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
gentbl_cc_library(
name = "TransformDialectIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
],
"include/mlir/Dialect/Transform/IR/TransformDialect.h.inc",
),
(
[
"-gen-dialect-defs",
],
"include/mlir/Dialect/Transform/IR/TransformDialect.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Transform/IR/TransformDialect.td",
deps = [":TransformDialectTdFiles"],
)
[mlir][transform] Introduce transform.sequence op Sequence is an important transform combination primitive that just indicates transform ops being applied in a row. The simplest version requires fails immediately if any transformation in the sequence fails. Introducing this operation allows one to start placing transform IR within other IR. Depends On D123135 Reviewed By: Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123664
2022-04-19 16:36:37 +02:00
gentbl_cc_library(
name = "TransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Transform/IR/TransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Transform/IR/TransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Transform/IR/TransformOps.td",
[mlir] Transform dialect: add named sequences Named sequences introduce an additional abstraction and reuse capability to the transform dialect. They can be though of as macros parameterized with handles that can be invoked in places where a transform dialect operation is expected. Such reuse was previously not possible in the dialect and required dynamic construction of the transform IR from the client language. Named sequences are intentionally restricted to disallow recursion, as it could make the dialect accidentally Turing-complete, which isn't desired at this point. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D146433
2023-03-20 16:18:35 +00:00
deps = [
":CallInterfacesTdFiles",
Add missing dependency for TransformDialect Differential Revision: https://reviews.llvm.org/D146589
2023-03-21 15:23:13 -07:00
":FunctionInterfacesTdFiles",
Fix dep error for transform dialect in bazel Fix dep error for transform dialect in bazel Reviewed By: anlunx Differential Revision: https://reviews.llvm.org/D146566
2023-03-21 14:40:17 -07:00
":TransformDialectTdFiles",
[mlir] Transform dialect: add named sequences Named sequences introduce an additional abstraction and reuse capability to the transform dialect. They can be though of as macros parameterized with handles that can be invoked in places where a transform dialect operation is expected. Such reuse was previously not possible in the dialect and required dynamic construction of the transform IR from the client language. Named sequences are intentionally restricted to disallow recursion, as it could make the dialect accidentally Turing-complete, which isn't desired at this point. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D146433
2023-03-20 16:18:35 +00:00
],
[mlir][transform] Introduce transform.sequence op Sequence is an important transform combination primitive that just indicates transform ops being applied in a row. The simplest version requires fails immediately if any transformation in the sequence fails. Introducing this operation allows one to start placing transform IR within other IR. Depends On D123135 Reviewed By: Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123664
2022-04-19 16:36:37 +02:00
)
[mlir] add types to the transform dialect Introduce a type system for the transform dialect. A transform IR type captures the expectations of the transform IR on the payload IR operations that are being transformed, such as being of a certain kind or implementing an interface that enables the transformation. This provides stricter checking and better readability of the transform IR than using the catch-all "handle" type. This change implements the basic support for a type system amendable to dialect extensions and adds a drop-in replacement for the unrestricted "handle" type. The actual switch of transform dialect ops to that type will happen in a separate commit. See https://discourse.llvm.org/t/rfc-type-system-for-the-transform-dialect/65702 Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D135164
2022-10-04 11:49:21 +00:00
gentbl_cc_library(
name = "TransformTypesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-typedef-decls"],
"include/mlir/Dialect/Transform/IR/TransformTypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/Transform/IR/TransformTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Transform/IR/TransformTypes.td",
deps = [":TransformDialectTdFiles"],
)
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
cc_library(
name = "TransformDialect",
srcs = glob(["lib/Dialect/Transform/IR/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/Transform/IR/*.h"]),
deps = [
Add missing dependency for TransformDialect Differential Revision: https://reviews.llvm.org/D146589
2023-03-21 15:23:13 -07:00
":Analysis",
Fix dep error for transform dialect in bazel Fix dep error for transform dialect in bazel Reviewed By: anlunx Differential Revision: https://reviews.llvm.org/D146566
2023-03-21 14:40:17 -07:00
":CallOpInterfaces",
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfaces",
[mlir] Use-after-free checker for the Transform dialect The Transform dialect uses the side effect modeling mechanism to record the effects of the transform ops on the mapping between Transform IR values and Payload IR ops. Introduce a checker pass that warns if a Transform IR value is used after it has been freed (consumed). This pass is mostly intended as a debugging aid in addition to the verification/assertion mechanisms in the transform interpreter. It reports all potential use-after-free situations. The implementation makes a series of simplifying assumptions to be simple and conservative. A more advanced implementation would rely on the data flow-like analysis associated with a side-effect resource rather than a value, which is currently not supported by the analysis infrastructure. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D126381
2022-05-25 15:57:22 +02:00
":ControlFlowInterfaces",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
":IR",
[mlir][transform] Add `transform.apply_licm` op This op applies loop-invariant code motion to the targeted loop-like op. Differential Revision: https://reviews.llvm.org/D154327
2023-07-03 15:24:35 +02:00
":LoopLikeInterface",
[mlir][transform] Add ApplyRegisteredPassOp transform op This transform op runs a pass on the target op. Differential Revision: https://reviews.llvm.org/D153143
2023-06-20 08:54:49 +02:00
":Pass",
[mlir] Connect Transform dialect to PDL This introduces a pair of ops to the Transform dialect that connect it to PDL patterns. Transform dialect relies on PDL for matching the Payload IR ops that are about to be transformed. For this purpose, it provides a container op for patterns, a "pdl_match" op and transform interface implementations that call into the pattern matching infrastructure. To enable the caching of compiled patterns, this also provides the extension mechanism for TransformState. Extensions allow one to store additional information in the TransformState and thus communicate it between different Transform dialect operations when they are applied. They can be added and removed when applying transform ops. An extension containing a symbol table in which the pattern names are resolved and a pattern compilation cache is introduced as the first client. Depends On D123664 Reviewed By: Mogball Differential Revision: https://reviews.llvm.org/D124007
2022-04-20 12:57:23 +02:00
":Rewrite",
[mlir] use side effects in the Transform dialect Currently, the sequence of Transform dialect operations only supports a single use of each operand (verified by the `transform.sequence` operation). This was originally motivated by the need to guard against accessing a payload IR operation associated with a transform IR value after this operation has likely been rewritten by a transformation. However, not all Transform dialect operations rewrite payload IR, in particular the "navigation" operation such as `transform.pdl_match` do not. Introduce memory effects to the Transform dialect operations to describe their effect on the payload IR and the mapping between payload IR opreations and transform IR values. Use these effects to replace the single-use rule, allowing repeated reads and disallowing use-after-free, where operations with the "free" effect are considered to "consume" the transform IR value and rewrite the corresponding payload IR operations). As an additional improvement, this enables code motion transformation on the transform IR itself. Reviewed By: Mogball Differential Revision: https://reviews.llvm.org/D124181
2022-04-22 10:26:53 +02:00
":SideEffectInterfaces",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
":Support",
[mlir][transform] failure propagation mode in sequence Introduce two different failure propagation mode in the Transform dialect's Sequence operation. These modes specify whether silenceable errors produced by nested ops are immediately propagated, thus stopping the sequence, or suppressed. The latter is useful in end-to-end transform application scenarios where the user cannot correct the transformation, but it is robust enough to silenceable failures. It can be combined with the "alternatives" operation. There is intentionally no default value to avoid favoring one mode over the other. Downstreams can update their tests using: S='s/sequence \(%.*\) {/sequence \1 failures(propagate) {/' T='s/sequence {/sequence failures(propagate) {/' git grep -l transform.sequence | xargs sed -i -e "$S" git grep -l transform.sequence | xargs sed -i -e "$T" Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D131774
2022-08-12 13:53:46 +00:00
":TransformDialectEnumsIncGen",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
":TransformDialectIncGen",
":TransformDialectInterfacesIncGen",
[mlir] Add transform.foreach_match Add a new transform op combinator that implements an "if-then-else" style of mechanism for applying transformations. Its main purpose is to serve as a higher-level driver when applying multiple transform scripts to potentially overlapping pieces of the payload IR. This is similar to how the various rewrite drivers operate in C++, but at a higher level and with more declarative expressions. This is not intended to replace existing pattern-based rewrites, but to to drive more complex transformations that are exposed in the transform dialect and are too complex to be expressed as simple declarative rewrites. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D148013
2023-04-03 12:59:49 +00:00
":TransformDialectMatchInterfacesIncGen",
[mlir] NFC: move DiagnosedSilenceableFailure to Utils in Transform dialect It was originally placed in TransformInterfaces for convenience, but it is really a generic utility. It may also create an include cycle between TransformTypes and TransformInterfaces if the latter needs to include the former because the former uses the failure util. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D140978
2023-01-02 17:11:40 +00:00
":TransformDialectUtils",
[mlir][transform] Introduce transform.sequence op Sequence is an important transform combination primitive that just indicates transform ops being applied in a row. The simplest version requires fails immediately if any transformation in the sequence fails. Introducing this operation allows one to start placing transform IR within other IR. Depends On D123135 Reviewed By: Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123664
2022-04-19 16:36:37 +02:00
":TransformOpsIncGen",
[mlir] add types to the transform dialect Introduce a type system for the transform dialect. A transform IR type captures the expectations of the transform IR on the payload IR operations that are being transformed, such as being of a certain kind or implementing an interface that enables the transformation. This provides stricter checking and better readability of the transform IR than using the catch-all "handle" type. This change implements the basic support for a type system amendable to dialect extensions and adds a drop-in replacement for the unrestricted "handle" type. The actual switch of transform dialect ops to that type will happen in a separate commit. See https://discourse.llvm.org/t/rfc-type-system-for-the-transform-dialect/65702 Reviewed By: nicolasvasilache Differential Revision: https://reviews.llvm.org/D135164
2022-10-04 11:49:21 +00:00
":TransformTypesIncGen",
[bazel] Fix mlir buildifier issues
2023-03-28 13:11:47 -07:00
":Transforms",
[mlir] Introduce Transform dialect This dialect provides operations that can be used to control transformation of the IR using a different portion of the IR. It refers to the IR being transformed as payload IR, and to the IR guiding the transformation as transform IR. The main use case for this dialect is orchestrating fine-grain transformations on individual operations or sets thereof. For example, it may involve finding loop-like operations with specific properties (e.g., large size) in the payload IR, applying loop tiling to those and only those operations, and then applying loop unrolling to the inner loops produced by the previous transformations. As such, it is not intended as a replacement for the pass infrastructure, nor for the pattern rewriting infrastructure. In the most common case, the transform IR will be processed and applied to payload IR by a pass. Transformations expressed by the transform dialect may be implemented using the pattern infrastructure or any other relevant MLIR component. This dialect is designed to be extensible, that is, clients of this dialect are allowed to inject additional operations into this dialect using the newly introduced in this patch `TransformDialectExtension` mechanism. This allows the dialect to avoid a dependency on the implementation of the transformation as well as to avoid introducing dialect-specific transform dialects. See https://discourse.llvm.org/t/rfc-interfaces-and-dialects-for-precise-ir-transformation-control/60927. Reviewed By: nicolasvasilache, Mogball, rriddle Differential Revision: https://reviews.llvm.org/D123135
2022-04-14 13:16:48 +02:00
"//llvm:Support",
],
)
[mlir] move PDL-related transform ops into an extension The initial bring-up of the Transform dialect relied on PDL to provide the default handle type (`!pdl.operation`) and the matching capability. Both are now provided natively by the Transform dialect removing the reason to have a hard dependency on the PDL dialect and its interpreter. Move PDL-related transform operations into a separate extension. This requires us to introduce a dialect state extension mechanism into the Transform dialect so it no longer needs to know about PDL constraint functions that may be injected by extensions similarly to operations and types. This mechanism will be reused to connect pattern application drivers and the Transform dialect. This completes the restructuring of the Transform dialect to remove overrilance on PDL. Note to downstreams: flow that are using `!pdl.operation` with Transform dialect operations will now require `transform::PDLExtension` to be applied to the transform dialect in order to provide the transform handle type interface for `!pdl.operation`. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D151104
2023-05-22 14:36:58 +00:00
td_library(
name = "TransformPDLExtensionTdFiles",
srcs = glob(["include/mlir/Dialect/Transform/PDLExtension/*.td"]),
deps = [
":PDLDialectTdFiles",
":TransformDialectTdFiles",
],
)
gentbl_cc_library(
name = "TransformPDLExtensionOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-op-decls",
],
"include/mlir/Dialect/Transform/PDLExtension/PDLExtensionOps.h.inc",
),
(
[
"-gen-op-defs",
],
"include/mlir/Dialect/Transform/PDLExtension/PDLExtensionOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Transform/PDLExtension/PDLExtensionOps.td",
deps = [":TransformPDLExtensionTdFiles"],
)
cc_library(
name = "TransformPDLExtension",
srcs = glob(["lib/Dialect/Transform/PDLExtension/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/Transform/PDLExtension/*.h"]),
deps = [
":IR",
":PDLDialect",
":PDLInterpDialect",
[MLIR] Reformat the Bazel build This patch normalizes formatting of the the root BUILD.bazel file by: 1) adjusting indentation a little bit, 2) alphabetically ordering dependencies. These small deviations were introduced by some yesterday's patches: * https://reviews.llvm.org/D151104 * https://reviews.llvm.org/D151346 * https://reviews.llvm.org/rG16fe2b37365c00b0c6d0ed22c2e6521f2d5de01a * https://reviews.llvm.org/rG4d1cd1d8caab13d6b76ce6fc4ff76a01a7931c34 Reviewed By: aartbik Differential Revision: https://reviews.llvm.org/D151499
2023-05-26 14:16:47 -07:00
":Rewrite",
[mlir] move PDL-related transform ops into an extension The initial bring-up of the Transform dialect relied on PDL to provide the default handle type (`!pdl.operation`) and the matching capability. Both are now provided natively by the Transform dialect removing the reason to have a hard dependency on the PDL dialect and its interpreter. Move PDL-related transform operations into a separate extension. This requires us to introduce a dialect state extension mechanism into the Transform dialect so it no longer needs to know about PDL constraint functions that may be injected by extensions similarly to operations and types. This mechanism will be reused to connect pattern application drivers and the Transform dialect. This completes the restructuring of the Transform dialect to remove overrilance on PDL. Note to downstreams: flow that are using `!pdl.operation` with Transform dialect operations will now require `transform::PDLExtension` to be applied to the transform dialect in order to provide the transform handle type interface for `!pdl.operation`. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D151104
