Skip to content

[mlir][vector][NFC] isDisjointTransferIndices: Use getConstantIntValue #65931

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
merged 1 commit into from
Sep 11, 2023
Merged

[mlir][vector][NFC] isDisjointTransferIndices: Use getConstantIntValue #65931

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
13 changes: 5 additions & 8 deletions mlir/lib/Dialect/Vector/IR/VectorOps.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -172,24 +172,21 @@ bool mlir::vector::isDisjointTransferIndices(
return false;
unsigned rankOffset = transferA.getLeadingShapedRank();
for (unsigned i = 0, e = transferA.indices().size(); i < e; i++) {
auto indexA = transferA.indices()[i].getDefiningOp<arith::ConstantOp>();
auto indexB = transferB.indices()[i].getDefiningOp<arith::ConstantOp>();
auto indexA = getConstantIntValue(transferA.indices()[i]);
auto indexB = getConstantIntValue(transferB.indices()[i]);
// If any of the indices are dynamic we cannot prove anything.
if (!indexA || !indexB)
if (!indexA.has_value() || !indexB.has_value())
continue;

if (i < rankOffset) {
// For leading dimensions, if we can prove that index are different we
// know we are accessing disjoint slices.
if (llvm::cast<IntegerAttr>(indexA.getValue()).getInt() !=
llvm::cast<IntegerAttr>(indexB.getValue()).getInt())
if (*indexA != *indexB)
return true;
} else {
// For this dimension, we slice a part of the memref we need to make sure
// the intervals accessed don't overlap.
int64_t distance =
std::abs(llvm::cast<IntegerAttr>(indexA.getValue()).getInt() -
llvm::cast<IntegerAttr>(indexB.getValue()).getInt());
int64_t distance = std::abs(*indexA - *indexB);
if (distance >= transferA.getVectorType().getDimSize(i - rankOffset))
return true;
}
Expand Down