[VectorCombine] New folding pattern for extract/binop/shuffle chains #145232
+678
−0
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
@llvm/pr-subscribers-llvm-transforms Author: Rajveer Singh Bharadwaj (Rajveer100) ChangesResolves #144654 This adds a new define i16 @<!-- -->test_reduce_v8i16(<8 x i16> %a0) local_unnamed_addr #<!-- -->0 {
%1 = shufflevector <8 x i16> %a0, <8 x i16> poison, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 poison, i32 poison, i32 poison, i32 poison>
%2 = tail call <8 x i16> @<!-- -->llvm.umin.v8i16(<8 x i16> %a0, <8 x i16> %1)
%3 = shufflevector <8 x i16> %2, <8 x i16> poison, <8 x i32> <i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
%4 = tail call <8 x i16> @<!-- -->llvm.umin.v8i16(<8 x i16> %2, <8 x i16> %3)
%5 = shufflevector <8 x i16> %4, <8 x i16> poison, <8 x i32> <i32 1, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
%6 = tail call <8 x i16> @<!-- -->llvm.umin.v8i16(<8 x i16> %4, <8 x i16> %5)
%7 = extractelement <8 x i16> %6, i64 0
ret i16 %7
}...which can be reduced to a llvm.vector.reduce.umin.v8i16(%a0) intrinsic call. Similar transformation for other ops when costs permit to do so. Full diff: https://github.com/llvm/llvm-project/pull/145232.diff 2 Files Affected:
diff --git a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
index 52cb1dbb33b86..aca939c4f534d 100644
--- a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
+++ b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
@@ -129,6 +129,7 @@ class VectorCombine {
bool foldShuffleOfIntrinsics(Instruction &I);
bool foldShuffleToIdentity(Instruction &I);
bool foldShuffleFromReductions(Instruction &I);
+ bool foldShuffleChainsToReduce(Instruction &I);
bool foldCastFromReductions(Instruction &I);
bool foldSelectShuffle(Instruction &I, bool FromReduction = false);
bool foldInterleaveIntrinsics(Instruction &I);
@@ -2910,6 +2911,130 @@ bool VectorCombine::foldShuffleFromReductions(Instruction &I) {
return foldSelectShuffle(*Shuffle, true);
}
+bool VectorCombine::foldShuffleChainsToReduce(Instruction &I) {
+ auto *SVI = dyn_cast<ShuffleVectorInst>(&I);
+ if (!SVI)
+ return false;
+
+ std::queue<Value *> Worklist;
+ SmallVector<Instruction *> ToEraseFromParent;
+
+ SmallVector<int> ShuffleMask;
+ bool IsShuffleOp = true;
+
+ Worklist.push(SVI);
+ SVI->getShuffleMask(ShuffleMask);
+
+ if (ShuffleMask.size() < 2)
+ return false;
+
+ Instruction *Prev0 = nullptr, *Prev1 = nullptr;
+ Instruction *LastOp = nullptr;
+
+ int MaskHalfPos = ShuffleMask.size() / 2;
+ bool IsFirst = true;
+
+ while (!Worklist.empty()) {
+ Value *V = Worklist.front();
+ Worklist.pop();
+
+ auto *CI = dyn_cast<Instruction>(V);
+ if (!CI)
+ return false;
+
+ if (auto *SV = dyn_cast<ShuffleVectorInst>(V)) {
+ if (!IsShuffleOp || MaskHalfPos < 1 || (!Prev1 && !IsFirst))
+ return false;
+
+ auto *Op0 = SV->getOperand(0);
+ auto *Op1 = SV->getOperand(1);
+ if (!Op0 || !Op1)
+ return false;
+
