Given that FMV support is required for the SME builtins to be built, the
FMV constructor as defined in:
compiler-rt/lib/builtins/cpu_model/aarch64.c
already initialises the feature bits, so there's no need to create
another one.
Based on the feedback from #118782, this switches most of the pointer
arithmetic in __llvm_profile_merge_from_buffer to work on uintptr_t
instead of const char *, only casting back to a pointer when performing
actual accesses.
This ensures that all the arithmetic is performed without any
assumptions about pointer overflow.
This re-applies 570ecdcf8b4, which was reverted in 74e8a37ff32 due to bot
failures. This commit renames sysv_resolve.cpp to resolve.cpp, which was the
cause of the config errors.
The unit tests may transiently depend on __orc_rt_log_error, which is usually
provided by an alias when loading the runtime through the JIT. The unit tests
statically link the runtime, so this patch provides a fixed definition.
This reapplies 570ecdcf8b4, which was reverted in 6073dd923b8 due to bot
failures.
The test failures on Linux were fixed by:
1. Removing an overly restrictive assertion (query dependence on a symbol no
longer implies a MaterializingInfo for that symbol)
2. Adding reentry and resolver files to the ORC runtime CMakeLists.txt for
Linux.
3. Adding the __orc_rt_reentry -> __orc_rt_sysv_reentry alias to ELFNixPlatform.
Since the sanitizer merge in commit r15-5164-gfa321004f3f628 of GCC
which entails LLVM commit 61a6439f35b6de28ff4aff4450d6fca970292fd5, GCCs
bootstrap is broken on s390 -m31. This is due to commit
ec68dc1ca4d967b599f1202855917d5ec9cae52f which introduces stricter type
checking which is why GCC bootstrap fails with
```
In file included from /gcc/src/libsanitizer/interception/interception.h:18,
from /gcc/src/libsanitizer/interception/interception_type_test.cpp:14:
/gcc/src/libsanitizer/interception/interception_type_test.cpp:30:61: error: static assertion failed
30 | COMPILER_CHECK((__sanitizer::is_same<::SSIZE_T, ::ssize_t>::value));
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~
/gcc/src/libsanitizer/sanitizer_common/sanitizer_internal_defs.h:363:44: note: in definition of macro 'COMPILER_CHECK'
363 | #define COMPILER_CHECK(pred) static_assert(pred, "")
| ^~~~
make[8]: *** [Makefile:469: interception_type_test.lo] Error 1
```
The culprit seems to be that we don't check for equality of type sizes
anymore but rather whether the types are indeed the same. On s390 -m31
we have that `sizeof(int)==sizeof(long)` holds which is why previously
the checks succeeded. They fail now because
```
size_t => unsigned long
ssize_t => long
ptrdiff_t => int
::SSIZE_T => __sanitizer::sptr => int
::PTRDIFF_T => __sanitizer::sptr => int
```
This is fixed by mapping `SSIZE_T` to `long` in the end.
```
#if defined(__s390__) && !defined(__s390x__)
typedef long ssize;
#else
typedef sptr ssize;
#endif
#define SSIZE_T __sanitizer::ssize
```
For some targets uptr is mapped to unsigned int and size_t to unsigned
long and sizeof(int)==sizeof(long) holds. Still, these are distinct
types and type checking may fail. Therefore, replace uptr by
usize/SIZE_T wherever a size_t is expected.
Part of #116957
…d reentry.
These utilities provide new, more generic and easier to use support for
lazy compilation in ORC.
LazyReexportsManager is an alternative to LazyCallThroughManager. It
takes requests for lazy re-entry points in the form of an alias map:
lazy-reexports = {
( <entry point symbol #1>, <implementation symbol #1> ),
( <entry point symbol #2>, <implementation symbol #2> ),
...
( <entry point symbol #n>, <implementation symbol #n> )
}
LazyReexportsManager then:
1. binds the entry points to the implementation names in an internal
table.
2. creates a JIT re-entry trampoline for each entry point.
3. creates a redirectable symbol for each of the entry point name and
binds redirectable symbol to the corresponding reentry trampoline.
When an entry point symbol is first called at runtime (which may be on
any thread of the JIT'd program) it will re-enter the JIT via the
trampoline and trigger a lookup for the implementation symbol stored in
LazyReexportsManager's internal table. When the lookup completes the
entry point symbol will be updated (via the RedirectableSymbolManager)
to point at the implementation symbol, and execution will proceed to the
implementation symbol.
Actual construction of the re-entry trampolines and redirectable symbols
is delegated to an EmitTrampolines functor and the
RedirectableSymbolsManager respectively.
