Skip to content

[LLVM Pulldown] Bump to rev 19b28074618c92fa4c4281eeee67c715abebbfd7 #1140

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Garra1980 merged 1 commit into from
Dec 19, 2025
Merged

[LLVM Pulldown] Bump to rev 19b28074618c92fa4c4281eeee67c715abebbfd7 #1140

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion build_tools/llvm_version.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1 +1 @@
aba8ebbda0912ef2037668aaa48cfbe59991576f
19b28074618c92fa4c4281eeee67c715abebbfd7
96 changes: 0 additions & 96 deletions build_tools/patches/extract_strided_slice_distribution_fix.patch
View file
Open in desktop

This file was deleted.

23 changes: 0 additions & 23 deletions build_tools/patches/wg_fa_support.patch
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,29 +107,6 @@ diff --git a/mlir/lib/Dialect/XeGPU/Transforms/XeGPUWgToSgDistribute.cpp b/mlir/
index 48bd0662b03f..76ed73fdfaef 100644
--- a/mlir/lib/Dialect/XeGPU/Transforms/XeGPUWgToSgDistribute.cpp
+++ b/mlir/lib/Dialect/XeGPU/Transforms/XeGPUWgToSgDistribute.cpp
@@ -1129,14 +1129,14 @@ struct WgToSgVectorShapeCastOp

// For rank reducing or increasing shape_cast ops, the lower rank layout
// must be a slice of higher rank layout.
- int64_t sourceRank = srcType.getRank();
- int64_t resultRank = sgShape.size();
- xegpu::DistributeLayoutAttr sourceLayout =
- xegpu::getDistributeLayoutAttr(op.getSource());
- if (sourceRank < resultRank && !sourceLayout.isSliceOf(layout))
- return failure();
- if (sourceRank > resultRank && !layout.isSliceOf(sourceLayout))
- return failure();
+ // int64_t sourceRank = srcType.getRank();
+ // int64_t resultRank = sgShape.size();
+ // xegpu::DistributeLayoutAttr sourceLayout =
+ // xegpu::getDistributeLayoutAttr(op.getSource());
+ // if (sourceRank < resultRank && !sourceLayout.isSliceOf(layout))
+ // return failure();
+ // if (sourceRank > resultRank && !layout.isSliceOf(sourceLayout))
+ // return failure();

SmallVector<Value> newShapeCastOps;
for (auto src : adaptor.getSource()) {
@@ -1238,13 +1238,13 @@ struct WgToSgVectorTransposeOp
SmallVector<int64_t> sourceSgLayout =
sourceLayout.getEffectiveSgLayoutAsInt();
Expand Down
1 change: 1 addition & 0 deletions test/Integration/Dialect/XeGPU/WG/flash_attention_fwd.mlir
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,7 @@
// RUN: --runner mlir-runner -e main \
// RUN: --entry-point-result=void \
// RUN: --shared-libs=%irunner_utils,%mlir_runner_utils,%mlir_c_runner_utils,%mlir_levelzero_runtime --filecheck

#q = #xegpu.layout<sg_layout = [8, 1], sg_data = [16, 64], inst_data = [8, 16]>
#k = #xegpu.layout<sg_layout = [8, 1], sg_data = [16, 64], inst_data = [16, 16]>
#v = #k
Expand Down
Loading