+ auto *FVT = dyn_cast<FixedVectorType>(Op1->getType());
+ if (!FVT || !isa<PoisonValue>(Op1))
+ return false;
+
+ SmallVector<int> CurrentMask;
+ SV->getShuffleMask(CurrentMask);
+
+ int64_t MaskSize = CurrentMask.size();
+ for (int MaskPos = 0; MaskPos != MaskSize; ++MaskPos) {
+ if (MaskPos < MaskHalfPos && CurrentMask[MaskPos] != MaskHalfPos + MaskPos)
+ return false;
+ if (MaskPos >= MaskHalfPos && CurrentMask[MaskPos] != -1)
+ return false;
+ }
+ MaskHalfPos /= 2;
+ Prev0 = SV;
+ } else if (auto *Call = dyn_cast<CallInst>(V)) {
+ if (IsShuffleOp || !Prev0)
+ return false;
+
+ auto *II = dyn_cast<IntrinsicInst>(Call);
+ if (!II)
+ return false;
+
+ switch (II->getIntrinsicID()) {
+ case Intrinsic::umin: {
+ auto *Op0 = Call->getOperand(0);
+ auto *Op1 = Call->getOperand(1);
+ if (!(Op0 == Prev0 && Op1 == Prev1) && !(Op0 == Prev1 && Op1 == Prev0) && !IsFirst)
+ return false;
+
+ if (!IsFirst)
+ Prev0 = Prev1;
+ else
+ IsFirst = false;
+ Prev1 = Call;
+ break;
+ }
+ default:
+ return false;
+ }
+ } else if (auto *ExtractElement = dyn_cast<ExtractElementInst>(CI)) {
+ if (!IsShuffleOp || !Prev0 || !Prev1 || MaskHalfPos != 0)
+ return false;
+
+ auto *Op0 = ExtractElement->getOperand(0);
+ auto *Op1 = ExtractElement->getOperand(1);
+ if (Op0 != Prev1)
+ return false;
+
+ if (auto *Op1Idx = dyn_cast<ConstantInt>(Op1)) {
+ if (Op1Idx->getValue() != 0)
+ return false;
+ } else {
+ return false;
+ }
+ LastOp = ExtractElement;
+ break;
+ }
+ IsShuffleOp ^= 1;
+ ToEraseFromParent.push_back(CI);
+
+ auto *NextI = CI->getNextNode();
+ if (!NextI)
+ return false;
+ Worklist.push(NextI);
+ }
+
+ if (!LastOp)
+ return false;
+
+ auto *ReducedResult = Builder.CreateIntrinsic(Intrinsic::vector_reduce_umin, {SVI->getType()}, {SVI->getOperand(0)});
+ replaceValue(*LastOp, *ReducedResult);
+
+ ToEraseFromParent.push_back(LastOp);
+
+ std::reverse(ToEraseFromParent.begin(), ToEraseFromParent.end());
+ // for (auto &Instr : ToEraseFromParent)
+ // eraseInstruction(*Instr);
+ // Instr->eraseFromParent();
+
+ return true;
+}
+
/// Determine if its more efficient to fold:
/// reduce(trunc(x)) -> trunc(reduce(x)).
/// reduce(sext(x)) -> sext(reduce(x)).
@@ -3607,6 +3732,7 @@ bool VectorCombine::run() {
MadeChange |= foldShuffleOfIntrinsics(I);
MadeChange |= foldSelectShuffle(I);
MadeChange |= foldShuffleToIdentity(I);
+ MadeChange |= foldShuffleChainsToReduce(I);
break;
case Instruction::BitCast:
MadeChange |= foldBitcastShuffle(I);
diff --git a/llvm/test/Transforms/VectorCombine/fold-shuffle-chains-to-reduce.ll b/llvm/test/Transforms/VectorCombine/fold-shuffle-chains-to-reduce.ll
new file mode 100644
index 0000000000000..6f21eb5097fde
--- /dev/null
+++ b/llvm/test/Transforms/VectorCombine/fold-shuffle-chains-to-reduce.ll
@@ -0,0 +1,18 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt < %s -passes=vector-combine -S | FileCheck %s
+
+define i16 @test_reduce_v8i16(<8 x i16> %a0) local_unnamed_addr #0 {