JITLinkReentryTrampolines.h provides a JITLink-based implementation of
the EmitTrampolines functor. (AArch64 only in this patch, but other
architectures will be added in the near future).
Register state save and reentry functionality is added to the ORC
runtime in the __orc_rt_sysv_resolve and __orc_rt_resolve_implementation
functions (the latter is generic, the former will need custom
implementations for each ABI and architecture to be supported, however
this should be much less effort than the existing OrcABISupport
approach, since the ORC runtime allows this code to be written as native
assembly).
The resulting system:
1. Works equally well for in-process and out-of-process JIT'd code.
2. Requires less boilerplate to set up.
Given an ObjectLinkingLayer and PlatformJD (JITDylib containing the ORC
runtime), setup is just:
```c++
auto RSMgr = JITLinkRedirectableSymbolManager::Create(OLL);
if (!RSMgr)
return RSMgr.takeError();
auto LRMgr = createJITLinkLazyReexportsManager(OLL, **RSMgr, PlatformJD);
if (!LRMgr)
return LRMgr.takeError();
```
after which lazy reexports can be introduced with:
```c++
JD.define(lazyReexports(LRMgr, <alias map>));
```
LazyObectLinkingLayer is updated to use this new method, but the LLVM-IR
level CompileOnDemandLayer will continue to use LazyCallThroughManager
and OrcABISupport until the new system supports a wider range of
architectures and ABIs.
The llvm-jitlink utility's -lazy option now uses the new scheme. Since
it depends on the ORC runtime, the lazy-link.ll testcase and associated
helpers are moved to the ORC runtime.
Add the `__memprof_default_options_str` variable, initialized via the
`-memprof-runtime-default-options` LLVM flag, to hold the default options string
for memprof. This allows us to set these options during compile time in
the clang invocation.
Also update the docs to describe the various ways to set these options.
These bounds checks work on the result of the pointer addition -- but
the pointer addition already asserts that no overflow may occur, so the
checks are optimized away after #118472. Avoid this by performing the
addition in a way that permits overflow.
warning: format specifies type 'void *' but the argument has type 'uptr'
(aka 'unsigned long long') [-Wformat] (observed at x86_64, in
AllocateTrampolineRegion)
warning: format specifies type 'char *' but the argument has type
'RVAPtr<char>' [-Wformat] (observed at x86_64, in
InternalGetProcAddress)
This patch extends the PGO infrastructure with an option to prefer the
instrumentation of loop entry blocks.
This option is a generalization of
19fb5b467b,
and helps to cover cases where the loop exit is never executed.
An example where this can occur are event handling loops.
Note that change does NOT change the default behavior.
At some point FreeBSD introduced libsys as a wrapper between syscalls
and libc, and then linking sanitized programs started failing with:
```text
# c++ -fsanitize=address main.cc
ld: error: undefined symbol: __elf_aux_vector
>>> referenced by sanitizer_linux_libcdep.cpp:950 (/usr/src/contrib/llvm-project/compiler-rt/lib/sanitizer_common/sanitizer_linux_libcdep.cpp:950)
>>> sanitizer_linux_libcdep.o:(__sanitizer::ReExec()) in archive /usr/lib/clang/17/lib/freebsd/libclang_rt.asan-x86_64.a
c++: error: linker command failed with exit code 1 (use -v to see invocation)
```
Mark `__elf_aux_vector` as weak in the internal sanitizer declaration,
so the linker will accept it at link time. The dynamic linker will then
take care of the symbol at run time.
Since #105495, we will build sanitizer_common when
COMPILER_RT_BUILD_CTX_PROFILE is set. But we have outstanding failures
when building sanitizer_common, so for now we'll disable ctx_profile
too.
I was trying to build compiler-rt with /usr/bin/clang-17 and the testsuite
failed due to the code in lit.common.cfg.py:
```
# GCC-ASan uses dynamic runtime by default (since config.bits is not set).
if config.compiler_id == "GNU":
gcc_dir = os.path.dirname(config.clang)
libasan_dir = os.path.join(gcc_dir, "..", "lib" + config.bits)
push_dynamic_library_lookup_path(config, libasan_dir)
```
Fix this in two ways: First, if the test compiler matches the library
compiler, set COMPILER_RT_TEST_COMPILER_ID to CMAKE_C_COMPILER_ID. Second,
relax the regex detecting clang to allow any kind of suffix.
Reviewed By: compnerd
Pull Request: https://github.com/llvm/llvm-project/pull/117812
We don't need 128-bit integers here, instead rewrite the fp_test.h helpers
to avoid the need for __int128. Also change the high argument for
compareResultF80() and F80FromRep80() to be uint16_t since values greater
than this do not make any sense. This should allow the compiler to detect
accidentally swapping the arguments.
