Skip to content

[LV] Fix MVE regression from #132190 #141736

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

Open
wants to merge 4 commits into
base: main
Choose a base branch
from
+148 −1
Open

[LV] Fix MVE regression from #132190 #141736

Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
f79a467
[LV] Fix MVE regression from #132190
SamTebbs33 May 28, 2025
a4a0728
Add test
SamTebbs33 May 29, 2025
d6837fe
Only compute reg usage if needed
SamTebbs33 May 29, 2025
760eb57
Interpret VPWidenPointerInductionRecipe as a scalar instead
SamTebbs33 May 30, 2025
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
27 changes: 21 additions & 6 deletions llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -959,6 +959,10 @@ class LoopVectorizationCostModel {
return expectedCost(UserVF).isValid();
}

/// \return True if maximizing vector bandwidth is enabled by the target or
/// user options.
bool useMaxBandwidth(TargetTransformInfo::RegisterKind RegKind);

/// \return The size (in bits) of the smallest and widest types in the code
/// that needs to be vectorized. We ignore values that remain scalar such as
/// 64 bit loop indices.
Expand Down Expand Up @@ -3944,6 +3948,14 @@ LoopVectorizationCostModel::computeMaxVF(ElementCount UserVF, unsigned UserIC) {
return FixedScalableVFPair::getNone();
}

bool LoopVectorizationCostModel::useMaxBandwidth(
TargetTransformInfo::RegisterKind RegKind) {
return MaximizeBandwidth || (MaximizeBandwidth.getNumOccurrences() == 0 &&
(TTI.shouldMaximizeVectorBandwidth(RegKind) ||
(UseWiderVFIfCallVariantsPresent &&
Legal->hasVectorCallVariants())));
Copy link
Contributor

@lukel97 lukel97 May 28, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@preames @wangpc-pp @topperc Making a note here that after this patch we'll likely regress cases like this on RVV unless we opt into shouldMaximizeVectorBandwidth https://godbolt.org/z/WcbWGooa4

}

ElementCount LoopVectorizationCostModel::getMaximizedVFForTarget(
unsigned MaxTripCount, unsigned SmallestType, unsigned WidestType,
ElementCount MaxSafeVF, bool FoldTailByMasking) {
Expand Down Expand Up @@ -4009,10 +4021,7 @@ ElementCount LoopVectorizationCostModel::getMaximizedVFForTarget(
ComputeScalableMaxVF ? TargetTransformInfo::RGK_ScalableVector
: TargetTransformInfo::RGK_FixedWidthVector;
ElementCount MaxVF = MaxVectorElementCount;
if (MaximizeBandwidth ||
(MaximizeBandwidth.getNumOccurrences() == 0 &&
(TTI.shouldMaximizeVectorBandwidth(RegKind) ||
(UseWiderVFIfCallVariantsPresent && Legal->hasVectorCallVariants())))) {
if (useMaxBandwidth(RegKind)) {
auto MaxVectorElementCountMaxBW = ElementCount::get(
llvm::bit_floor(WidestRegister.getKnownMinValue() / SmallestType),
ComputeScalableMaxVF);
Expand Down Expand Up @@ -4384,7 +4393,10 @@ VectorizationFactor LoopVectorizationPlanner::selectVectorizationFactor() {

/// Don't consider the VF if it exceeds the number of registers for the
/// target.
if (RU.exceedsMaxNumRegs(TTI))
if (CM.useMaxBandwidth(VF.isScalable()
? TargetTransformInfo::RGK_ScalableVector
: TargetTransformInfo::RGK_FixedWidthVector) &&
RU.exceedsMaxNumRegs(TTI))
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think this is the only use of RU, would be good to skip the register pressure computation when it's not needed

Same below

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good idea, done! This has also gotten the riscv tests back to what they were before merging #132190.

continue;

InstructionCost C = CM.expectedCost(VF);
Expand Down Expand Up @@ -7458,7 +7470,10 @@ VectorizationFactor LoopVectorizationPlanner::computeBestVF() {
InstructionCost Cost = cost(*P, VF);
VectorizationFactor CurrentFactor(VF, Cost, ScalarCost);

if (RU.exceedsMaxNumRegs(TTI)) {
if (CM.useMaxBandwidth(VF.isScalable()
? TargetTransformInfo::RGK_ScalableVector
: TargetTransformInfo::RGK_FixedWidthVector) &&
RU.exceedsMaxNumRegs(TTI)) {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you explain this logic briefly? Maybe I'm misunderstanding the goal here, but if we exceed the number of registers but don't have maxBandwidth enabled, then we'd try to vectorize anyway? Does the call to CM.useMaxBandwidth want to be negated?

In other words; Why are we only considering whether we exceed the number of registers when maxBandwidth is enabled?

Copy link
Collaborator Author

@SamTebbs33 SamTebbs33 May 29, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That understanding is correct. This was the behaviour before my previous PR accidentally changed it. I'll be looking into another sensible solution for this specific reproducer but this fixes the regression.

LLVM_DEBUG(dbgs() << "LV(REG): Not considering vector loop of width "
<< VF << " because it uses too many registers\n");
continue;
Expand Down
Loading