+; CHECK-LABEL: define i16 @test_reduce_v8i16(
+; CHECK-SAME: <8 x i16> [[A0:%.*]]) local_unnamed_addr {
+; CHECK-NEXT: [[TMP1:%.*]] = call i16 @llvm.vector.reduce.umin.v8i16(<8 x i16> [[A0]])
+; CHECK-NEXT: ret i16 [[TMP1]]
+;
+ %1 = shufflevector <8 x i16> %a0, <8 x i16> poison, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 poison, i32 poison, i32 poison, i32 poison>
+ %2 = tail call <8 x i16> @llvm.umin.v8i16(<8 x i16> %a0, <8 x i16> %1)
+ %3 = shufflevector <8 x i16> %2, <8 x i16> poison, <8 x i32> <i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
+ %4 = tail call <8 x i16> @llvm.umin.v8i16(<8 x i16> %2, <8 x i16> %3)
+ %5 = shufflevector <8 x i16> %4, <8 x i16> poison, <8 x i32> <i32 1, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
+ %6 = tail call <8 x i16> @llvm.umin.v8i16(<8 x i16> %4, <8 x i16> %5)
+ %7 = extractelement <8 x i16> %6, i64 0
+ ret i16 %7
+}
|
|
@llvm/pr-subscribers-vectorizers Author: Rajveer Singh Bharadwaj (Rajveer100) ChangesResolves #144654 This adds a new define i16 @<!-- -->test_reduce_v8i16(<8 x i16> %a0) local_unnamed_addr #<!-- -->0 {
%1 = shufflevector <8 x i16> %a0, <8 x i16> poison, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 poison, i32 poison, i32 poison, i32 poison>
%2 = tail call <8 x i16> @<!-- -->llvm.umin.v8i16(<8 x i16> %a0, <8 x i16> %1)
%3 = shufflevector <8 x i16> %2, <8 x i16> poison, <8 x i32> <i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
%4 = tail call <8 x i16> @<!-- -->llvm.umin.v8i16(<8 x i16> %2, <8 x i16> %3)
%5 = shufflevector <8 x i16> %4, <8 x i16> poison, <8 x i32> <i32 1, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
%6 = tail call <8 x i16> @<!-- -->llvm.umin.v8i16(<8 x i16> %4, <8 x i16> %5)
%7 = extractelement <8 x i16> %6, i64 0
ret i16 %7
}...which can be reduced to a llvm.vector.reduce.umin.v8i16(%a0) intrinsic call. Similar transformation for other ops when costs permit to do so. Full diff: https://github.com/llvm/llvm-project/pull/145232.diff 2 Files Affected:
diff --git a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
index 52cb1dbb33b86..aca939c4f534d 100644
--- a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
+++ b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
@@ -129,6 +129,7 @@ class VectorCombine {
bool foldShuffleOfIntrinsics(Instruction &I);
bool foldShuffleToIdentity(Instruction &I);
bool foldShuffleFromReductions(Instruction &I);
+ bool foldShuffleChainsToReduce(Instruction &I);
bool foldCastFromReductions(Instruction &I);
bool foldSelectShuffle(Instruction &I, bool FromReduction = false);
bool foldInterleaveIntrinsics(Instruction &I);
@@ -2910,6 +2911,130 @@ bool VectorCombine::foldShuffleFromReductions(Instruction &I) {
return foldSelectShuffle(*Shuffle, true);
}
+bool VectorCombine::foldShuffleChainsToReduce(Instruction &I) {
+ auto *SVI = dyn_cast<ShuffleVectorInst>(&I);
+ if (!SVI)
+ return false;
+
+ std::queue<Value *> Worklist;