Reviewed By: biabbas, alexander-shaposhnikov
Pull Request: https://github.com/llvm/llvm-project/pull/117818
This is a complex one - `syscall` is used when people want to bypass
libc and make the call directly
However, this call:
* Has a variable amount of arguments (up to 6, typically)
* Has arguments that can be any type, (think of whatever arguments go in
to the libc call, or see more details here
https://syscalls.mebeim.net/?table=x86/64/x64/latest)
I've tried to put in a couple tests to ensure we aren't mucking with the
underlying functionality and they seem to be working.
It appears already some lines above with this comment:
"Cannot overwrite control-instruction. Return 0 to indicate failure.".
Replacing just the comment in the first appearance.
Found after creating the test in #113085.
Similar to `__asan_default_options`, users can specify default options
upon building the instrumented binaries by providing their own
definition of `__xray_default_options` which returns the option strings.
This is useful in cases where setting the `XRAY_OPTIONS` environment
variable might be difficult. Plus, it's a convenient way to populate
XRay options when you always want the instrumentation to be enabled.
It turns out we were not passing -m32 to the check_c_source_compiles()
invocation since CMAKE_REQUIRE_FLAGS needs to be string separated list
and
we were passing a ;-separated CMake list which appears to be parsed by
CMake as 'ignore all arguments beyond the first'.
Fix this by transforming the list to a command line first.
With this change, Clang 17 no longer claims to support _Float16 for
i386.
Using a `--section-start` address in the test was causing link errors on
some targets. Now it uses a linker script to move the note after `.bss`,
which should still have the kind of memory offset we're looking for.
This is a follow-up to #114907 to fix buildbot errors.
This relands #86209 which was reverted because ./bin/llvm no longer
accepted test paths in the source tree instead of the build tree. This was
happening because `add_subdirectory(${LLVM_MAIN_SRC_DIR}/utils/llvm-lit`
was called before all tsst suites were registered, and therefore it was
missing the source->build dir mappings.
Original commit message:
I am currently trying to test the LLVM runtimes (including compiler-rt)
against an installed LLVM tree rather than a build tree (since that is
no longer available). Currently, the runtimes build of compiler-rt assumes
that LLVM_BINARY_DIR is writable since it uses configure_file() to write
there during the CMake configure stage. Instead, generate this file inside
CMAKE_CURRENT_BINARY_DIR, which will match LLVM_BINARY_DIR when invoked
from llvm/runtimes/CMakeLists.txt.
I also needed to make a minor change to the hwasan tests: hwasan_symbolize
was previously found in the LLVM_BINARY_DIR, but since it is generated as
part of the compiler-rt build it is now inside the CMake build directory
instead. I fixed this by passing the output directory to lit as
config.compiler_rt_bindir and using llvm_config.add_tool_substitutions().
For testing that we no longer write to the LLVM install directory as
part of testing or configuration, I created a read-only bind mount and
configured the runtimes builds as follows:
```
$ sudo mount --bind --read-only ~/llvm-install /tmp/upstream-llvm-readonly
$ cmake -DCMAKE_BUILD_TYPE=Debug \
-DCMAKE_C_COMPILER=/tmp/upstream-llvm-readonly/bin/clang \
-DCMAKE_CXX_COMPILER=/tmp/upstream-llvm-readonly/bin/clang++ \
-DLLVM_INCLUDE_TESTS=TRUE -DLLVM_ENABLE_ASSERTIONS=TRUE \
-DCOMPILER_RT_INCLUDE_TESTS=TRUE -DCOMPILER_RT_DEBUG=OFF \
-DLLVM_ENABLE_RUNTIMES=compiler-rt \
-DCMAKE_DISABLE_FIND_PACKAGE_LLVM=TRUE \
-DCMAKE_DISABLE_FIND_PACKAGE_Clang=TRUE \
-G Ninja -S ~/upstream-llvm-project/runtimes \
-B ~/upstream-llvm-project/runtimes/cmake-build-debug-llvm-git
```
Pull Request: https://github.com/llvm/llvm-project/pull/114307
I've run into an issue where TSan can't be used on some code without
turning off deadlock detection because a thread tries to hold too many
mutexes. It would be preferable to be able to use deadlock detection as
that is a major benefit of TSan.
Its mentioned in https://github.com/google/sanitizers/issues/950 that
the 64 mutex limit was an arbitrary number. I've increased it to 128 and
all the tests still pass. Considering the increasing number of cores on
CPUs and how programs can now use more threads to take advantage of it,
I think raising the limit to 128 would be some good future proofing
---------
Co-authored-by: Vitaly Buka <vitalybuka@google.com>
