[ConstantFolding] Generalize constant folding for vector_interleave2 to interleave3-8. #168473

topperc · 2025-11-18T02:12:03Z

No description provided.

llvmbot · 2025-11-18T02:12:58Z

@llvm/pr-subscribers-llvm-transforms

Author: Craig Topper (topperc)

Changes

Full diff: https://github.com/llvm/llvm-project/pull/168473.diff

2 Files Affected:

(modified) llvm/lib/Analysis/ConstantFolding.cpp (+20-7)
(modified) llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll (+48)

diff --git a/llvm/lib/Analysis/ConstantFolding.cpp b/llvm/lib/Analysis/ConstantFolding.cpp
index da32542cf7870..a13df6c5bf552 100755
--- a/llvm/lib/Analysis/ConstantFolding.cpp
+++ b/llvm/lib/Analysis/ConstantFolding.cpp
@@ -1659,6 +1659,12 @@ bool llvm::canConstantFoldCallTo(const CallBase *Call, const Function *F) {
   case Intrinsic::vector_extract:
   case Intrinsic::vector_insert:
   case Intrinsic::vector_interleave2:
+  case Intrinsic::vector_interleave3:
+  case Intrinsic::vector_interleave4:
+  case Intrinsic::vector_interleave5:
+  case Intrinsic::vector_interleave6:
+  case Intrinsic::vector_interleave7:
+  case Intrinsic::vector_interleave8:
   case Intrinsic::vector_deinterleave2:
   // Target intrinsics
   case Intrinsic::amdgcn_perm:
@@ -4207,16 +4213,23 @@ static Constant *ConstantFoldFixedVectorCall(
     }
     return ConstantVector::get(Result);
   }
-  case Intrinsic::vector_interleave2: {
+  case Intrinsic::vector_interleave2:
+  case Intrinsic::vector_interleave3:
+  case Intrinsic::vector_interleave4:
+  case Intrinsic::vector_interleave5:
+  case Intrinsic::vector_interleave6:
+  case Intrinsic::vector_interleave7:
+  case Intrinsic::vector_interleave8: {
     unsigned NumElements =
         cast<FixedVectorType>(Operands[0]->getType())->getNumElements();
+    unsigned NumOperands = Operands.size();
     for (unsigned I = 0; I < NumElements; ++I) {
-      Constant *Elt0 = Operands[0]->getAggregateElement(I);
-      Constant *Elt1 = Operands[1]->getAggregateElement(I);
-      if (!Elt0 || !Elt1)
-        return nullptr;
-      Result[2 * I] = Elt0;
-      Result[2 * I + 1] = Elt1;
+      for (unsigned J = 0; J < NumOperands; ++J) {
+        Constant *Elt = Operands[J]->getAggregateElement(I);
+        if (!Elt)
+          return nullptr;
+        Result[NumOperands * I + J] = Elt;
+      }
     }
     return ConstantVector::get(Result);
   }
diff --git a/llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll b/llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll
index 14543f339db5d..848f0d17ff373 100644
--- a/llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll
+++ b/llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll
@@ -51,6 +51,54 @@ define <8 x i32> @fold_vector_interleave2() {
   ret <8 x i32> %1
 }
 
