Add gfx1150/gfx1151 to the supported arch list for hipBLASLt #2580
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… test_c10d_nccl.py (#2522) Cherry-pick of #2447 Co-authored-by: akashveramd <[email protected]>
… rocm version (#2529) Cherry-pick of #2518 Co-authored-by: Ethan Wee <[email protected]>
Cherry-pick of #2450 --------- Co-authored-by: iupaikov-amd <[email protected]>
Fixes SWDEV-543698 (https://ontrack-internal.amd.com/browse/SWDEV-543698) Cherry-picked from #2502 This PR fixes the errors like below: ``` [rank3]: RuntimeError: The following operation failed in the TorchScript interpreter. [rank3]: Traceback of TorchScript (most recent call last): [rank3]: RuntimeError: /tmp/comgr-28f951/input/CompileSourceACC062:67:7: error: unknown type name 'uint32_t'; did you mean '__hip_internal::uint32_t'? [rank3]: 67 | uint32_t int32; [rank3]: | ^~~~~~~~ [rank3]: | __hip_internal::uint32_t ``` Earlier uint32_t was defined in HIP headers in std namespace. Now it is moved to __hip_internal namespace in hip headers. This change is made in ROCm 7.0.
Cherry-pick of #2535 --------- Co-authored-by: Ethan Wee <[email protected]>
…2556) Cherry-pick of #2554 Co-authored-by: Jerry Mannil <[email protected]>
#2561) Cherry-pick of #2553 Co-authored-by: Jerry Mannil <[email protected]>
…IOpen symbols (#2415) Changing TORCH_CUDA_CPP_API macros to TORCH_HIP_CPP in MIOpen header files according to closed [PR](pytorch#156479) in pytorch upstream. --------- Signed-off-by: Jagadish Krishnamoorthy <[email protected]> Co-authored-by: Jithun Nair <[email protected]> Co-authored-by: Ethan Wee <[email protected]> Co-authored-by: Jithun Nair <[email protected]> Co-authored-by: Xinya Zhang <[email protected]> Co-authored-by: Prachi Gupta <[email protected]> Co-authored-by: Dmitry Nikolaev <[email protected]> Co-authored-by: iupaikov-amd <[email protected]> Co-authored-by: omkar kakarparthi <[email protected]> Co-authored-by: Jeff Daily <[email protected]> Co-authored-by: rocm-mici <[email protected]> Co-authored-by: Michael Halkenhäuser <[email protected]> Co-authored-by: Hashem Hashemi <[email protected]> Co-authored-by: Jagadish Krishnamoorthy <[email protected]> Co-authored-by: Andres Lugo <[email protected]> Co-authored-by: pmaybank <[email protected]> Co-authored-by: pnikolic-amd <[email protected]> Co-authored-by: Jerry Mannil <[email protected]> Co-authored-by: Sampsa Riikonen <[email protected]> Co-authored-by: Ramya Ramineni <[email protected]> Co-authored-by: pytorchbot <[email protected]> Co-authored-by: Nikita Shulga <[email protected]> Co-authored-by: rzou <[email protected]> Co-authored-by: atalman <[email protected]> Co-authored-by: David Berard <[email protected]> Co-authored-by: Frank Lin <[email protected]> Co-authored-by: Wei Wang <[email protected]> Co-authored-by: Cao E <[email protected]> Co-authored-by: Sidharth Subbarao <[email protected]> Co-authored-by: Camyll Harajli <[email protected]> Co-authored-by: Aditya Tewari <[email protected]> Co-authored-by: Christopher Sidebottom <[email protected]> Co-authored-by: Ryo Suzuki <[email protected]> Co-authored-by: Ye Tao <[email protected]> Co-authored-by: Nikita Shulga <[email protected]> Co-authored-by: Eli Uriegas <[email protected]> Co-authored-by: Luca Wehrstedt <[email protected]> Co-authored-by: Robert Hardwick <[email protected]> Co-authored-by: Svetlana Karslioglu <[email protected]> Co-authored-by: Michael Lazos <[email protected]> Co-authored-by: bobrenjc93 <[email protected]> Co-authored-by: Nikita Shulga <[email protected]> Co-authored-by: Jane Xu <[email protected]> Co-authored-by: charlifu <[email protected]> Co-authored-by: Jack Taylor <[email protected]> Co-authored-by: Nichols A. Romero <[email protected]> Co-authored-by: Copilot <[email protected]> Co-authored-by: Sriram Kumar <[email protected]> Co-authored-by: Divin Honnappa <[email protected]> Co-authored-by: akashveramd <[email protected]>
This commit fixes Windows build issue related to trying to rocm-core/rocm_version.h or hip/hip_version.h. This is cherry-picked commit from upstream main ([PR](pytorch#156486))
mstankov-amd
requested review from
jeffdaily and
jithunnair-amd
as code owners
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PyTorch uses hipBLASLt based on a filter function. If the architecture is not in the list, it will automatically switch to hipBLAS.
With this PR, we add gfx1150/gfx1151 to the supported list.