2023-05-22 14:36:58 +00:00
":SideEffectInterfaces",
":Support",
":TransformDialect",
":TransformPDLExtensionOpsIncGen",
"//llvm:Support",
],
)
[mlir] Use-after-free checker for the Transform dialect The Transform dialect uses the side effect modeling mechanism to record the effects of the transform ops on the mapping between Transform IR values and Payload IR ops. Introduce a checker pass that warns if a Transform IR value is used after it has been freed (consumed). This pass is mostly intended as a debugging aid in addition to the verification/assertion mechanisms in the transform interpreter. It reports all potential use-after-free situations. The implementation makes a series of simplifying assumptions to be simple and conservative. A more advanced implementation would rely on the data flow-like analysis associated with a side-effect resource rather than a value, which is currently not supported by the analysis infrastructure. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D126381
2022-05-25 15:57:22 +02:00
td_library(
name = "TransformDialectTransformsTdFiles",
srcs = glob(["include/mlir/Dialect/Transform/Transforms/*.td"]),
deps = [
":PassBaseTdFiles",
":TransformDialectTdFiles",
],
)
gentbl_cc_library(
name = "TransformDialectTransformsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Transform",
],
"include/mlir/Dialect/Transform/Transforms/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Transform/Transforms/Passes.td",
deps = [":TransformDialectTransformsTdFiles"],
)
cc_library(
name = "TransformDialectTransforms",
srcs = glob(["lib/Dialect/Transform/Transforms/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/Transform/Transforms/*.h"]),
deps = [
":Analysis",
":IR",
[mlir] provide a base class for transform interpreter passes The transform dialect infrastructure does not provide a default interpreter pass and instead expects users to create their own to ensure all relevant extensions and dependent dialects are loaded. Provide a base class for implementing such passes that includes the additional facilities for debugging and is aware of the multithreaded nature of pass execution. Reviewed By: pifon2a, nicolasvasilache Differential Revision: https://reviews.llvm.org/D142729
2023-01-27 16:01:26 +00:00
":Parser",
[mlir] Use-after-free checker for the Transform dialect The Transform dialect uses the side effect modeling mechanism to record the effects of the transform ops on the mapping between Transform IR values and Payload IR ops. Introduce a checker pass that warns if a Transform IR value is used after it has been freed (consumed). This pass is mostly intended as a debugging aid in addition to the verification/assertion mechanisms in the transform interpreter. It reports all potential use-after-free situations. The implementation makes a series of simplifying assumptions to be simple and conservative. A more advanced implementation would rely on the data flow-like analysis associated with a side-effect resource rather than a value, which is currently not supported by the analysis infrastructure. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D126381
2022-05-25 15:57:22 +02:00
":Pass",
":SideEffectInterfaces",
[mlir] provide a base class for transform interpreter passes The transform dialect infrastructure does not provide a default interpreter pass and instead expects users to create their own to ensure all relevant extensions and dependent dialects are loaded. Provide a base class for implementing such passes that includes the additional facilities for debugging and is aware of the multithreaded nature of pass execution. Reviewed By: pifon2a, nicolasvasilache Differential Revision: https://reviews.llvm.org/D142729
2023-01-27 16:01:26 +00:00
":Support",
[mlir] Use-after-free checker for the Transform dialect The Transform dialect uses the side effect modeling mechanism to record the effects of the transform ops on the mapping between Transform IR values and Payload IR ops. Introduce a checker pass that warns if a Transform IR value is used after it has been freed (consumed). This pass is mostly intended as a debugging aid in addition to the verification/assertion mechanisms in the transform interpreter. It reports all potential use-after-free situations. The implementation makes a series of simplifying assumptions to be simple and conservative. A more advanced implementation would rely on the data flow-like analysis associated with a side-effect resource rather than a value, which is currently not supported by the analysis infrastructure. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D126381
2022-05-25 15:57:22 +02:00
":TransformDialect",
":TransformDialectTransformsIncGen",
"//llvm:Support",
],
)
[mlir][transform] Add PackedOrDynamicIndexList helper This customer parser/printer is similar to DynamicIndexList, but has special syntax for the case where one handle represents the entire list. Example: ``` // Regular index list [10, 20, %val] // Packed handle (no square parentheses) %val ``` Differential Revision: https://reviews.llvm.org/D138825
2022-12-19 07:57:46 +01:00
cc_library(
name = "TransformDialectUtils",
[mlir] NFC: move DiagnosedSilenceableFailure to Utils in Transform dialect It was originally placed in TransformInterfaces for convenience, but it is really a generic utility. It may also create an include cycle between TransformTypes and TransformInterfaces if the latter needs to include the former because the former uses the failure util. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D140978
2023-01-02 17:11:40 +00:00
srcs = glob(["lib/Dialect/Transform/Utils/*cpp"]),
hdrs = glob(["include/mlir/Dialect/Transform/Utils/*.h"]),
[mlir][transform] Add PackedOrDynamicIndexList helper This customer parser/printer is similar to DynamicIndexList, but has special syntax for the case where one handle represents the entire list. Example: ``` // Regular index list [10, 20, %val] // Packed handle (no square parentheses) %val ``` Differential Revision: https://reviews.llvm.org/D138825
2022-12-19 07:57:46 +01:00
includes = ["include"],
deps = [
":DialectUtils",
":IR",
":Support",
":ViewLikeInterface",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "ComplexOpsTdFiles",
srcs = [
"include/mlir/Dialect/Complex/IR/ComplexBase.td",
"include/mlir/Dialect/Complex/IR/ComplexOps.td",
],
includes = ["include"],
deps = [
[mlir][bazel] Fix bazel build files This adds a missing dependency on builtin dialect td files. Differential Revision: https://reviews.llvm.org/D130890
2022-08-01 11:40:39 +02:00
":BuiltinDialectTdFiles",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
[bazel] Port 5caa941f683270c3e5078ec46f208212b565c505
2022-07-27 16:12:58 +02:00
gentbl_cc_library(
name = "ComplexAttributesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/Complex/IR/ComplexAttributes.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/Complex/IR/ComplexAttributes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Complex/IR/ComplexAttributes.td",
deps = [":ComplexOpsTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "ComplexBaseIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=complex",
],
"include/mlir/Dialect/Complex/IR/ComplexOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=complex",
],
"include/mlir/Dialect/Complex/IR/ComplexOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Complex/IR/ComplexBase.td",
deps = [":ComplexOpsTdFiles"],
)
gentbl_cc_library(
name = "ComplexOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Complex/IR/ComplexOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Complex/IR/ComplexOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Complex/IR/ComplexOps.td",
deps = [":ComplexOpsTdFiles"],
)
cc_library(
name = "ComplexDialect",
srcs = glob(
[
"lib/Dialect/Complex/IR/*.cpp",
"lib/Dialect/Complex/IR/*.h",
],
),
hdrs = ["include/mlir/Dialect/Complex/IR/Complex.h"],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[bazel] Port 5caa941f683270c3e5078ec46f208212b565c505
2022-07-27 16:12:58 +02:00
":ComplexAttributesIncGen",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ComplexBaseIncGen",
":ComplexOpsIncGen",
":IR",
[bazel] Port 480cd4cb
2022-01-26 23:24:45 +01:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfaces",
"//llvm:Support",
],
)
cc_library(
name = "ComplexToLLVM",
srcs = glob([
"lib/Conversion/ComplexToLLVM/*.cpp",
"lib/Conversion/ComplexToLLVM/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/ComplexToLLVM/*.h",
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ComplexDialect",
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[Bazel] Fixes for b5d847b1b95750d0af40cfc8c71a8fec50bb8613 and 6412a13539ab2914eed8e1df83c399b9a16e3408
2021-07-07 16:50:23 -07:00
":LLVMCommonConversion",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":LLVMDialect",
":Pass",
":Support",
":Transforms",
"//llvm:Core",
"//llvm:Support",
],
)
[mlir][complex] Add pow/sqrt/tanh ops and lowering to libm Lowering through libm gives us a baseline version, even though it's not going to be particularly fast. This is similar to what we do for some math dialect ops. Differential Revision: https://reviews.llvm.org/D125550
2022-05-13 17:12:11 +02:00
cc_library(
name = "ComplexToLibm",
srcs = glob([
"lib/Conversion/ComplexToLibm/*.cpp",
"lib/Conversion/ComplexToLibm/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
[mlir][complex] Add pow/sqrt/tanh ops and lowering to libm Lowering through libm gives us a baseline version, even though it's not going to be particularly fast. This is similar to what we do for some math dialect ops. Differential Revision: https://reviews.llvm.org/D125550
2022-05-13 17:12:11 +02:00
hdrs = glob([
"include/mlir/Conversion/ComplexToLibm/*.h",
]),
includes = ["include"],
deps = [
":ComplexDialect",
":ConversionPassIncGen",
":DialectUtils",
":FuncDialect",
":IR",
":Pass",
":Support",
":Transforms",
"//llvm:Core",
"//llvm:Support",
],
)
[mlir][spirv][complex] Convert complex ops to SPIR-V ops This commit adds conversion from complex construction and extraction ops to SPIR-V. Other arithemtic ops can be done via ComplexToStandard patterns. Reviewed By: kuhar Differential Revision: https://reviews.llvm.org/D147193
2023-03-30 08:49:57 -07:00
cc_library(
name = "ComplexToSPIRV",
srcs = glob([
"lib/Conversion/ComplexToSPIRV/*.cpp",
"lib/Conversion/ComplexToSPIRV/*.h",
]),
hdrs = glob([
"include/mlir/Conversion/ComplexToSPIRV/*.h",
]),
includes = ["include"],
deps = [
":ComplexDialect",
":ConversionPassIncGen",
":IR",
":Pass",
":SPIRVCommonConversion",
":SPIRVConversion",
":SPIRVDialect",
":Support",
":Transforms",
"//llvm:Core",
"//llvm:Support",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "ComplexToStandard",
srcs = glob([
"lib/Conversion/ComplexToStandard/*.cpp",
"lib/Conversion/ComplexToStandard/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/ComplexToStandard/*.h",
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ComplexDialect",
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":MathDialect",
":Pass",
":Transforms",
],
)
exports_files([
"include/mlir/Interfaces/CallInterfaces.h",
"include/mlir/Interfaces/CastInterfaces.h",
"include/mlir/Interfaces/ControlFlowInterfaces.h",
"include/mlir/Transforms/InliningUtils.h",
])
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
td_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithOpsTdFiles",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
srcs = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/IR/ArithBase.td",
"include/mlir/Dialect/Arith/IR/ArithOps.td",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
],
includes = ["include"],
deps = [
[bazel][mlir][arith] build fixes for arith fastmath flags
2022-10-26 17:45:25 -04:00
":ArithOpsInterfacesTdFiles",
[mlir][bazel] Fix bazel build files This adds a missing dependency on builtin dialect td files. Differential Revision: https://reviews.llvm.org/D130890
2022-08-01 11:40:39 +02:00
":BuiltinDialectTdFiles",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":CastInterfacesTdFiles",
[bazel] Port 75bfc6f29579
2022-06-14 23:45:52 +02:00
":InferIntRangeInterfaceTdFiles",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterfaceTdFiles",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
":VectorInterfacesTdFiles",
],
)
[bazel][mlir][arith] build fixes for arith fastmath flags
2022-10-26 17:45:25 -04:00
td_library(
name = "ArithOpsInterfacesTdFiles",
srcs = [
"include/mlir/Dialect/Arith/IR/ArithOpsInterfaces.td",
],
includes = ["include"],
deps = [
":OpBaseTdFiles",
],
)
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
gentbl_cc_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithBaseIncGen",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=arith",
],
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/IR/ArithOpsDialect.h.inc",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
),
(
[
"-gen-dialect-defs",
"-dialect=arith",
],
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/IR/ArithOpsDialect.cpp.inc",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
),
(
["-gen-enum-decls"],
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/IR/ArithOpsEnums.h.inc",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
),
(
["-gen-enum-defs"],
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/IR/ArithOpsEnums.cpp.inc",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
),
],
tblgen = ":mlir-tblgen",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
td_file = "include/mlir/Dialect/Arith/IR/ArithBase.td",
deps = [":ArithOpsTdFiles"],
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
)
gentbl_cc_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithOpsIncGen",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/IR/ArithOps.h.inc",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
),
(
["-gen-op-defs"],
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/IR/ArithOps.cpp.inc",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
),
[bazel][mlir][arith] build fixes for arith fastmath flags
2022-10-26 17:45:25 -04:00
(
[
"-gen-attrdef-decls",
"-attrdefs-dialect=arith",
],
"include/mlir/Dialect/Arith/IR/ArithOpsAttributes.h.inc",
),
(
[
"-gen-attrdef-defs",
"-attrdefs-dialect=arith",
],
"include/mlir/Dialect/Arith/IR/ArithOpsAttributes.cpp.inc",
),
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
],
tblgen = ":mlir-tblgen",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
td_file = "include/mlir/Dialect/Arith/IR/ArithOps.td",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithOpsTdFiles",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
],
)
[bazel][mlir][arith] build fixes for arith fastmath flags
2022-10-26 17:45:25 -04:00
gentbl_cc_library(
name = "ArithOpsInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Dialect/Arith/IR/ArithOpsInterfaces.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Dialect/Arith/IR/ArithOpsInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Arith/IR/ArithOpsInterfaces.td",
deps = [
":ArithOpsInterfacesTdFiles",
],
)
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
gentbl_cc_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithCanonicalizationIncGen",
strip_include_prefix = "include/mlir/Dialect/Arith/IR",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
tbl_outs = [
(
["-gen-rewriters"],
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/IR/ArithCanonicalization.inc",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
),
],
tblgen = ":mlir-tblgen",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
td_file = "lib/Dialect/Arith/IR/ArithCanonicalization.td",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithOpsTdFiles",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
":CastInterfacesTdFiles",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncTdFiles",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
],
)
cc_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithDialect",
[mlir][arith] Add ValueBoundsOpInterface impls These ops are useful for unit testing. (They do not fold/canonicalize with affine.apply etc.) Differential Revision: https://reviews.llvm.org/D145696
2023-04-06 10:46:48 +09:00
srcs = [
"lib/Dialect/Arith/IR/ArithDialect.cpp",
"lib/Dialect/Arith/IR/ArithOps.cpp",
"lib/Dialect/Arith/IR/InferIntRangeInterfaceImpls.cpp",
],
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
hdrs = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/IR/Arith.h",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
"include/mlir/Transforms/InliningUtils.h",
],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithBaseIncGen",
":ArithCanonicalizationIncGen",
":ArithOpsIncGen",
[bazel][mlir][arith] build fixes for arith fastmath flags
2022-10-26 17:45:25 -04:00
":ArithOpsInterfacesIncGen",
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfaces",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
":CommonFolders",
":IR",
Revert "Revert "[MLIR] Update bazel build file after 455305624884cf9237143e2ba0635fcc5ba5206"" This reverts commit 9ef7ae507387f133cf34ca0bc0edd3f009a3ff71.