+ SmallVector<Instruction *> ToEraseFromParent;
+
+ SmallVector<int> ShuffleMask;
+ bool IsShuffleOp = true;
+
+ Worklist.push(SVI);
+ SVI->getShuffleMask(ShuffleMask);
+
+ if (ShuffleMask.size() < 2)
+ return false;
+
+ Instruction *Prev0 = nullptr, *Prev1 = nullptr;
+ Instruction *LastOp = nullptr;
+
+ int MaskHalfPos = ShuffleMask.size() / 2;
+ bool IsFirst = true;
+
+ while (!Worklist.empty()) {
+ Value *V = Worklist.front();
+ Worklist.pop();
+
+ auto *CI = dyn_cast<Instruction>(V);
+ if (!CI)
+ return false;
+
+ if (auto *SV = dyn_cast<ShuffleVectorInst>(V)) {
+ if (!IsShuffleOp || MaskHalfPos < 1 || (!Prev1 && !IsFirst))
+ return false;
+
+ auto *Op0 = SV->getOperand(0);
+ auto *Op1 = SV->getOperand(1);
+ if (!Op0 || !Op1)
+ return false;
+
+ auto *FVT = dyn_cast<FixedVectorType>(Op1->getType());
+ if (!FVT || !isa<PoisonValue>(Op1))
+ return false;
+
+ SmallVector<int> CurrentMask;
+ SV->getShuffleMask(CurrentMask);
+
+ int64_t MaskSize = CurrentMask.size();
+ for (int MaskPos = 0; MaskPos != MaskSize; ++MaskPos) {
+ if (MaskPos < MaskHalfPos && CurrentMask[MaskPos] != MaskHalfPos + MaskPos)
+ return false;
+ if (MaskPos >= MaskHalfPos && CurrentMask[MaskPos] != -1)
+ return false;
+ }
+ MaskHalfPos /= 2;
+ Prev0 = SV;
+ } else if (auto *Call = dyn_cast<CallInst>(V)) {
+ if (IsShuffleOp || !Prev0)
+ return false;
+
+ auto *II = dyn_cast<IntrinsicInst>(Call);
+ if (!II)
+ return false;
+
+ switch (II->getIntrinsicID()) {
+ case Intrinsic::umin: {
+ auto *Op0 = Call->getOperand(0);
+ auto *Op1 = Call->getOperand(1);
+ if (!(Op0 == Prev0 && Op1 == Prev1) && !(Op0 == Prev1 && Op1 == Prev0) && !IsFirst)
+ return false;
+
+ if (!IsFirst)
+ Prev0 = Prev1;
+ else
+ IsFirst = false;
+ Prev1 = Call;
+ break;
+ }
+ default:
+ return false;
+ }
+ } else if (auto *ExtractElement = dyn_cast<ExtractElementInst>(CI)) {
+ if (!IsShuffleOp || !Prev0 || !Prev1 || MaskHalfPos != 0)
+ return false;
+
+ auto *Op0 = ExtractElement->getOperand(0);
+ auto *Op1 = ExtractElement->getOperand(1);
+ if (Op0 != Prev1)
+ return false;
+
+ if (auto *Op1Idx = dyn_cast<ConstantInt>(Op1)) {
+ if (Op1Idx->getValue() != 0)
+ return false;
+ } else {
+ return false;
+ }
+ LastOp = ExtractElement;
+ break;
+ }
+ IsShuffleOp ^= 1;
+ ToEraseFromParent.push_back(CI);
+
+ auto *NextI = CI->getNextNode();
+ if (!NextI)
+ return false;
+ Worklist.push(NextI);
+ }
+
+ if (!LastOp)
+ return false;
+
+ auto *ReducedResult = Builder.CreateIntrinsic(Intrinsic::vector_reduce_umin, {SVI->getType()}, {SVI->getOperand(0)});
+ replaceValue(*LastOp, *ReducedResult);
+
+ ToEraseFromParent.push_back(LastOp);
+
+ std::reverse(ToEraseFromParent.begin(), ToEraseFromParent.end());
+ // for (auto &Instr : ToEraseFromParent)
+ // eraseInstruction(*Instr);
+ // Instr->eraseFromParent();
+
+ return true;
+}
+
/// Determine if its more efficient to fold:
/// reduce(trunc(x)) -> trunc(reduce(x)).
/// reduce(sext(x)) -> sext(reduce(x)).
