Add `--@local_config_cuda//cuda:override_include_cuda_libs` to override settings for TF wheel.
Forbid building TF wheel with `--@local_config_cuda//cuda:include_cuda_libs=true`
PiperOrigin-RevId: 666848518
Tests fixed include:
- `test_globally_sharded_key_array_8x4_multi_device`
- Issue was in `replicate_trailing_dims` where an `xc.OpSharding` was always created. Fixed by creating an equivalent SDY sharding.
- `test_aot_out_info`
- Issue was there was no mesh since there weren't any NamedShardings. Fixed by not asserting a mesh tuple exists in `lower_jaxpr_to_module` when adding the sdy MeshOp (there won't be any propagation)
- `test_concurrent_pjit`
- In Shardy if there was a tensor dimension of size 0, we'd emit a verification error if the dimension is sharded on an axes. But if the axis is of size 1, then JAX says this is okay. So have shardy assume the same.
- `test_globally_sharded_key_array_result_8x4_single_device`
- This tests adds a WSC when no `mesh_shape_tuple` exists (`"sdy.sharding_constraint"(%8) <{sharding = #sdy.sharding<@mesh, [{?}, {?}, {}]>}>`), so we should create a mesh named `mesh` with a single device id in case it doesn't exist.
- `testLowerCostAnalysis`
- This calls into `mlir_module_to_xla_computation` which calls its own MLIR parsing function in `//third_party/tensorflow/compiler/xla/python/mlir.cc`. Needed to register the SDY dialect in it.
- `testShardingConstraintWithArray`
- This calls `.compiler_ir(dialect="hlo")` which calls `PyMlirModuleToXlaComputation` which converts the MLIR to HLO, but the Sdy dialect is still inside. Export it before converting it to HLO.
PiperOrigin-RevId: 666777167
Interactive causes bazel to output more
useful info when running locally.
* Fix issue with rocm el8 repo urls
Work around quirk with rocm version
when it ends with 0
* Fix package name conflict
Ubu22 and higher have a package name conflict
between the debian versions and the AMD provided
versions.
* [ROCm] Use clang env
The assumption is that QR Factorization will never fail from LAPACK's side because all necessary verification is happening right before the call.
PiperOrigin-RevId: 666241215
It is possible that a null pointer is inserted into the cache and not updated with a valid kernel call
in case there is an error later during initialization. This change updates the cache to store either
an error or a valid kernel call.
PiperOrigin-RevId: 666161091
Regular conversion instructions have a ridiculously low throughput on Hopper,
so replacing them with some bit tricks yields a much faster implementation.
Co-authored-by: Benjamin Chetioui <bchetioui@google.com>
PiperOrigin-RevId: 665893696
