[llvm] [LAA] Support backward dependences with non-constant distance. (PR #91525)
via llvm-commits
llvm-commits at lists.llvm.org
Thu May 9 04:26:36 PDT 2024
github-actions[bot] wrote:
<!--LLVM CODE FORMAT COMMENT: {clang-format}-->
:warning: C/C++ code formatter, clang-format found issues in your code. :warning:
<details>
<summary>
You can test this locally with the following command:
</summary>
``````````bash
git-clang-format --diff 1494d8849fa2aef575dabd431b0060639f4a57c1 464ec4786467b01cbaca6427d1948e7e97bab182 -- llvm/include/llvm/Analysis/LoopAccessAnalysis.h llvm/lib/Analysis/LoopAccessAnalysis.cpp llvm/lib/Transforms/Scalar/LoopFlatten.cpp llvm/lib/Transforms/Scalar/LoopVersioningLICM.cpp llvm/unittests/Transforms/Vectorize/VPlanSlpTest.cpp
``````````
</details>
<details>
<summary>
View the diff from clang-format here.
</summary>
``````````diff
diff --git a/llvm/include/llvm/Analysis/LoopAccessAnalysis.h b/llvm/include/llvm/Analysis/LoopAccessAnalysis.h
index ccc2694ed8..6ebd0fb847 100644
--- a/llvm/include/llvm/Analysis/LoopAccessAnalysis.h
+++ b/llvm/include/llvm/Analysis/LoopAccessAnalysis.h
@@ -316,7 +316,10 @@ private:
/// RecordDependences is true.
SmallVector<Dependence, 8> Dependences;
- /// The maximum width of a target's vector registers multiplied by 2 to also roughly account for additional interleaving. Is used to decide if a backwards dependence with non-constant stride should be classified as backwards-vectorizable or unknown (triggering a runtime check).
+ /// The maximum width of a target's vector registers multiplied by 2 to also
+ /// roughly account for additional interleaving. Is used to decide if a
+ /// backwards dependence with non-constant stride should be classified as
+ /// backwards-vectorizable or unknown (triggering a runtime check).
unsigned MaxTargetVectorWidthInBits = 0;
/// Check whether there is a plausible dependence between the two
diff --git a/llvm/lib/Analysis/LoopAccessAnalysis.cpp b/llvm/lib/Analysis/LoopAccessAnalysis.cpp
index be8deae251..35061508c3 100644
--- a/llvm/lib/Analysis/LoopAccessAnalysis.cpp
+++ b/llvm/lib/Analysis/LoopAccessAnalysis.cpp
@@ -3048,7 +3048,8 @@ LoopAccessInfo::LoopAccessInfo(Loop *L, ScalarEvolution *SE,
TypeSize FixedWidth =
TTI->getRegisterBitWidth(TargetTransformInfo::RGK_FixedWidthVector);
if (FixedWidth.isNonZero()) {
- // Scale the vector width by 2 as rough estimate to also consider interleaving.
+ // Scale the vector width by 2 as rough estimate to also consider
+ // interleaving.
MaxTargetVectorWidthInBits = FixedWidth.getFixedValue() * 2;
}
``````````
</details>
https://github.com/llvm/llvm-project/pull/91525
More information about the llvm-commits
mailing list