@@ -3607,6 +3732,7 @@ bool VectorCombine::run() {
MadeChange |= foldShuffleOfIntrinsics(I);
MadeChange |= foldSelectShuffle(I);
MadeChange |= foldShuffleToIdentity(I);
+ MadeChange |= foldShuffleChainsToReduce(I);
break;
case Instruction::BitCast:
MadeChange |= foldBitcastShuffle(I);
diff --git a/llvm/test/Transforms/VectorCombine/fold-shuffle-chains-to-reduce.ll b/llvm/test/Transforms/VectorCombine/fold-shuffle-chains-to-reduce.ll
new file mode 100644
index 0000000000000..6f21eb5097fde
--- /dev/null
+++ b/llvm/test/Transforms/VectorCombine/fold-shuffle-chains-to-reduce.ll
@@ -0,0 +1,18 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
+; RUN: opt < %s -passes=vector-combine -S | FileCheck %s
+
+define i16 @test_reduce_v8i16(<8 x i16> %a0) local_unnamed_addr #0 {
+; CHECK-LABEL: define i16 @test_reduce_v8i16(
+; CHECK-SAME: <8 x i16> [[A0:%.*]]) local_unnamed_addr {
+; CHECK-NEXT: [[TMP1:%.*]] = call i16 @llvm.vector.reduce.umin.v8i16(<8 x i16> [[A0]])
+; CHECK-NEXT: ret i16 [[TMP1]]
+;
+ %1 = shufflevector <8 x i16> %a0, <8 x i16> poison, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 poison, i32 poison, i32 poison, i32 poison>
+ %2 = tail call <8 x i16> @llvm.umin.v8i16(<8 x i16> %a0, <8 x i16> %1)
+ %3 = shufflevector <8 x i16> %2, <8 x i16> poison, <8 x i32> <i32 2, i32 3, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
+ %4 = tail call <8 x i16> @llvm.umin.v8i16(<8 x i16> %2, <8 x i16> %3)
+ %5 = shufflevector <8 x i16> %4, <8 x i16> poison, <8 x i32> <i32 1, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison, i32 poison>
+ %6 = tail call <8 x i16> @llvm.umin.v8i16(<8 x i16> %4, <8 x i16> %5)
+ %7 = extractelement <8 x i16> %6, i64 0
+ ret i16 %7
+}
|
Rajveer100
force-pushed
the
instcombine-opt
branch
from
20aa7b1 to
fea1941
Compare
RKSimon
reviewed
Jun 22, 2025
RKSimon
reviewed
Jun 23, 2025
Rajveer100
force-pushed
the
instcombine-opt
branch
from
fea1941 to
b184ba5
Compare
|
I have updated the implementation itself, re-checking for other potential issues and adding cost analysis/tests. |
Rajveer100
force-pushed
the
instcombine-opt
branch
2 times, most recently
from
d130a70 to
1da772b
Compare
|
@RKSimon |
|
@Rajveer100 Please can you investigate the CI failures? |
Rajveer100
commented
Jun 26, 2025
RKSimon
reviewed
Jun 26, 2025
llvm/test/Transforms/VectorCombine/X86/shuffle-chain-reduction-umin.ll
Outdated
Show resolved
Hide resolved
RKSimon
reviewed
Jun 26, 2025
Rajveer100
force-pushed
the
instcombine-opt
branch
from
1da772b to
dea9e0b
Compare
RKSimon
reviewed
Jun 27, 2025
RKSimon
reviewed
Jun 27, 2025
Rajveer100
force-pushed
the
instcombine-opt
branch
2 times, most recently
from
0116da2 to
44a3268
Compare
Rajveer100
force-pushed
the
instcombine-opt
branch
from
b24a5b8 to
61d835b
Compare
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
RKSimon
reviewed
Aug 18, 2025
Rajveer100
force-pushed
the
instcombine-opt
branch
from
4c3357c to
6cde39b
Compare
Rajveer100
force-pushed
the
instcombine-opt
branch
from
6cde39b to
b7e08e7
Compare
RKSimon
reviewed
Aug 18, 2025
Rajveer100
force-pushed
the
instcombine-opt
branch
from
b7e08e7 to
2a352b1
Compare
RKSimon
reviewed
Aug 18, 2025
RKSimon
reviewed
Aug 18, 2025
Rajveer100
force-pushed
the
instcombine-opt
branch
from
2a352b1 to
a6acccd
Compare
|
@RKSimon |
|
@dtcxzyw |
dtcxzyw
approved these changes
Aug 24, 2025
|
CI failures are unrelated to the changes. It looks like it's due to |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/186/builds/11842 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/51/builds/22175 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/66/builds/18338 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/72/builds/14435 Here is the relevant piece of the build log for the reference |
dtcxzyw
reviewed
Aug 24, 2025
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/168/builds/15535 Here is the relevant piece of the build log for the reference |
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/145/builds/9280 Here is the relevant piece of the build log for the reference |
Rajveer100
added a commit
that referenced
this pull request
Aug 24, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Resolves #144654
Part of #143088
This adds a new
foldShuffleChainsToReducefor horizontal reduction of patterns like:...which can be reduced to a llvm.vector.reduce.umin.v8i16(%a0) intrinsic call.
Similar transformation for other ops when costs permit to do so.