2023-01-20 21:56:25 +01:00
":InferIntRangeCommon",
[bazel] Port 75bfc6f29579
2022-06-14 23:45:52 +02:00
":InferIntRangeInterface",
[bazel] Port ead1107257e6
2022-02-03 01:41:03 +01:00
":InferTypeOpInterface",
[MLIR] Split arith dialect from the std dialect Create the Arithmetic dialect that contains basic integer and floating point arithmetic operations. Ops that did not meet this criterion were moved to the Math dialect. First of two atomic patches to remove integer and floating point operations from the standard dialect. Ops will be removed from the standard dialect in a subsequent patch. Reviewed By: ftynse, silvas Differential Revision: https://reviews.llvm.org/D110200
2021-09-21 20:54:07 +00:00
":SideEffectInterfaces",
":VectorInterfaces",
"//llvm:Support",
],
)
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
gentbl_cc_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithPassIncGen",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"-name=Arith",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
],
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/Transforms/Passes.h.inc",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
),
],
tblgen = ":mlir-tblgen",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
td_file = "include/mlir/Dialect/Arith/Transforms/Passes.td",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
deps = [":PassBaseTdFiles"],
)
cc_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithTransforms",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
srcs = glob([
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"lib/Dialect/Arith/Transforms/*.cpp",
"lib/Dialect/Arith/Transforms/*.h",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
]),
[bazel] Port 7fa1d743d073
2022-09-09 09:31:13 +02:00
hdrs = glob([
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"include/mlir/Dialect/Arith/Transforms/*.h",
[bazel] Port 7fa1d743d073
2022-09-09 09:31:13 +02:00
]),
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
includes = ["include"],
deps = [
[bazel] Port b0b00432093b
2022-06-14 23:55:50 +02:00
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithPassIncGen",
":ArithUtils",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationDialect",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":BufferizationTransforms",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[bazel] Port 7fa1d743d073
2022-09-09 09:31:13 +02:00
":FuncTransforms",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":IR",
Revert "Fix bazel deps.". Original change has been reverted. This reverts commit b3999fd261805a2c8fc14f9421b7d9e8af251a13.
2022-09-08 22:11:44 +00:00
":InferIntRangeInterface",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
":MemRefDialect",
":Pass",
[bazel][mlir] Add missing dependency for D138178
2022-11-17 14:10:16 -05:00
":Support",
[mlir][Arith] ValueBoundsOpInterface: Reify with Arith ops This revision adds an additional `reifyValueBounds` helper that reifies the IR with Arith ops instead of Affine ops. This is needed to support value bounds for integer types different from `index` in a subsequent revision. Differential Revision: https://reviews.llvm.org/D146524
2023-04-18 16:44:44 +09:00
":TensorDialect",
[mlir][bufferize] Move arith BufferizableOpInterface impl to arith dialect Also switch the implementation of `-arith-bufferize` to BufferizableOpInterface. Differential Revision: https://reviews.llvm.org/D118325
2022-01-28 01:11:22 +09:00
":TransformUtils",
Revert "Fix bazel deps.". Original change has been reverted. This reverts commit b3999fd261805a2c8fc14f9421b7d9e8af251a13.
2022-09-08 22:11:44 +00:00
":Transforms",
[mlir][Arith] ValueBoundsOpInterface: Reify with Arith ops This revision adds an additional `reifyValueBounds` helper that reifies the IR with Arith ops instead of Affine ops. This is needed to support value bounds for integer types different from `index` in a subsequent revision. Differential Revision: https://reviews.llvm.org/D146524
2023-04-18 16:44:44 +09:00
":ValueBoundsOpInterface",
[bazel] Port 7fa1d743d073
2022-09-09 09:31:13 +02:00
":VectorDialect",
[mlir][bufferize] Move arith BufferizableOpInterface impl to arith dialect Also switch the implementation of `-arith-bufferize` to BufferizableOpInterface. Differential Revision: https://reviews.llvm.org/D118325
2022-01-28 01:11:22 +09:00
"//llvm:Support",
[MLIR] Replace std ops with arith dialect ops Precursor: https://reviews.llvm.org/D110200 Removed redundant ops from the standard dialect that were moved to the `arith` or `math` dialects. Renamed all instances of operations in the codebase and in tests. Reviewed By: rriddle, jpienaar Differential Revision: https://reviews.llvm.org/D110797
2021-10-12 23:14:57 +00:00
],
)
[bazel] Port ead1107257e6
2022-02-03 01:41:03 +01:00
cc_library(
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
name = "ArithUtils",
[bazel] Port ead1107257e6
2022-02-03 01:41:03 +01:00
srcs = glob([
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
"lib/Dialect/Arith/Utils/*.cpp",
[bazel] Port ead1107257e6
2022-02-03 01:41:03 +01:00
]),
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
hdrs = ["include/mlir/Dialect/Arith/Utils/Utils.h"],
[bazel] Port ead1107257e6
2022-02-03 01:41:03 +01:00
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[bazel] fix build of ArithUtils
2023-07-18 09:20:13 +02:00
":ComplexDialect",
[bazel] Port ead1107257e6
2022-02-03 01:41:03 +01:00
":IR",
[mlir][bazel] Update post ControlFlow dialect split
2022-02-06 17:21:48 -08:00
"//llvm:Support",
[bazel] Port ead1107257e6
2022-02-03 01:41:03 +01:00
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "MathOpsTdFiles",
srcs = [
"include/mlir/Dialect/Math/IR/MathBase.td",
"include/mlir/Dialect/Math/IR/MathOps.td",
],
includes = ["include"],
deps = [
[bazel] Port 589764a38264
2022-11-04 19:06:44 +01:00
":ArithOpsTdFiles",
[mlir] Move trait to InferTypeOpInterface Step towards removing the hard coded behavior for this trait and to instead use common interface. Differential Revision: https://reviews.llvm.org/D114208
2021-11-21 14:41:11 -08:00
":InferTypeOpInterfaceTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
":VectorInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "MathBaseIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=math",
],
"include/mlir/Dialect/Math/IR/MathOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=math",
],
"include/mlir/Dialect/Math/IR/MathOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Math/IR/MathBase.td",
deps = [":MathOpsTdFiles"],
)
gentbl_cc_library(
name = "MathOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Math/IR/MathOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Math/IR/MathOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Math/IR/MathOps.td",
deps = [":MathOpsTdFiles"],
)
[Bazel][mlir] Port ee8b8d6b58e1ff45d1bcbad38c6cf458f872b38d
2023-06-18 17:51:56 +00:00
gentbl_cc_library(
name = "MathPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Math",
],
"include/mlir/Dialect/Math/Transforms/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Math/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
cc_library(
name = "MathDialect",
srcs = glob(
[
"lib/Dialect/Math/IR/*.cpp",
"lib/Dialect/Math/IR/*.h",
],
),
hdrs = [
"include/mlir/Dialect/Math/IR/Math.h",
],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Update the Bazel build files for "[mlir][Math] Replace some constant ..."
