Vinayaka Bandishti
01c755ff80
Make optimize llvm common to both gpu-to-hsaco/cubin
...
Before serializing, optimizations on llvm were only called on path to
hsaco, and not cubin. Define opt-level for `gpu-to-cubin` pass as well,
and move call to optimize llvm to a common place.
Reviewed By: bondhugula
Differential Revision: https://reviews.llvm.org/D151554
2023-06-05 10:32:51 +05:30
Artem Belevich
d4ba4c6af7
Revert unintentionally committed "Use nvptxcompile library."
...
This reverts commit 5f66348e59aa7ce5e5780a972b3875268c45d57c.
2023-03-17 14:23:42 -07:00
Artem Belevich
5f66348e59
Use nvptxcompile library.
...
Differential Revision: https://reviews.llvm.org/D145527
2023-03-17 14:08:53 -07:00
Ivan Radanov Ivanov
e01c7f092f
[MLIR] Revert default NVIDIA GPU version
...
Due to integration tests failing revert mlir::SerializeToCubinPass defaults to old ones (changed in https://reviews.llvm.org/D134153 )
Reviewed By: akuegel
Differential Revision: https://reviews.llvm.org/D134414
2022-09-22 10:19:38 +02:00
Ivan Radanov Ivanov
f9211330f6
[MLIR] Set default NVIDIA GPU version
2022-09-21 18:10:59 -04:00
Ivan Radanov Ivanov
2f7a774ed7
[MLIR] Add a create function for mlir::SerializeToCubinPass
...
Differential Revision: https://reviews.llvm.org/D134153
2022-09-21 18:02:59 -04:00
Jeff Niu
b7f93c2809
[mlir] (NFC) run clang-format on all files
2022-07-14 13:32:13 -07:00
Mogball
d7ef488bb6
[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 22:49:03 +00:00
River Riddle
1269f96d2e
[mlir] Add MLIR_DEFINE_EXPLICIT_INTERNAL_INLINE_TYPE_ID to SerializeToCubinPass
...
This pass is defined in an anonymous namespace and requires an explicit TypeID
2022-04-04 14:28:10 -07:00
Mehdi Amini
b5e22e6d42
Migrate MLIR test passes to the new registration API
...
Make sure they all define getArgument()/getDescription().
Depends On D104421
Differential Revision: https://reviews.llvm.org/D104426
2021-06-16 23:42:17 +00:00
Christian Sigg
2224221fb3
[mlir] Add NVVM to CUBIN conversion to mlir-opt
...
If MLIR_CUDA_RUNNER_ENABLED, register a 'gpu-to-cubin' conversion pass to mlir-opt.
The next step is to switch CUDA integration tests from mlir-cuda-runner to mlir-opt + mlir-cpu-runner and remove mlir-cuda-runner.
Depends On D98279
Reviewed By: herhut, rriddle, mehdi_amini
Differential Revision: https://reviews.llvm.org/D98203
2021-03-11 10:07:11 +01:00
