github/ROCm
1
1
Fork 0
mirror of https://github.com/ROCm/ROCm.git synced 2026-04-05 03:01:17 -04:00
Code Issues Packages Projects Releases Wiki Activity

revert PR#245 and #241 for gemm performance regression (#265)

Browse Source 
* Revert "[MFMA][Dot] Swizzle mfma dot operands  (#245)"

This reverts commit 4d0deef45f.

* revert the swizzle and slice PRs for gemm performance regression

* fix build error after reverting the PR#245(commit 4d0deef45f) Swizzle mamf dot operands
This commit is contained in:
Shucai Xiao
2023-07-24 14:09:46 -05:00
committed by GitHub
parent 5bdf71313c
commit 356595f3dc
3 changed files with 115 additions and 215 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
python/triton/compiler/compiler.py
View File

@@ -85,7 +85,9 @@ def optimize_ttgir(mod, num_stages, arch):
pm.add_tritongpu_accelerate_matmul_pass(80)
pm.add_tritongpu_remove_layout_conversions_pass()
pm.add_tritongpu_optimize_dot_operands_pass()
pm.add_tritongpu_pipeline_pass(num_stages)
# TODO enable this pass for AMD GPU when it is ready
if not is_hip():
pm.add_tritongpu_pipeline_pass(num_stages)
pm.add_tritongpu_prefetch_pass()
pm.add_tritongpu_optimize_dot_operands_pass()
pm.add_tritongpu_remove_layout_conversions_pass()