2022-04-12 13:47:51 +02:00
":CommonFolders",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
[bazel] Port 92a836da0759
2022-04-28 22:51:27 +02:00
":InferTypeOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MathBaseIncGen",
":MathOpsIncGen",
":SideEffectInterfaces",
":VectorInterfaces",
"//llvm:Support",
],
)
cc_library(
name = "MathTransforms",
srcs = glob([
"lib/Dialect/Math/Transforms/*.cpp",
"lib/Dialect/Math/Transforms/*.h",
]),
hdrs = glob(["include/mlir/Dialect/Math/Transforms/*.h"]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":DialectUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":MathDialect",
[Bazel][mlir] Port ee8b8d6b58e1ff45d1bcbad38c6cf458f872b38d
2023-06-18 17:51:56 +00:00
":MathPassIncGen",
[Bazel][mlir] Fix layering check after 11db162db07d6083b79f4724e649a8c2c69913e1
2023-06-19 09:26:17 +02:00
":Pass",
[bazel] Add a missing dependency after f3bdb56d61e3
2022-06-01 22:49:34 +02:00
":SCFDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":VectorUtils",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":X86VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
cc_library(
name = "MathToLibm",
srcs = glob([
"lib/Conversion/MathToLibm/*.cpp",
"lib/Conversion/MathToLibm/*.h",
[bazel] Drop ConversionPassDetail, it shouldn't be needed after 67d0d7ac0acb0665d6a09f61278fbcf51f0114c2
2022-08-31 13:31:11 +02:00
]),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
hdrs = glob([
"include/mlir/Conversion/MathToLibm/*.h",
]),
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ConversionPassIncGen",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":LLVMDialect",
":MathDialect",
":Pass",
":Support",
":Transforms",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
[mlir][vector][NFC] Split into IR, Transforms and Utils This reduces the dependencies of the MLIRVector target and makes the dialect consistent with other dialects. Differential Revision: https://reviews.llvm.org/D118533
2022-01-31 19:10:51 +09:00
":VectorUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Core",
"//llvm:Support",
],
)
td_library(
name = "MemRefOpsTdFiles",
srcs = [
"include/mlir/Dialect/MemRef/IR/MemRefBase.td",
"include/mlir/Dialect/MemRef/IR/MemRefOps.td",
],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithOpsTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":CastInterfacesTdFiles",
":ControlFlowInterfacesTdFiles",
":CopyOpInterfaceTdFiles",
[Bazel] Add missing tablegen. (For MemorySlotTypeInterfaces) Summary: Reviewers: khasanovaa Subscribers:
2023-05-22 13:07:45 +02:00
":MemorySlotInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":OpBaseTdFiles",
[mlir][interfaces] Add ShapedDimOpInterface This interface is implemented by memref.dim and tensor.dim. This change makes it possible to remove a build dependency of the Affine dialect on the Tensor dialect (and maybe also the MemRef dialect in the future). Differential Revision: https://reviews.llvm.org/D133595
2022-10-03 13:53:16 +09:00
":ShapedOpInterfacesTdFiles",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":SideEffectInterfacesTdFiles",
":ViewLikeInterfaceTdFiles",
],
)
gentbl_cc_library(
name = "MemRefBaseIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=memref",
],
"include/mlir/Dialect/MemRef/IR/MemRefOpsDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=memref",
],
"include/mlir/Dialect/MemRef/IR/MemRefOpsDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/MemRef/IR/MemRefBase.td",
deps = [":MemRefOpsTdFiles"],
)
gentbl_cc_library(
name = "MemRefOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/MemRef/IR/MemRefOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/MemRef/IR/MemRefOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/MemRef/IR/MemRefOps.td",
[mlir][interfaces] Add ShapedDimOpInterface This interface is implemented by memref.dim and tensor.dim. This change makes it possible to remove a build dependency of the Affine dialect on the Tensor dialect (and maybe also the MemRef dialect in the future). Differential Revision: https://reviews.llvm.org/D133595
2022-10-03 13:53:16 +09:00
deps = [
":MemRefOpsTdFiles",
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
)
cc_library(
name = "MemRefDialect",
srcs = glob(
[
"lib/Dialect/MemRef/IR/*.cpp",
"lib/Dialect/MemRef/IR/*.h",
],
),
hdrs = [
"include/mlir/Dialect/MemRef/IR/MemRef.h",
[bazel] Fix build after 0bf120a82040f7ffaba0f0ab72a983f1cd9343ab
2023-05-24 11:14:18 +02:00
"include/mlir/Dialect/MemRef/IR/MemRefMemorySlot.h",
[mlir][memref] Add ValueBoundsOpInterface impls Differential Revision: https://reviews.llvm.org/D145695
2023-04-06 10:35:12 +09:00
"include/mlir/Dialect/MemRef/IR/ValueBoundsOpInterfaceImpl.h",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithUtils",
[mlir][IR][NFC] Move CastOpInterface helpers to mlir/Interfaces These helpers should not be part of the IR build unit. The interface is now implemented on `builtin.unrealized_conversion_cast` with an external model. Also rename the CastOpInterfaces Bazel target name to CastInterfaces to be consistent with the CMake target name. Differential Revision: https://reviews.llvm.org/D146972
2023-06-02 08:32:01 +02:00
":CastInterfaces",
Update BUILD.bazel for Mem2Reg deps. Reviewed By: csigg Differential Revision: https://reviews.llvm.org/D149930
2023-05-05 11:45:05 +02:00
":ComplexDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ControlFlowInterfaces",
":CopyOpInterface",
[Bazel] Update for 0813700de1 Updates Bazel BUILD files for changes in https://github.com/llvm/llvm-project/commit/0813700de1 Differential Revision: https://reviews.llvm.org/D104990
2021-06-27 08:30:54 -07:00
":DialectUtils",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":InferTypeOpInterface",
":MemRefBaseIncGen",
":MemRefOpsIncGen",
[Bazel] Add missing tablegen. (For MemorySlotTypeInterfaces) Summary: Reviewers: khasanovaa Subscribers:
2023-05-22 13:07:45 +02:00
":MemorySlotInterfaces",
[mlir][interfaces] Add ShapedDimOpInterface This interface is implemented by memref.dim and tensor.dim. This change makes it possible to remove a build dependency of the Affine dialect on the Tensor dialect (and maybe also the MemRef dialect in the future). Differential Revision: https://reviews.llvm.org/D133595
2022-10-03 13:53:16 +09:00
":ShapedOpInterfaces",
[bazel] Fix build after 0bf120a82040f7ffaba0f0ab72a983f1cd9343ab
2023-05-24 11:14:18 +02:00
":Support",
[mlir][memref] Add ValueBoundsOpInterface impls Differential Revision: https://reviews.llvm.org/D145695
2023-04-06 10:35:12 +09:00
":ValueBoundsOpInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":ViewLikeInterface",
"//llvm:Support",
[bazel] add TargetParser
2022-12-20 20:09:00 +01:00
"//llvm:TargetParser",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
)
[mlir][MemRef] Move narrow type emulation common methods to MemRefUtils. It also unifies the computation of StridedLayoutAttr. If the stride is static known value, we can just use it. Differential Revision: https://reviews.llvm.org/D155017
2023-07-11 14:13:07 -07:00
cc_library(
name = "MemRefUtils",
srcs = glob(
[
"lib/Dialect/MemRef/Utils/*.cpp",
],
),
hdrs = [
"include/mlir/Dialect/MemRef/Utils/MemRefUtils.h",
],
includes = ["include"],
deps = [
":AffineDialect",
":ArithUtils",
":MemRefDialect",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
gentbl_cc_library(
name = "MemRefPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=MemRef",
],
"include/mlir/Dialect/MemRef/Transforms/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/MemRef/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "MemRefTransforms",
srcs = glob(
[
"lib/Dialect/MemRef/Transforms/*.cpp",
"lib/Dialect/MemRef/Transforms/*.h",
],
),
hdrs = glob(["include/mlir/Dialect/MemRef/Transforms/*.h"]),
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir][memref] Add helper to make alloca ops independent Add a helper function that makes dynamic sizes of `memref.alloca` ops independent of a given set of values. This functionality can be used to make dynamic allocations hoistable from loops. Differential Revision: https://reviews.llvm.org/D149316
2023-05-04 14:07:19 +09:00
":AffineTransforms",
[bazel] Update MLIR deps I believe this is due to D117839, D117848, maybe others. Latest build failure: https://buildkite.com/llvm-project/upstream-bazel-rbe/builds/18053
2022-01-24 21:21:37 -08:00
":AffineUtils",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
":ArithTransforms",
":ArithUtils",
[mlir][memref] Bufferize memref.tensor_store op This change adds the BufferizableOpInterface implementation for memref.tensor_store. Differential Revision: https://reviews.llvm.org/D144080
2023-02-15 15:26:14 +01:00
":BufferizationDialect",
[mlir] Add RuntimeVerifiableOpInterface and transform Static op verification cannot detect cases where an op is valid at compile time but may be invalid at runtime. An example of such an op is `memref::ExpandShapeOp`. Invalid at compile time: `memref.expand_shape %m [[0, 1]] : memref<11xf32> into memref<2x5xf32>` Valid at compile time (because we do not know any better): `memref.expand_shape %m [[0, 1]] : memref<?xf32> into memref<?x5xf32>`. This op may or may not be valid at runtime depending on the runtime shape of `%m`. Invalid runtime ops such as the one above are hard to debug because they can crash the program execution at a seemingly unrelated position or (even worse) compute an invalid result without crashing. This revision adds a new op interface `RuntimeVerifiableOpInterface` that can be implemented by ops that provide additional runtime verification. Such runtime verification can be computationally expensive, so it is only generated on an opt-in basis by running `-generate-runtime-verification`. A simple runtime verifier for `memref::ExpandShapeOp` is provided as an example. Differential Revision: https://reviews.llvm.org/D138576
2022-12-21 10:51:10 +01:00
":ControlFlowDialect",
[bazel] Add missing dependency after 1b002d27683522e15d6ef3bccb74e751bbf56e74
2022-08-28 09:56:17 +02:00
":DialectUtils",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir][memref] Fold subview into GPU subgroup MMA load/store ops This commits adds support for folding subview into GPU subgroup MMA load/store ops. Reviewed By: ThomasRaoux Differential Revision: https://reviews.llvm.org/D146150
2023-03-15 17:49:27 +00:00
":GPUDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":IR",
":InferTypeOpInterface",
[mlir][memref] Add transformation to do loop multi-buffering This transformation is useful to break dependency between consecutive loop iterations by increasing the size of a temporary buffer. This is usually combined with heavy software pipelining. Differential Revision: https://reviews.llvm.org/D119406
2022-02-15 23:49:32 -08:00
":LoopLikeInterface",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":MemRefDialect",
":MemRefPassIncGen",
[mlir][MemRef] Move narrow type emulation common methods to MemRefUtils. It also unifies the computation of StridedLayoutAttr. If the stride is static known value, we can just use it. Differential Revision: https://reviews.llvm.org/D155017
2023-07-11 14:13:07 -07:00
":MemRefUtils",
[mlir][MemRef] Add patterns to extract address computations This patch adds patterns to rewrite memory accesses such that the resulting accesses are only using a base pointer. E.g., ```mlir memref.load %base[%off0, ...] ``` Will be rewritten in: ```mlir %new_base = memref.subview %base[%off0,...][1,...][1,...] memref.load %new_base[%c0,...] ``` The idea behind these patterns is to offer a way to more gradually lower address computations. These patterns are the exact opposite of FoldMemRefAliasOps. I've implemented the support of only five operations in this patch: - memref.load - memref.store - nvgpu.ldmatrix - vector.transfer_read - vector.transfer_write Going forward we may want to provide an interface for these rewritings (and the ones in FoldMemRefAliasOps.) One step at a time! Differential Revision: https://reviews.llvm.org/D146724
2023-03-23 15:41:14 +01:00
":NVGPUDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":Pass",
[mlir] Add RuntimeVerifiableOpInterface and transform Static op verification cannot detect cases where an op is valid at compile time but may be invalid at runtime. An example of such an op is `memref::ExpandShapeOp`. Invalid at compile time: `memref.expand_shape %m [[0, 1]] : memref<11xf32> into memref<2x5xf32>` Valid at compile time (because we do not know any better): `memref.expand_shape %m [[0, 1]] : memref<?xf32> into memref<?x5xf32>`. This op may or may not be valid at runtime depending on the runtime shape of `%m`. Invalid runtime ops such as the one above are hard to debug because they can crash the program execution at a seemingly unrelated position or (even worse) compute an invalid result without crashing. This revision adds a new op interface `RuntimeVerifiableOpInterface` that can be implemented by ops that provide additional runtime verification. Such runtime verification can be computationally expensive, so it is only generated on an opt-in basis by running `-generate-runtime-verification`. A simple runtime verifier for `memref::ExpandShapeOp` is provided as an example. Differential Revision: https://reviews.llvm.org/D138576
2022-12-21 10:51:10 +01:00
":RuntimeVerifiableOpInterface",
[MLIR][MemRef] Add missing #include for FailureOr FailureOr was used without including correct headers, so the code only works if the user of Transform.h includes the correct headers first. Reviewed By: jyknight Differential Revision: https://reviews.llvm.org/D147069
2023-03-28 11:06:54 -07:00
":Support",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
":TensorDialect",
":Transforms",
[mlir][memref] Add helper to make alloca ops independent Add a helper function that makes dynamic sizes of `memref.alloca` ops independent of a given set of values. This functionality can be used to make dynamic allocations hoistable from loops. Differential Revision: https://reviews.llvm.org/D149316
2023-05-04 14:07:19 +09:00
":ValueBoundsOpInterface",
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":VectorDialect",
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
"//llvm:Support",
],
)
[bazel] Port 3f050f6ac4626c9bf99bfdf5e4d7162ba7ad5cdc
2022-09-29 00:12:46 +02:00
td_library(
name = "MemRefTransformOpsTdFiles",
srcs = [
"include/mlir/Dialect/MemRef/TransformOps/MemRefTransformOps.td",
],
includes = ["include"],
deps = [
":TransformDialectTdFiles",
],
)
gentbl_cc_library(
name = "MemRefTransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/MemRef/TransformOps/MemRefTransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/MemRef/TransformOps/MemRefTransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/MemRef/TransformOps/MemRefTransformOps.td",
deps = [
":MemRefTransformOpsTdFiles",
],
)
cc_library(
name = "MemRefTransformOps",
srcs = glob(["lib/Dialect/MemRef/TransformOps/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/MemRef/TransformOps/*.h"]),
includes = ["include"],
deps = [
":AffineDialect",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[bazel] Port 3f050f6ac4626c9bf99bfdf5e4d7162ba7ad5cdc
2022-09-29 00:12:46 +02:00
":IR",
