[MLIR][XeGPU] Add unroll patterns and blocking pass for XeGPU [2/N] #140163
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
chencha3
changed the title
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
chencha3
changed the title
chencha3
changed the title
chencha3
changed the title
Garra1980
reviewed
May 16, 2025
Garra1980
reviewed
May 16, 2025
Garra1980
reviewed
May 16, 2025
Garra1980
reviewed
May 16, 2025
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/140/builds/24257 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/116/builds/13634 Here is the relevant piece of the build log for the reference |
ronlieb
pushed a commit
that referenced
this pull request
Jun 2, 2025
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/55/builds/12220 Here is the relevant piece of the build log for the reference |
llvm-sync bot
pushed a commit
to arm/arm-toolchain
that referenced
this pull request
Jun 2, 2025
…s for XeGPU [2/N]" (#142459) Reverts llvm/llvm-project#140163
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/35/builds/10589 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/169/builds/11924 Here is the relevant piece of the build log for the reference |
|
(Sorry, I missed that you had already reverted this. Thanks!) This patch has caused a use-after-free: |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/130/builds/13535 Here is the relevant piece of the build log for the reference |
No worries. I am sorry for the inconvenience. |
llvm-sync bot
pushed a commit
to arm/arm-toolchain
that referenced
this pull request
Jun 3, 2025
…GPU [2/N] (#142477) Bring back llvm/llvm-project#140163 with fixes
sallto
pushed a commit
to sallto/llvm-project
that referenced
this pull request
Jun 3, 2025
…lvm#140163) This PR introduces the initial implementation of a blocking pass for XeGPU programs. The pass leverages unroll patterns from both the XeGPU and Vector dialects. --------- Co-authored-by: Adam Siemieniuk <[email protected]>
sallto
pushed a commit
to sallto/llvm-project
that referenced
this pull request
Jun 3, 2025
sallto
pushed a commit
to sallto/llvm-project
that referenced
this pull request
Jun 3, 2025
…lvm#142477) Bring back llvm#140163 with fixes
llvm-sync bot
pushed a commit
to arm/arm-toolchain
that referenced
this pull request
Jun 10, 2025
…and transpose (#143389) This PR adds blocking support for vector dialect operations (`reduce`, `broadcast`, and `transpose`) in the XeGPU based IR. It simply assigned the shape specified by "inst_data" as its target shape of the unrolling to implement the blocking. It is based on llvm/llvm-project#140163.
rorth
pushed a commit
to rorth/llvm-project
that referenced
this pull request
Jun 11, 2025
…lvm#140163) This PR introduces the initial implementation of a blocking pass for XeGPU programs. The pass leverages unroll patterns from both the XeGPU and Vector dialects. --------- Co-authored-by: Adam Siemieniuk <[email protected]>
rorth
pushed a commit
to rorth/llvm-project
that referenced
this pull request
Jun 11, 2025
rorth
pushed a commit
to rorth/llvm-project
that referenced
this pull request
Jun 11, 2025
…lvm#142477) Bring back llvm#140163 with fixes
rorth
pushed a commit
to rorth/llvm-project
that referenced
this pull request
Jun 11, 2025
…se (llvm#143389) This PR adds blocking support for vector dialect operations (`reduce`, `broadcast`, and `transpose`) in the XeGPU based IR. It simply assigned the shape specified by "inst_data" as its target shape of the unrolling to implement the blocking. It is based on llvm#140163.
DhruvSrivastavaX
pushed a commit
to DhruvSrivastavaX/lldb-for-aix
that referenced
this pull request
Jun 12, 2025
…lvm#140163) This PR introduces the initial implementation of a blocking pass for XeGPU programs. The pass leverages unroll patterns from both the XeGPU and Vector dialects. --------- Co-authored-by: Adam Siemieniuk <[email protected]>
DhruvSrivastavaX
pushed a commit
to DhruvSrivastavaX/lldb-for-aix
that referenced
this pull request
Jun 12, 2025
DhruvSrivastavaX
pushed a commit
to DhruvSrivastavaX/lldb-for-aix
that referenced
this pull request
Jun 12, 2025
…lvm#142477) Bring back llvm#140163 with fixes
tomtor
pushed a commit
to tomtor/llvm-project
that referenced
this pull request
Jun 14, 2025
…se (llvm#143389) This PR adds blocking support for vector dialect operations (`reduce`, `broadcast`, and `transpose`) in the XeGPU based IR. It simply assigned the shape specified by "inst_data" as its target shape of the unrolling to implement the blocking. It is based on llvm#140163.
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.
This PR introduces the initial implementation of a blocking pass for XeGPU programs. The pass leverages unroll patterns from both the XeGPU and Vector dialects. It requires that every
OpResultis associated with a validLayoutAttrfor whichisSgLayout()returns true. The pass uses theinst_datafield to determine the target shape for unrolling; ifinst_datais absent, unrolling for that Value is skipped. In this PR, only the following ops are enabled:createNd,Dpas,LoadNd,StoreNd,PrefetchNd,UpdateNd, andelementwise.