[GPU][Codegen] Allowing mfma for narrow problem config sizes #19615
Conversation
This file contains 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
Signed-off-by: jerryyin <[email protected]>
jerryyin
requested review from
MaheshRavishankar,
qedawkins,
kuhar,
Groverkss and
antiagainst
as code owners
qedawkins
requested changes
Jan 7, 2025
compiler/src/iree/compiler/Codegen/LLVMGPU/test/ROCDL/pipeline_warp_reduction.mlir
Outdated
Show resolved
Hide resolved
compiler/src/iree/compiler/Codegen/LLVMGPU/test/ROCDL/config_vector_distribute_gfx942.mlir
Show resolved
Hide resolved
|
I experimented with resnet50 again and can confirm between:
makes no difference in full model performance. |
jerryyin
force-pushed
the
users/zyin/remove-intrinsic-size-limitation
branch
from
22102d7 to
f2d5d51
Compare
This make sure tiny gemm can still be lowered through warp reduction pipeline. Signed-off-by: jerryyin <[email protected]>
qedawkins
approved these changes
Jan 7, 2025
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Outdated suggestions cannot be applied.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.
The motivation of this PR is convolution performance for resnet50 configs. With this PR (and a few pending ones), conv performance with igemm pipeline get decent speedup in situation where a standalone dimension size is smaller than intrinsic size. (Take dispatch 69 as example, the select tile m:7, n:512, k:4608 will be rejected from mfma because m tile is smaller than intrinsic size of 16). This happens because previously we are too defensive about when to use intrinsic: in situation when alignment is not required, we still enforce mfma to be picked up only when m/n/k tiles are all larger than intrinsic size.
With @nirvedhmeshram's #19271 and #19484, padding is allowed in tile and fuse matmul and igemm tile and fuse pipelines, it is no longer necessary to be as conservative as before. I am therefore getting rid of the conditional check that blocks mfma from being picked up.
This will impact a few pipelines that use
canTargetIntrinsic():LLVMGPUPadAndVectorDistributewill allow narrow m/n/k dimension sizes for batch matmuliree-codegen-rocdl-configuration-pipeline, will allow narrow m/n/k dimension sizes for matmul (instead of warp reduction)