[mlir][MemRef][Transform] Don't apply multibuffer on "useless" allocs `alloc`s that have users outside of loops are guaranteed to fail in `multibuffer`. Instead of exposing ourselves to that failure in the transform dialect, filter out the `alloc`s that fall in this category. To be able to do this filtering we have to change the `multibuffer` transform op from `TransformEachOpTrait` to a plain `TransformOp`. This is because `TransformEachOpTrait` expects that every successful `applyToOne` returns a non-empty result. Couple of notes: - I changed the assembly syntax to make sure we only get `alloc` ops as input. (And added a test case to make sure we reject invalid inputs.) - `multibuffer` can still fail pretty easily when you know its limitations. See the updated `op failed to multibuffer` test case for instance. Longer term, instead of leaking/coupling the actual implementation (in this case the checks normally done in `memref::multiBuffer`) with the transform dialect (the added check in `::apply`), we may want to refactor how we structure the underlying implementation. E.g., we could imagine a `canApply` method for all the implementations that we want to hook up in the transform dialect. This has some implications on how not to duplicate work between `canApply` and the actual implementation but I thought I throw that here to have us think about it :). Differential Revision: https://reviews.llvm.org/D143747
2023-02-10 17:21:38 +01:00
":LoopLikeInterface",
[bazel] Port 3f050f6ac4626c9bf99bfdf5e4d7162ba7ad5cdc
2022-09-29 00:12:46 +02:00
":MemRefDialect",
":MemRefTransformOpsIncGen",
":MemRefTransforms",
[mlir][MemRef] Add patterns to extract address computations This patch adds patterns to rewrite memory accesses such that the resulting accesses are only using a base pointer. E.g., ```mlir memref.load %base[%off0, ...] ``` Will be rewritten in: ```mlir %new_base = memref.subview %base[%off0,...][1,...][1,...] memref.load %new_base[%c0,...] ``` The idea behind these patterns is to offer a way to more gradually lower address computations. These patterns are the exact opposite of FoldMemRefAliasOps. I've implemented the support of only five operations in this patch: - memref.load - memref.store - nvgpu.ldmatrix - vector.transfer_read - vector.transfer_write Going forward we may want to provide an interface for these rewritings (and the ones in FoldMemRefAliasOps.) One step at a time! Differential Revision: https://reviews.llvm.org/D146724
2023-03-23 15:41:14 +01:00
":NVGPUDialect",
[mlir][memref] Add helper to make alloca ops independent Add a helper function that makes dynamic sizes of `memref.alloca` ops independent of a given set of values. This functionality can be used to make dynamic allocations hoistable from loops. Differential Revision: https://reviews.llvm.org/D149316
2023-05-04 14:07:19 +09:00
":SCFDialect",
[bazel] Port 3f050f6ac4626c9bf99bfdf5e4d7162ba7ad5cdc
2022-09-29 00:12:46 +02:00
":TransformDialect",
[mlir][MemRef] Add patterns to extract address computations This patch adds patterns to rewrite memory accesses such that the resulting accesses are only using a base pointer. E.g., ```mlir memref.load %base[%off0, ...] ``` Will be rewritten in: ```mlir %new_base = memref.subview %base[%off0,...][1,...][1,...] memref.load %new_base[%c0,...] ``` The idea behind these patterns is to offer a way to more gradually lower address computations. These patterns are the exact opposite of FoldMemRefAliasOps. I've implemented the support of only five operations in this patch: - memref.load - memref.store - nvgpu.ldmatrix - vector.transfer_read - vector.transfer_write Going forward we may want to provide an interface for these rewritings (and the ones in FoldMemRefAliasOps.) One step at a time! Differential Revision: https://reviews.llvm.org/D146724
2023-03-23 15:41:14 +01:00
":TransformUtils",
":VectorDialect",
[bazel] Port 3f050f6ac4626c9bf99bfdf5e4d7162ba7ad5cdc
2022-09-29 00:12:46 +02:00
"//llvm:Support",
],
)
[mlir] Introduce ml_program dialect. Differential Revision: https://reviews.llvm.org/D120203
2022-04-13 20:16:04 -07:00
##---------------------------------------------------------------------------##
# MLProgram dialect
##---------------------------------------------------------------------------##
td_library(
name = "MLProgramOpsTdFiles",
srcs = [
[mlir] Add GlobalOp, GlobalLoadConstOp to ml_program. The approach I took was to define a dialect 'extern' attribute that a GlobalOp can take as a value to signify external linkage. I think this approach should compose well and should also work with wherever the OpaqueElements work goes in the future (since that is just another kind of attribute). I special cased the GlobalOp parser/printer for this case because it is significantly easier on the eyes. In the discussion, Jeff Niu had proposed an alternative syntax for GlobalOp that I ended up not taking. I did try to implement it but a) I don't think it made anything easier to read in the common case, and b) it made the parsing/printing logic a lot more complicated (I think I would need a completely custom parser/printer to do it well). Please have a look at the common cases where the global type and initial value type match: I don't think how I have it is too bad. The less common cases seem ok to me. I chose to only implement the direct, constant load op since that is non side effecting and there was still discussion pending on that. Differential Revision: https://reviews.llvm.org/D124318
2022-04-22 19:59:34 -07:00
"include/mlir/Dialect/MLProgram/IR/MLProgramAttributes.td",
Adjust BUILD files for [MLIR][GPU] Add NvGpu mma.sync path to the VectorToGPU pass
2022-05-20 18:03:20 +02:00
"include/mlir/Dialect/MLProgram/IR/MLProgramBase.td",
[mlir] Introduce ml_program dialect. Differential Revision: https://reviews.llvm.org/D120203
2022-04-13 20:16:04 -07:00
"include/mlir/Dialect/MLProgram/IR/MLProgramOps.td",
[mlir] Add global_load and global_store ops to ml_program. * Adds simple, non-atomic, non-volatile, non-synchronized direct load/store ops. Differential Revision: https://reviews.llvm.org/D126230
2022-05-31 13:46:06 -07:00
"include/mlir/Dialect/MLProgram/IR/MLProgramTypes.td",
[mlir] Introduce ml_program dialect. Differential Revision: https://reviews.llvm.org/D120203
2022-04-13 20:16:04 -07:00
],
includes = ["include"],
deps = [
[mlir][bazel] Fix bazel build files This adds a missing dependency on builtin dialect td files. Differential Revision: https://reviews.llvm.org/D130890
2022-08-01 11:40:39 +02:00
":BuiltinDialectTdFiles",
[mlir] Introduce ml_program dialect. Differential Revision: https://reviews.llvm.org/D120203
2022-04-13 20:16:04 -07:00
":CallInterfacesTdFiles",
":ControlFlowInterfacesTdFiles",
":FunctionInterfacesTdFiles",
":OpBaseTdFiles",
":RegionKindInterfaceIncGen",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "MLProgramOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/MLProgram/IR/MLProgramOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/MLProgram/IR/MLProgramOps.cpp.inc",
),
(
["-gen-dialect-decls"],
"include/mlir/Dialect/MLProgram/IR/MLProgramOpsDialect.h.inc",
),
(
["-gen-dialect-defs"],
"include/mlir/Dialect/MLProgram/IR/MLProgramOpsDialect.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/MLProgram/IR/MLProgramOps.td",
deps = [":MLProgramOpsTdFiles"],
)
[mlir] Add GlobalOp, GlobalLoadConstOp to ml_program. The approach I took was to define a dialect 'extern' attribute that a GlobalOp can take as a value to signify external linkage. I think this approach should compose well and should also work with wherever the OpaqueElements work goes in the future (since that is just another kind of attribute). I special cased the GlobalOp parser/printer for this case because it is significantly easier on the eyes. In the discussion, Jeff Niu had proposed an alternative syntax for GlobalOp that I ended up not taking. I did try to implement it but a) I don't think it made anything easier to read in the common case, and b) it made the parsing/printing logic a lot more complicated (I think I would need a completely custom parser/printer to do it well). Please have a look at the common cases where the global type and initial value type match: I don't think how I have it is too bad. The less common cases seem ok to me. I chose to only implement the direct, constant load op since that is non side effecting and there was still discussion pending on that. Differential Revision: https://reviews.llvm.org/D124318
2022-04-22 19:59:34 -07:00
gentbl_cc_library(
name = "MLProgramAttributesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-attrdef-decls"],
"include/mlir/Dialect/MLProgram/IR/MLProgramAttributes.h.inc",
),
(
["-gen-attrdef-defs"],
"include/mlir/Dialect/MLProgram/IR/MLProgramAttributes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/MLProgram/IR/MLProgramAttributes.td",
deps = [":MLProgramOpsTdFiles"],
)
[mlir] Add global_load and global_store ops to ml_program. * Adds simple, non-atomic, non-volatile, non-synchronized direct load/store ops. Differential Revision: https://reviews.llvm.org/D126230
2022-05-31 13:46:06 -07:00
gentbl_cc_library(
name = "MLProgramTypesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-typedef-decls"],
"include/mlir/Dialect/MLProgram/IR/MLProgramTypes.h.inc",
),
(
["-gen-typedef-defs"],
"include/mlir/Dialect/MLProgram/IR/MLProgramTypes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/MLProgram/IR/MLProgramTypes.td",
deps = [":MLProgramOpsTdFiles"],
)
[mlir] Introduce ml_program dialect. Differential Revision: https://reviews.llvm.org/D120203
2022-04-13 20:16:04 -07:00
cc_library(
name = "MLProgramDialect",
srcs = glob([
"lib/Dialect/MLProgram/IR/*.cpp",
"lib/Dialect/MLProgram/IR/*.h",
]),
hdrs = glob([
"include/mlir/Dialect/MLProgram/IR/*.h",
]),
includes = ["include"],
deps = [
":ControlFlowInterfaces",
":IR",
[mlir] Add GlobalOp, GlobalLoadConstOp to ml_program. The approach I took was to define a dialect 'extern' attribute that a GlobalOp can take as a value to signify external linkage. I think this approach should compose well and should also work with wherever the OpaqueElements work goes in the future (since that is just another kind of attribute). I special cased the GlobalOp parser/printer for this case because it is significantly easier on the eyes. In the discussion, Jeff Niu had proposed an alternative syntax for GlobalOp that I ended up not taking. I did try to implement it but a) I don't think it made anything easier to read in the common case, and b) it made the parsing/printing logic a lot more complicated (I think I would need a completely custom parser/printer to do it well). Please have a look at the common cases where the global type and initial value type match: I don't think how I have it is too bad. The less common cases seem ok to me. I chose to only implement the direct, constant load op since that is non side effecting and there was still discussion pending on that. Differential Revision: https://reviews.llvm.org/D124318
2022-04-22 19:59:34 -07:00
":MLProgramAttributesIncGen",
[mlir] Introduce ml_program dialect. Differential Revision: https://reviews.llvm.org/D120203
2022-04-13 20:16:04 -07:00
":MLProgramOpsIncGen",
[mlir] Add global_load and global_store ops to ml_program. * Adds simple, non-atomic, non-volatile, non-synchronized direct load/store ops. Differential Revision: https://reviews.llvm.org/D126230
2022-05-31 13:46:06 -07:00
":MLProgramTypesIncGen",
[mlir] Introduce ml_program dialect. Differential Revision: https://reviews.llvm.org/D120203
2022-04-13 20:16:04 -07:00
":Pass",
":Support",
"//llvm:Support",
],
)
##---------------------------------------------------------------------------##
# Allocation interfaces
##---------------------------------------------------------------------------##
Revert "Revert "[mlir] Move AllocationOpInterface to Bufferize/IR/AllocationOpInterface.td."" This reverts and fixes commit de18b7dee6a81e5e790c8e8060065b1ef72d13ed.
2021-11-23 10:04:47 +01:00
td_library(
name = "AllocationOpInterfaceTdFiles",
srcs = ["include/mlir/Dialect/Bufferization/IR/AllocationOpInterface.td"],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
gentbl_cc_library(
name = "AllocationOpInterfaceIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-interface-decls"],
"include/mlir/Dialect/Bufferization/IR/AllocationOpInterface.h.inc",
),
(
["-gen-op-interface-defs"],
"include/mlir/Dialect/Bufferization/IR/AllocationOpInterface.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Bufferization/IR/AllocationOpInterface.td",
deps = [":AllocationOpInterfaceTdFiles"],
)
cc_library(
name = "AllocationOpInterface",
srcs = ["lib/Dialect/Bufferization/IR/AllocationOpInterface.cpp"],
hdrs = ["include/mlir/Dialect/Bufferization/IR/AllocationOpInterface.h"],
includes = ["include"],
deps = [
":AllocationOpInterfaceIncGen",
":IR",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
],
)
td_library(
name = "BufferizationOpsTdFiles",
srcs = [
"include/mlir/Dialect/Bufferization/IR/BufferizationBase.td",
[bazel] Fix Bufferization dialect build D137833 added a new .td file and updated existing files to use it. It broke bazel build. Differential Revision: https://reviews.llvm.org/D137961
2022-11-14 18:57:52 +01:00
"include/mlir/Dialect/Bufferization/IR/BufferizationEnums.td",
[bazel] Add another missing dependency after D137833 While there run buildifier.
2022-11-14 19:02:56 +01:00
"include/mlir/Dialect/Bufferization/IR/BufferizationOps.td",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
],
includes = ["include"],
deps = [
":AllocationOpInterfaceTdFiles",
":CopyOpInterfaceTdFiles",
[mlir][bufferization] Add bufferization.alloc_tensor op This change adds a new op `alloc_tensor` to the bufferization dialect. During bufferization, this op is always lowered to a buffer allocation (unless it is "eliminated" by a pre-processing pass). It is useful to have such an op in tensor land, because it allows users to model tensor SSA use-def chains (which drive bufferization decisions) and because tensor SSA use-def chains can be analyzed by One-Shot Bufferize, while memref values cannot. This change also replaces all uses of linalg.init_tensor in bufferization-related code with bufferization.alloc_tensor. linalg.init_tensor and bufferization.alloc_tensor are similar, but the purpose of the former one is just to carry a shape. It does not indicate a memory allocation. linalg.init_tensor is not suitable for modelling SSA use-def chains for bufferization purposes, because linalg.init_tensor is marked as not having side effects (in contrast to alloc_tensor). As such, it is legal to move linalg.init_tensor ops around/CSE them/etc. This is not desirable for alloc_tensor; it represents an explicit buffer allocation while still in tensor land and such allocations should not suddenly disappear or get moved around when running the canonicalizer/CSE/etc. BEGIN_PUBLIC No public commit message needed for presubmit. END_PUBLIC Differential Revision: https://reviews.llvm.org/D126003
2022-05-21 02:31:07 +02:00
":InferTypeOpInterfaceTdFiles",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "BufferizationBaseIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=bufferization",
],
"include/mlir/Dialect/Bufferization/IR/BufferizationOpsDialect.h.inc",
),
(
[
"-gen-dialect-defs",
"-dialect=bufferization",
],
"include/mlir/Dialect/Bufferization/IR/BufferizationOpsDialect.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Bufferization/IR/BufferizationBase.td",
deps = [":BufferizationOpsTdFiles"],
)
Update bazel files.
2022-11-15 07:48:51 +01:00
td_library(
name = "BufferizationEnumsTdFiles",
srcs = [
"include/mlir/Dialect/Bufferization/IR/BufferizationEnums.td",
],
includes = ["include"],
deps = [
":OpBaseTdFiles",
],
)
[bazel] Fix Bufferization dialect build D137833 added a new .td file and updated existing files to use it. It broke bazel build. Differential Revision: https://reviews.llvm.org/D137961
2022-11-14 18:57:52 +01:00
gentbl_cc_library(
name = "BufferizationEnumsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-enum-decls"],
"include/mlir/Dialect/Bufferization/IR/BufferizationEnums.h.inc",
),
(
["-gen-enum-defs"],
"include/mlir/Dialect/Bufferization/IR/BufferizationEnums.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Bufferization/IR/BufferizationEnums.td",
Update bazel files.
2022-11-15 07:48:51 +01:00
deps = [":BufferizationEnumsTdFiles"],
[bazel] Fix Bufferization dialect build D137833 added a new .td file and updated existing files to use it. It broke bazel build. Differential Revision: https://reviews.llvm.org/D137961
2022-11-14 18:57:52 +01:00
)
[mlir][bufferization] Add OneShotBufferize transform op This commit allows for One-Shot Bufferize to be used through the transform dialect. No op handle is currently returned for the bufferized IR. Differential Revision: https://reviews.llvm.org/D125098
2022-06-09 13:00:08 +02:00
td_library(
name = "BufferizationTransformOpsTdFiles",
srcs = [
"include/mlir/Dialect/Bufferization/TransformOps/BufferizationTransformOps.td",
],
includes = ["include"],
deps = [
[bazel] Add another missing dependency after D137833 While there run buildifier.
2022-11-14 19:02:56 +01:00
":BufferizationOpsTdFiles",
[mlir][bufferization] Add OneShotBufferize transform op This commit allows for One-Shot Bufferize to be used through the transform dialect. No op handle is currently returned for the bufferized IR. Differential Revision: https://reviews.llvm.org/D125098
2022-06-09 13:00:08 +02:00
":TransformDialectTdFiles",
],
)
gentbl_cc_library(
name = "BufferizationTransformOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Bufferization/TransformOps/BufferizationTransformOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Bufferization/TransformOps/BufferizationTransformOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Bufferization/TransformOps/BufferizationTransformOps.td",
deps = [
":BufferizationTransformOpsTdFiles",
],
)
cc_library(
name = "BufferizationTransformOps",
srcs = [
"lib/Dialect/Bufferization/TransformOps/BufferizationTransformOps.cpp",
],
hdrs = [
"include/mlir/Dialect/Bufferization/TransformOps/BufferizationTransformOps.h",
],
includes = ["include"],
deps = [
":BufferizationDialect",
[bazel] Add another missing dependency after D137833 While there run buildifier.
2022-11-14 19:02:56 +01:00
":BufferizationEnumsIncGen",
[mlir][bufferization] Add OneShotBufferize transform op This commit allows for One-Shot Bufferize to be used through the transform dialect. No op handle is currently returned for the bufferized IR. Differential Revision: https://reviews.llvm.org/D125098
2022-06-09 13:00:08 +02:00
":BufferizationTransformOpsIncGen",
":BufferizationTransforms",
":IR",
[mlir][bufferization] OneShotBufferizeOp: Add options to use linalg.copy This new option allows users to specify a custom memcpy op. Differential Revision: https://reviews.llvm.org/D155280
2023-07-14 11:57:54 +02:00
":LinalgDialect",
[mlir][bufferization] Add OneShotBufferize transform op This commit allows for One-Shot Bufferize to be used through the transform dialect. No op handle is currently returned for the bufferized IR. Differential Revision: https://reviews.llvm.org/D125098
2022-06-09 13:00:08 +02:00
":MemRefDialect",
":Parser",
":SideEffectInterfaces",
[mlir][Bazel] Add missing dependency to TensorDialect.
2022-12-19 10:03:44 +01:00
":TensorDialect",
[mlir][bufferization] Add OneShotBufferize transform op This commit allows for One-Shot Bufferize to be used through the transform dialect. No op handle is currently returned for the bufferized IR. Differential Revision: https://reviews.llvm.org/D125098
2022-06-09 13:00:08 +02:00
":TransformDialect",
"//llvm:Support",
],
)
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
gentbl_cc_library(
name = "BufferizationOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["-gen-op-decls"],
"include/mlir/Dialect/Bufferization/IR/BufferizationOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/Bufferization/IR/BufferizationOps.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Bufferization/IR/BufferizationOps.td",
[mlir][bufferize][NFC] Implement BufferizableOpInterface on bufferization ops directly No longer go through an external model. Also put BufferizableOpInterface into the same build target as the BufferizationDialect. This allows for some code reuse between BufferizationOps canonicalizers and BufferizableOpInterface implementations. Differential Revision: https://reviews.llvm.org/D117987
2022-01-25 00:51:41 +09:00
deps = [
":BufferizableOpInterfaceTdFiles",
":BufferizationOpsTdFiles",
],
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
)
cc_library(
name = "BufferizationDialect",
[mlir][bufferization] Move one-shot bufferization to Bufferization dialect This commit is the first step towards unifying core bufferization and One-Shot Bufferize. This commit does not move over the implementations of BufferizableOpInterface yet. This will be done in separate commits. This change does also not move the unit tests yet. The tests will be moved together with op interface implementations and split into separate files. Differential Revision: https://reviews.llvm.org/D117641
2022-01-20 18:14:59 +09:00
srcs = [
[mlir][bufferize][NFC] Implement BufferizableOpInterface on bufferization ops directly No longer go through an external model. Also put BufferizableOpInterface into the same build target as the BufferizationDialect. This allows for some code reuse between BufferizationOps canonicalizers and BufferizableOpInterface implementations. Differential Revision: https://reviews.llvm.org/D117987
2022-01-25 00:51:41 +09:00
"lib/Dialect/Bufferization/IR/BufferizableOpInterface.cpp",
[mlir][bufferization] Move one-shot bufferization to Bufferization dialect This commit is the first step towards unifying core bufferization and One-Shot Bufferize. This commit does not move over the implementations of BufferizableOpInterface yet. This will be done in separate commits. This change does also not move the unit tests yet. The tests will be moved together with op interface implementations and split into separate files. Differential Revision: https://reviews.llvm.org/D117641
2022-01-20 18:14:59 +09:00
"lib/Dialect/Bufferization/IR/BufferizationDialect.cpp",
"lib/Dialect/Bufferization/IR/BufferizationOps.cpp",
],
[mlir][bufferize][NFC] Implement BufferizableOpInterface on bufferization ops directly No longer go through an external model. Also put BufferizableOpInterface into the same build target as the BufferizationDialect. This allows for some code reuse between BufferizationOps canonicalizers and BufferizableOpInterface implementations. Differential Revision: https://reviews.llvm.org/D117987
2022-01-25 00:51:41 +09:00
hdrs = [
"include/mlir/Dialect/Bufferization/IR/BufferizableOpInterface.h",
"include/mlir/Dialect/Bufferization/IR/Bufferization.h",
[mlir][bufferize] Provide default BufferizableOpInterface impl for destination style ops tensor.insert and tensor.insert_slice (as destination style ops) do no longer need to implement the entire BufferizableOpInterface. Differential Revision: https://reviews.llvm.org/D136347
2022-10-27 10:40:24 +02:00
"include/mlir/Dialect/Bufferization/IR/DstBufferizableOpInterfaceImpl.h",
[mlir][bufferize][NFC] Implement BufferizableOpInterface on bufferization ops directly No longer go through an external model. Also put BufferizableOpInterface into the same build target as the BufferizationDialect. This allows for some code reuse between BufferizationOps canonicalizers and BufferizableOpInterface implementations. Differential Revision: https://reviews.llvm.org/D117987
2022-01-25 00:51:41 +09:00
],
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
includes = ["include"],
deps = [
[mlir] (NFC) Clean up bazel and CMake target names All dialect targets in bazel have been named *Dialect and all dialect targets in CMake have been named MLIR*Dialect.
2022-06-13 06:50:55 +00:00
":AffineDialect",
[mlir] Fix BufferizationToMemRef build.
2021-11-30 13:10:37 +01:00
":AllocationOpInterface",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir][bufferize][NFC] Implement BufferizableOpInterface on bufferization ops directly No longer go through an external model. Also put BufferizableOpInterface into the same build target as the BufferizationDialect. This allows for some code reuse between BufferizationOps canonicalizers and BufferizableOpInterface implementations. Differential Revision: https://reviews.llvm.org/D117987
2022-01-25 00:51:41 +09:00
":BufferizableOpInterfaceIncGen",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":BufferizationBaseIncGen",
[bazel] Fix Bufferization dialect build D137833 added a new .td file and updated existing files to use it. It broke bazel build. Differential Revision: https://reviews.llvm.org/D137961
2022-11-14 18:57:52 +01:00
":BufferizationEnumsIncGen",
[bazel] Add another missing dependency after D137833 While there run buildifier.
2022-11-14 19:02:56 +01:00
":BufferizationOpsIncGen",
[mlir][bufferize] Add isRepetitiveRegion to BufferizableOpInterface This method allows to declare regions as "repetitive" even if the parent op does not implement the RegionBranchOpInterface. This is needed to support loop-like ops that have parallel semantics but do not branch between regions. Differential Revision: https://reviews.llvm.org/D133113
2022-09-02 14:32:04 +02:00
":ControlFlowInterfaces",
[mlir][Bazel] Add missing dependencies. When building with layering_check enabled, there needs to be a dependency for each header include.
2022-04-25 09:16:16 +02:00
":CopyOpInterface",
[mlir][bufferize] Provide default BufferizableOpInterface impl for destination style ops tensor.insert and tensor.insert_slice (as destination style ops) do no longer need to implement the entire BufferizableOpInterface. Differential Revision: https://reviews.llvm.org/D136347
2022-10-27 10:40:24 +02:00
":DestinationStyleOpInterface",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":IR",
[mlir][bufferization] Add bufferization.alloc_tensor op This change adds a new op `alloc_tensor` to the bufferization dialect. During bufferization, this op is always lowered to a buffer allocation (unless it is "eliminated" by a pre-processing pass). It is useful to have such an op in tensor land, because it allows users to model tensor SSA use-def chains (which drive bufferization decisions) and because tensor SSA use-def chains can be analyzed by One-Shot Bufferize, while memref values cannot. This change also replaces all uses of linalg.init_tensor in bufferization-related code with bufferization.alloc_tensor. linalg.init_tensor and bufferization.alloc_tensor are similar, but the purpose of the former one is just to carry a shape. It does not indicate a memory allocation. linalg.init_tensor is not suitable for modelling SSA use-def chains for bufferization purposes, because linalg.init_tensor is marked as not having side effects (in contrast to alloc_tensor). As such, it is legal to move linalg.init_tensor ops around/CSE them/etc. This is not desirable for alloc_tensor; it represents an explicit buffer allocation while still in tensor land and such allocations should not suddenly disappear or get moved around when running the canonicalizer/CSE/etc. BEGIN_PUBLIC No public commit message needed for presubmit. END_PUBLIC Differential Revision: https://reviews.llvm.org/D126003
2022-05-21 02:31:07 +02:00
":InferTypeOpInterface",
[mlir] Fix BufferizationToMemRef build.
2021-11-30 13:10:37 +01:00
":MemRefDialect",
[mlir][bufferization][sparse] put restriction on sparse tensor allocation Putting some direct use restrictions on tensor allocations in the sparse case enables the use of simplifying assumptions in the bufferization analysis. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D128463
2022-06-23 12:17:01 -07:00
":SparseTensorDialect",
[mlir][bufferize][NFC] Implement BufferizableOpInterface on bufferization ops directly No longer go through an external model. Also put BufferizableOpInterface into the same build target as the BufferizationDialect. This allows for some code reuse between BufferizationOps canonicalizers and BufferizableOpInterface implementations. Differential Revision: https://reviews.llvm.org/D117987
2022-01-25 00:51:41 +09:00
":Support",
[mlir] Move memref.[tensor_load|buffer_cast|clone] to "bufferization" dialect. https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712 Differential Revision: https://reviews.llvm.org/D114552
2021-11-25 11:42:16 +01:00
":TensorDialect",
"//llvm:Support",
Revert "Revert "[mlir] Move AllocationOpInterface to Bufferize/IR/AllocationOpInterface.td."" This reverts and fixes commit de18b7dee6a81e5e790c8e8060065b1ef72d13ed.
2021-11-23 10:04:47 +01:00
],
)
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
gentbl_cc_library(
name = "BufferizationPassIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Bufferization",
],
"include/mlir/Dialect/Bufferization/Transforms/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/Bufferization/Transforms/Passes.td",
deps = [":PassBaseTdFiles"],
)
cc_library(
name = "BufferizationTransforms",
srcs = glob(
[
"lib/Dialect/Bufferization/Transforms/*.cpp",
"lib/Dialect/Bufferization/Transforms/*.h",
],
),
hdrs = glob(["include/mlir/Dialect/Bufferization/Transforms/*.h"]),
includes = ["include"],
deps = [
":AllocationOpInterface",
":Analysis",
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":BufferizationDialect",
Update bazel files.
2022-11-15 07:48:51 +01:00
":BufferizationEnumsIncGen",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":BufferizationPassIncGen",
[mlir][bufferization] Move one-shot bufferization to Bufferization dialect This commit is the first step towards unifying core bufferization and One-Shot Bufferize. This commit does not move over the implementations of BufferizableOpInterface yet. This will be done in separate commits. This change does also not move the unit tests yet. The tests will be moved together with op interface implementations and split into separate files. Differential Revision: https://reviews.llvm.org/D117641
2022-01-20 18:14:59 +09:00
":ControlFlowInterfaces",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":IR",
[bazel] Update MLIR deps I believe this is due to D117839, D117848, maybe others. Latest build failure: https://buildkite.com/llvm-project/upstream-bazel-rbe/builds/18053
2022-01-24 21:21:37 -08:00
":LoopLikeInterface",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":MemRefDialect",
[mlir][MemRef] Move narrow type emulation common methods to MemRefUtils. It also unifies the computation of StridedLayoutAttr. If the stride is static known value, we can just use it. Differential Revision: https://reviews.llvm.org/D155017
2023-07-11 14:13:07 -07:00
":MemRefUtils",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":Pass",
[mlir] Remove `Transforms/SideEffectUtils.h` and move the methods into `Interface/SideEffectInterfaces.h`. The methods in `SideEffectUtils.h` (and their implementations in `SideEffectUtils.cpp`) seem to have similar intent to methods already existing in `SideEffectInterfaces.h`. Move the decleration (and implementation) from `SideEffectUtils.h` (and `SideEffectUtils.cpp`) into `SideEffectInterfaces.h` (and `SideEffectInterface.cpp`). Also drop the `SideEffectInterface::hasNoEffect` method in favor of `mlir::isMemoryEffectFree` which actually recurses into the operation instead of just relying on the `hasRecursiveMemoryEffectTrait` exclusively. Differential Revision: https://reviews.llvm.org/D137857
2022-11-11 17:10:38 +00:00
":SideEffectInterfaces",
[mlir][bufferization] Add bufferization.alloc_tensor op This change adds a new op `alloc_tensor` to the bufferization dialect. During bufferization, this op is always lowered to a buffer allocation (unless it is "eliminated" by a pre-processing pass). It is useful to have such an op in tensor land, because it allows users to model tensor SSA use-def chains (which drive bufferization decisions) and because tensor SSA use-def chains can be analyzed by One-Shot Bufferize, while memref values cannot. This change also replaces all uses of linalg.init_tensor in bufferization-related code with bufferization.alloc_tensor. linalg.init_tensor and bufferization.alloc_tensor are similar, but the purpose of the former one is just to carry a shape. It does not indicate a memory allocation. linalg.init_tensor is not suitable for modelling SSA use-def chains for bufferization purposes, because linalg.init_tensor is marked as not having side effects (in contrast to alloc_tensor). As such, it is legal to move linalg.init_tensor ops around/CSE them/etc. This is not desirable for alloc_tensor; it represents an explicit buffer allocation while still in tensor land and such allocations should not suddenly disappear or get moved around when running the canonicalizer/CSE/etc. BEGIN_PUBLIC No public commit message needed for presubmit. END_PUBLIC Differential Revision: https://reviews.llvm.org/D126003
2022-05-21 02:31:07 +02:00
":TensorDialect",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
":Transforms",
Move BufferViewFlowAnalysis to the Bufferization dialect. It's only used from there, and this lets us remove the dependency from Analysis to the Arith dialect. Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D132928
2022-08-30 13:15:27 +02:00
":ViewLikeInterface",
[mlir] Move bufferization-related passes to `bufferization` dialect. [RFC](https://llvm.discourse.group/t/rfc-dialect-for-bufferization-related-ops/4712) Differential Revision: https://reviews.llvm.org/D114698
2021-11-29 13:45:01 +01:00
"//llvm:Support",
],
)
[mlir] Add bazel build for BufferizationToMemRef.
2021-11-30 13:27:24 +01:00
cc_library(
name = "BufferizationToMemRef",
srcs = [
"lib/Conversion/BufferizationToMemRef/BufferizationToMemRef.cpp",
],
hdrs = [
"include/mlir/Conversion/BufferizationToMemRef/BufferizationToMemRef.h",
],
includes = ["include"],
deps = [
[mlir][arith] Change dialect name from Arithmetic to Arith Suggested by @lattner in https://discourse.llvm.org/t/rfc-define-precise-arith-semantics/65507/22. Tested with: `ninja check-mlir check-mlir-integration check-mlir-mlir-spirv-cpu-runner check-mlir-mlir-vulkan-runner check-mlir-examples` and `bazel build --config=generic_clang @llvm-project//mlir:all`. Reviewed By: lattner, Mogball, rriddle, jpienaar, mehdi_amini Differential Revision: https://reviews.llvm.org/D134762
2022-09-29 11:14:47 -04:00
":ArithDialect",
[mlir] Add bazel build for BufferizationToMemRef.
2021-11-30 13:27:24 +01:00
":BufferizationDialect",
":ConversionPassIncGen",
[mlir] more Bazel changes for 23aa5a744666
2022-03-02 16:16:14 +01:00
":FuncDialect",
[mlir] Add bazel build for BufferizationToMemRef.
2021-11-30 13:27:24 +01:00
":IR",
":MemRefDialect",
":Pass",
[mlir][bufferization] Add lowering of bufferization.dealloc to memref.dealloc Adds a generic lowering that suppors all cases of bufferization.dealloc and one specialized, more efficient lowering for the simple case. Using a helper function with for loops in the general case enables O(|num_dealloc_memrefs|+|num_retain_memrefs|) size of the lowered code. Depends on D155467 Reviewed By: springerm Differential Revision: https://reviews.llvm.org/D155468
2023-07-19 14:13:27 +00:00
":SCFDialect",
[mlir] Add bazel build for BufferizationToMemRef.
2021-11-30 13:27:24 +01:00
":Support",
":Transforms",
],
)
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
td_library(
name = "DLTIDialectTdFiles",
srcs = [
"include/mlir/Dialect/DLTI/DLTI.td",
"include/mlir/Dialect/DLTI/DLTIBase.td",
],
includes = ["include"],
deps = [":OpBaseTdFiles"],
)
gentbl_cc_library(
name = "DLTIBaseIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=dlti",
],
"include/mlir/Dialect/DLTI/DLTIDialect.h.inc",
),
[mlir] Generare .cpp.inc files for dialects. * Previously, we were only generating .h.inc files. We foresee the need to also generate implementations and this is a step towards that. * Discussed in https://llvm.discourse.group/t/generating-cpp-inc-files-for-dialects/3732/2 * Deviates from the discussion above by generating a default constructor in the .cpp.inc file (and adding a tablegen bit that disables this in case if this is user provided). * Generating the destructor started as a way to flush out the missing includes (produces a link error), but it is a strict improvement on its own that is worth doing (i.e. by emitting key methods in the .cpp file, we root vtables in one translation unit, which is a non-controversial improvement). Differential Revision: https://reviews.llvm.org/D105070
2021-06-28 22:54:11 +00:00
(
[
"-gen-dialect-defs",
"-dialect=dlti",
],
"include/mlir/Dialect/DLTI/DLTIDialect.cpp.inc",
),
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/DLTI/DLTIBase.td",
deps = [":OpBaseTdFiles"],
)
cc_library(
name = "DLTIDialect",
srcs = glob(["lib/Dialect/DLTI/*.cpp"]),
hdrs = glob(["include/mlir/Dialect/DLTI/*.h"]),
includes = ["include"],
deps = [
":DLTIBaseIncGen",
":DataLayoutInterfaces",
":IR",
"//llvm:Support",
],
)
gentbl_cc_library(
name = "ReducerIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-pass-decls",
"-name=Reducer",
],
"include/mlir/Reducer/Passes.h.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Reducer/Passes.td",
deps = [
":PassBaseTdFiles",
":ReducerTdFiles",
],
)
cc_library(
name = "Reducer",
srcs = glob(["lib/Reducer/*.cpp"]),
hdrs = glob(["include/mlir/Reducer/*.h"]),
includes = ["include"],
deps = [
":IR",
":Pass",
":ReducerIncGen",
":Rewrite",
":Support",
":TransformUtils",
"//llvm:Support",
],
)
cc_library(
name = "MlirReduceLib",
srcs = ["lib/Tools/mlir-reduce/MlirReduceMain.cpp"],
[mlir] fix bazel build breakage Reviewed By: Peiming Differential Revision: https://reviews.llvm.org/D135105
2022-10-03 14:34:14 -07:00
hdrs = [
[mlir] Fix typo s/utilties/utilities/ (including in file name) Reviewed By: rriddle Differential Revision: https://reviews.llvm.org/D136887
2022-10-27 16:45:14 -04:00
"include/mlir/Tools/ParseUtilities.h",
[mlir][Bazel] Port d85f6e5d57f38f0cbbc756484e3a93ae89b11195
2022-10-07 13:49:28 +02:00
"include/mlir/Tools/mlir-reduce/MlirReduceMain.h",
[mlir] fix bazel build breakage Reviewed By: Peiming Differential Revision: https://reviews.llvm.org/D135105
2022-10-03 14:34:14 -07:00
],
Introduce a Bazel build configuration This patch introduces configuration for a Bazel BUILD in a side directory in the monorepo. This is following the approval of https://github.com/llvm/llvm-www/blob/main/proposals/LP0002-BazelBuildConfiguration.md As detailed in the README, the Bazel BUILD is not supported by the community in general, and is maintained only by interested parties. It follows the requirements of the LLVM peripheral tier: https://llvm.org/docs/SupportPolicy.html#peripheral-tier. This is largely copied from https://github.com/google/llvm-bazel, with a few filepath tweaks and the addition of the README. Reviewed By: echristo, keith, dblaikie, kuhar Differential Revision: https://reviews.llvm.org/D90352
2021-05-18 15:42:25 -07:00
includes = ["include"],
deps = [
":IR",
":Parser",
":Pass",
":Reducer",
":Rewrite",
":Support",
"//llvm:Support",
],
)
cc_binary(
name = "mlir-reduce",
srcs = ["tools/mlir-reduce/mlir-reduce.cpp"],
includes = ["include"],
local_defines = ["MLIR_INCLUDE_TESTS"],
stamp = 0,
deps = [
":AllPassesAndDialects",
":IR",
":MlirReduceLib",
"//llvm:Support",
"//mlir/test:TestDialect",
],
)
[mlir][PDLL] Add an initial frontend for PDLL This is a new pattern rewrite frontend designed from the ground up to support MLIR constructs, and to target PDL. This frontend language was proposed in https://llvm.discourse.group/t/rfc-pdll-a-new-declarative-rewrite-frontend-for-mlir/4798 This commit starts sketching out the base structure of the frontend, and is intended to be a minimal starting point for building up the language. It essentially contains support for defining a pattern, variables, and erasing an operation. The features mentioned in the proposal RFC (including IDE support) will be added incrementally in followup commits. I intend to upstream the documentation for the language in a followup when a bit more of the pieces have been landed. Differential Revision: https://reviews.llvm.org/D115093
2021-12-16 01:48:19 +00:00
Bazel fixes for 81f2f4dfb2922e4f7af8bbfd8b653eda7c1f1339
2022-03-04 01:02:18 +00:00
cc_library(
name = "PDLLODS",
srcs = glob(
[
"lib/Tools/PDLL/ODS/*.cpp",
"lib/Tools/PDLL/ODS/*.h",
],
),
hdrs = glob(["include/mlir/Tools/PDLL/ODS/*.h"]),
includes = ["include"],
deps = [
":Support",
"//llvm:Support",
[bazel] add TargetParser
2022-12-20 20:09:00 +01:00
"//llvm:TargetParser",
Bazel fixes for 81f2f4dfb2922e4f7af8bbfd8b653eda7c1f1339
2022-03-04 01:02:18 +00:00
],
)
[mlir][PDLL] Add an initial frontend for PDLL This is a new pattern rewrite frontend designed from the ground up to support MLIR constructs, and to target PDL. This frontend language was proposed in https://llvm.discourse.group/t/rfc-pdll-a-new-declarative-rewrite-frontend-for-mlir/4798 This commit starts sketching out the base structure of the frontend, and is intended to be a minimal starting point for building up the language. It essentially contains support for defining a pattern, variables, and erasing an operation. The features mentioned in the proposal RFC (including IDE support) will be added incrementally in followup commits. I intend to upstream the documentation for the language in a followup when a bit more of the pieces have been landed. Differential Revision: https://reviews.llvm.org/D115093
2021-12-16 01:48:19 +00:00
cc_library(
name = "PDLLAST",
srcs = glob(
[
"lib/Tools/PDLL/AST/*.cpp",
"lib/Tools/PDLL/AST/*.h",
],
),
hdrs = glob(["include/mlir/Tools/PDLL/AST/*.h"]),
includes = ["include"],
deps = [
[mlir][Bazel] Update BUILD.bazel file
2022-01-25 12:33:42 +01:00
":Support",
[mlir][PDLL] Add an initial frontend for PDLL This is a new pattern rewrite frontend designed from the ground up to support MLIR constructs, and to target PDL. This frontend language was proposed in https://llvm.discourse.group/t/rfc-pdll-a-new-declarative-rewrite-frontend-for-mlir/4798 This commit starts sketching out the base structure of the frontend, and is intended to be a minimal starting point for building up the language. It essentially contains support for defining a pattern, variables, and erasing an operation. The features mentioned in the proposal RFC (including IDE support) will be added incrementally in followup commits. I intend to upstream the documentation for the language in a followup when a bit more of the pieces have been landed. Differential Revision: https://reviews.llvm.org/D115093
2021-12-16 01:48:19 +00:00
"//llvm:Support",
[bazel] add TargetParser
2022-12-20 20:09:00 +01:00
"//llvm:TargetParser",
[mlir][PDLL] Add an initial frontend for PDLL This is a new pattern rewrite frontend designed from the ground up to support MLIR constructs, and to target PDL. This frontend language was proposed in https://llvm.discourse.group/t/rfc-pdll-a-new-declarative-rewrite-frontend-for-mlir/4798 This commit starts sketching out the base structure of the frontend, and is intended to be a minimal starting point for building up the language. It essentially contains support for defining a pattern, variables, and erasing an operation. The features mentioned in the proposal RFC (including IDE support) will be added incrementally in followup commits. I intend to upstream the documentation for the language in a followup when a bit more of the pieces have been landed. Differential Revision: https://reviews.llvm.org/D115093
2021-12-16 01:48:19 +00:00
],
)
[bazel] Port g95b4e88b1db3
2022-02-27 00:11:42 +01:00
cc_library(
name = "PDLLCodeGen",
srcs = glob(
[
"lib/Tools/PDLL/CodeGen/*.cpp",
"lib/Tools/PDLL/CodeGen/*.h",
],
),
hdrs = glob(["include/mlir/Tools/PDLL/CodeGen/*.h"]),
includes = ["include"],
deps = [
[bazel] Add new rule for c60b897d22b2feab3282c4fc2b390bc87560c7de
2022-07-25 20:26:19 -07:00
":AsmParser",
[bazel] Port g95b4e88b1db3
2022-02-27 00:11:42 +01:00
":IR",
":PDLDialect",
":PDLLAST",
More bazel fixes
2022-03-04 04:15:38 +00:00
":PDLLODS",
[bazel] Port g95b4e88b1db3
2022-02-27 00:11:42 +01:00
":Support",
"//llvm:Support",
],
)
[mlir][PDLL] Add an initial frontend for PDLL This is a new pattern rewrite frontend designed from the ground up to support MLIR constructs, and to target PDL. This frontend language was proposed in https://llvm.discourse.group/t/rfc-pdll-a-new-declarative-rewrite-frontend-for-mlir/4798 This commit starts sketching out the base structure of the frontend, and is intended to be a minimal starting point for building up the language. It essentially contains support for defining a pattern, variables, and erasing an operation. The features mentioned in the proposal RFC (including IDE support) will be added incrementally in followup commits. I intend to upstream the documentation for the language in a followup when a bit more of the pieces have been landed. Differential Revision: https://reviews.llvm.org/D115093
2021-12-16 01:48:19 +00:00
cc_library(
name = "PDLLParser",
srcs = glob(
[
"lib/Tools/PDLL/Parser/*.cpp",
"lib/Tools/PDLL/Parser/*.h",
],
),
hdrs = glob(["include/mlir/Tools/PDLL/Parser/*.h"]),
includes = ["include"],
deps = [
":PDLLAST",
Bazel fixes for 81f2f4dfb2922e4f7af8bbfd8b653eda7c1f1339
2022-03-04 01:02:18 +00:00
":PDLLODS",
[mlir][PDLL] Add an initial frontend for PDLL This is a new pattern rewrite frontend designed from the ground up to support MLIR constructs, and to target PDL. This frontend language was proposed in https://llvm.discourse.group/t/rfc-pdll-a-new-declarative-rewrite-frontend-for-mlir/4798 This commit starts sketching out the base structure of the frontend, and is intended to be a minimal starting point for building up the language. It essentially contains support for defining a pattern, variables, and erasing an operation. The features mentioned in the proposal RFC (including IDE support) will be added incrementally in followup commits. I intend to upstream the documentation for the language in a followup when a bit more of the pieces have been landed. Differential Revision: https://reviews.llvm.org/D115093
2021-12-16 01:48:19 +00:00
":Support",
":TableGen",
"//llvm:Support",
Bazel fixes for 81f2f4dfb2922e4f7af8bbfd8b653eda7c1f1339
2022-03-04 01:02:18 +00:00
"//llvm:TableGen",
[bazel] Add missing :Support dependency after 1b8224537070
2023-01-06 12:51:01 +01:00
"//llvm:TargetParser",
[mlir][PDLL] Add an initial frontend for PDLL This is a new pattern rewrite frontend designed from the ground up to support MLIR constructs, and to target PDL. This frontend language was proposed in https://llvm.discourse.group/t/rfc-pdll-a-new-declarative-rewrite-frontend-for-mlir/4798 This commit starts sketching out the base structure of the frontend, and is intended to be a minimal starting point for building up the language. It essentially contains support for defining a pattern, variables, and erasing an operation. The features mentioned in the proposal RFC (including IDE support) will be added incrementally in followup commits. I intend to upstream the documentation for the language in a followup when a bit more of the pieces have been landed. Differential Revision: https://reviews.llvm.org/D115093
2021-12-16 01:48:19 +00:00
],
)
cc_binary(
name = "mlir-pdll",
srcs = [
"tools/mlir-pdll/mlir-pdll.cpp",
],
deps = [
[bazel] Port g95b4e88b1db3
2022-02-27 00:11:42 +01:00
":IR",
[mlir][PDLL] Add an initial frontend for PDLL This is a new pattern rewrite frontend designed from the ground up to support MLIR constructs, and to target PDL. This frontend language was proposed in https://llvm.discourse.group/t/rfc-pdll-a-new-declarative-rewrite-frontend-for-mlir/4798 This commit starts sketching out the base structure of the frontend, and is intended to be a minimal starting point for building up the language. It essentially contains support for defining a pattern, variables, and erasing an operation. The features mentioned in the proposal RFC (including IDE support) will be added incrementally in followup commits. I intend to upstream the documentation for the language in a followup when a bit more of the pieces have been landed. Differential Revision: https://reviews.llvm.org/D115093
2021-12-16 01:48:19 +00:00
":PDLLAST",
[bazel] Port g95b4e88b1db3
2022-02-27 00:11:42 +01:00
":PDLLCodeGen",
[mlir][Bazel] Adjust regarding StandardToLLVM -> FuncToLLVM rename. Differential Revision: https://reviews.llvm.org/D121185
2022-03-08 08:55:55 +01:00
":PDLLODS",
[mlir][PDLL] Add an initial frontend for PDLL This is a new pattern rewrite frontend designed from the ground up to support MLIR constructs, and to target PDL. This frontend language was proposed in https://llvm.discourse.group/t/rfc-pdll-a-new-declarative-rewrite-frontend-for-mlir/4798 This commit starts sketching out the base structure of the frontend, and is intended to be a minimal starting point for building up the language. It essentially contains support for defining a pattern, variables, and erasing an operation. The features mentioned in the proposal RFC (including IDE support) will be added incrementally in followup commits. I intend to upstream the documentation for the language in a followup when a bit more of the pieces have been landed. Differential Revision: https://reviews.llvm.org/D115093
2021-12-16 01:48:19 +00:00
":PDLLParser",
":Support",
"//llvm:Support",
"//llvm:config",
],
)
[bazel][mlir] Add MLIR PDLL LSP server target Add targets for PDLL LSP server.
2022-03-19 17:53:37 -07:00
cc_binary(
name = "mlir-pdll-lsp-server",
srcs = ["tools/mlir-pdll-lsp-server/mlir-pdll-lsp-server.cpp"],
includes = ["include"],
deps = [
":MlirPdllLspServerLib",
":Support",
],
)
Add missing PDLL libs.
2022-11-08 12:14:32 +01:00
td_library(
name = "DialectConversionPdllFiles",
srcs = ["include/mlir/Transforms/DialectConversion.pdll"],
includes = ["include"],
)
[mlir] Add UBDialect to BUILD.bazel.
2023-07-20 15:08:19 +02:00
td_library(
name = "UBDialectTdFiles",
srcs = [
"include/mlir/Dialect/UB/IR/UBOps.td",
"include/mlir/Dialect/UB/IR/UBOpsInterfaces.td",
],
includes = ["include"],
deps = [
":OpBaseTdFiles",
":SideEffectInterfacesTdFiles",
],
)
gentbl_cc_library(
name = "UBOpsInterfacesIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
["--gen-attr-interface-decls"],
"include/mlir/Dialect/UB/IR/UBOpsInterfaces.h.inc",
),
(
["--gen-attr-interface-defs"],
"include/mlir/Dialect/UB/IR/UBOpsInterfaces.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/UB/IR/UBOpsInterfaces.td",
deps = [":UBDialectTdFiles"],
)
gentbl_cc_library(
name = "UBOpsIncGen",
strip_include_prefix = "include",
tbl_outs = [
(
[
"-gen-dialect-decls",
"-dialect=ub",
],
"include/mlir/Dialect/UB/IR/UBOpsDialect.h.inc",
),
(
[
"-gen-dialect-defs",
"-dialect=ub",
],
"include/mlir/Dialect/UB/IR/UBOpsDialect.cpp.inc",
),
(
["-gen-op-decls"],
"include/mlir/Dialect/UB/IR/UBOps.h.inc",
),
(
["-gen-op-defs"],
"include/mlir/Dialect/UB/IR/UBOps.cpp.inc",
),
(
["--gen-attrdef-decls"],
"include/mlir/Dialect/UB/IR/UBOpsAttributes.h.inc",
),
(
["--gen-attrdef-defs"],
"include/mlir/Dialect/UB/IR/UBOpsAttributes.cpp.inc",
),
],
tblgen = ":mlir-tblgen",
td_file = "include/mlir/Dialect/UB/IR/UBOps.td",
deps = [":UBDialectTdFiles"],
)
cc_library(
name = "UBDialect",
srcs = ["lib/Dialect/UB/IR/UBOps.cpp"],
hdrs = ["include/mlir/Dialect/UB/IR/UBOps.h"],
includes = ["include"],
deps = [
":IR",
":SideEffectInterfaces",
":UBOpsIncGen",
":UBOpsInterfacesIncGen",
[bazel] Fix dependency path
2023-07-20 17:05:28 +02:00
"//llvm:Support",
[mlir] Add UBDialect to BUILD.bazel.
2023-07-20 15:08:19 +02:00
],
)
[bazel][mlir] Port 0f446adf678212b6daedebc5eb1bd2849b1efad1
2023-07-24 19:21:18 +02:00
cc_library(
name = "UBToLLVM",
srcs = glob([
"lib/Conversion/UBToLLVM/*.cpp",
"lib/Conversion/UBToLLVM/*.h",
]),
hdrs = glob([
"include/mlir/Conversion/UBToLLVM/*.h",
]),
includes = ["include"],
deps = [
":ConversionPassIncGen",
":IR",
":LLVMCommonConversion",
":LLVMDialect",
":Pass",
":UBDialect",
],
)
[bazel] Add UPToSPIRV library/dep Added in D156163
2023-07-24 19:33:33 -07:00
cc_library(
name = "UBToSPIRV",
srcs = glob([
"lib/Conversion/UBToSPIRV/*.cpp",
"lib/Conversion/UBToSPIRV/*.h",
]),
hdrs = glob([
"include/mlir/Conversion/UBToSPIRV/*.h",
]),
includes = ["include"],
deps = [
":ConversionPassIncGen",
":Pass",
":SPIRVConversion",
":SPIRVDialect",
":UBDialect",
"//llvm:Core",
],
)
Reference in New Issue Copy Permalink