+define <12 x i32> @fold_vector_interleave3() {
+; CHECK-LABEL: define <12 x i32> @fold_vector_interleave3() {
+; CHECK-NEXT:    ret <12 x i32> <i32 1, i32 5, i32 9, i32 2, i32 6, i32 10, i32 3, i32 7, i32 11, i32 4, i32 8, i32 12>
+;
+  %1 = call <12 x i32> @llvm.vector.interleave3.v12i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>)
+  ret <12 x i32> %1
+}
+
+define <16 x i32> @fold_vector_interleave4() {
+; CHECK-LABEL: define <16 x i32> @fold_vector_interleave4() {
+; CHECK-NEXT:    ret <16 x i32> <i32 1, i32 5, i32 9, i32 13, i32 2, i32 6, i32 10, i32 14, i32 3, i32 7, i32 11, i32 15, i32 4, i32 8, i32 12, i32 16>
+;
+  %1 = call <16 x i32> @llvm.vector.interleave4.v16i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>)
+  ret <16 x i32> %1
+}
+
+define <20 x i32> @fold_vector_interleave5() {
+; CHECK-LABEL: define <20 x i32> @fold_vector_interleave5() {
+; CHECK-NEXT:    ret <20 x i32> <i32 1, i32 5, i32 9, i32 13, i32 17, i32 2, i32 6, i32 10, i32 14, i32 18, i32 3, i32 7, i32 11, i32 15, i32 19, i32 4, i32 8, i32 12, i32 16, i32 20>
+;
+  %1 = call <20 x i32> @llvm.vector.interleave5.v20i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>, <4 x i32> <i32 17, i32 18, i32 19, i32 20>)
+  ret <20 x i32> %1
+}
+
+define <24 x i32> @fold_vector_interleave6() {
+; CHECK-LABEL: define <24 x i32> @fold_vector_interleave6() {
+; CHECK-NEXT:    ret <24 x i32> <i32 1, i32 5, i32 9, i32 13, i32 17, i32 21, i32 2, i32 6, i32 10, i32 14, i32 18, i32 22, i32 3, i32 7, i32 11, i32 15, i32 19, i32 23, i32 4, i32 8, i32 12, i32 16, i32 20, i32 24>
+;
+  %1 = call <24 x i32> @llvm.vector.interleave6.v24i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>, <4 x i32> <i32 17, i32 18, i32 19, i32 20>, <4 x i32> <i32 21, i32 22, i32 23, i32 24>)
+  ret <24 x i32> %1
+}
+
+define <28 x i32> @fold_vector_interleave7() {
+; CHECK-LABEL: define <28 x i32> @fold_vector_interleave7() {
+; CHECK-NEXT:    ret <28 x i32> <i32 1, i32 5, i32 9, i32 13, i32 17, i32 21, i32 25, i32 2, i32 6, i32 10, i32 14, i32 18, i32 22, i32 26, i32 3, i32 7, i32 11, i32 15, i32 19, i32 23, i32 27, i32 4, i32 8, i32 12, i32 16, i32 20, i32 24, i32 28>
+;
+  %1 = call <28 x i32> @llvm.vector.interleave7.v28i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>, <4 x i32> <i32 17, i32 18, i32 19, i32 20>, <4 x i32> <i32 21, i32 22, i32 23, i32 24>, <4 x i32> <i32 25, i32 26, i32 27, i32 28>)
+  ret <28 x i32> %1
+}
+
+define <32 x i32> @fold_vector_interleave8() {
+; CHECK-LABEL: define <32 x i32> @fold_vector_interleave8() {
+; CHECK-NEXT:    ret <32 x i32> <i32 1, i32 5, i32 9, i32 13, i32 17, i32 21, i32 25, i32 29, i32 2, i32 6, i32 10, i32 14, i32 18, i32 22, i32 26, i32 30, i32 3, i32 7, i32 11, i32 15, i32 19, i32 23, i32 27, i32 31, i32 4, i32 8, i32 12, i32 16, i32 20, i32 24, i32 28, i32 32>
+;
+  %1 = call <32 x i32> @llvm.vector.interleave8.v32i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>, <4 x i32> <i32 17, i32 18, i32 19, i32 20>, <4 x i32> <i32 21, i32 22, i32 23, i32 24>, <4 x i32> <i32 25, i32 26, i32 27, i32 28>, <4 x i32> <i32 29, i32 30, i32 31, i32 32>)
+  ret <32 x i32> %1
+}
+
 define {<4 x i32>, <4 x i32>} @fold_vector_deinterleave2() {
 ; CHECK-LABEL: define { <4 x i32>, <4 x i32> } @fold_vector_deinterleave2() {
 ; CHECK-NEXT:    ret { <4 x i32>, <4 x i32> } { <4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8> }

llvmbot · 2025-11-18T02:12:58Z

@llvm/pr-subscribers-llvm-analysis

Author: Craig Topper (topperc)

Changes

Full diff: https://github.com/llvm/llvm-project/pull/168473.diff

2 Files Affected:

(modified) llvm/lib/Analysis/ConstantFolding.cpp (+20-7)
(modified) llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll (+48)

diff --git a/llvm/lib/Analysis/ConstantFolding.cpp b/llvm/lib/Analysis/ConstantFolding.cpp
index da32542cf7870..a13df6c5bf552 100755
--- a/llvm/lib/Analysis/ConstantFolding.cpp
+++ b/llvm/lib/Analysis/ConstantFolding.cpp
@@ -1659,6 +1659,12 @@ bool llvm::canConstantFoldCallTo(const CallBase *Call, const Function *F) {
   case Intrinsic::vector_extract:
   case Intrinsic::vector_insert:
   case Intrinsic::vector_interleave2:
+  case Intrinsic::vector_interleave3:
+  case Intrinsic::vector_interleave4:
+  case Intrinsic::vector_interleave5:
+  case Intrinsic::vector_interleave6:
+  case Intrinsic::vector_interleave7:
+  case Intrinsic::vector_interleave8:
   case Intrinsic::vector_deinterleave2:
   // Target intrinsics
   case Intrinsic::amdgcn_perm:
@@ -4207,16 +4213,23 @@ static Constant *ConstantFoldFixedVectorCall(
     }
     return ConstantVector::get(Result);
   }
-  case Intrinsic::vector_interleave2: {
+  case Intrinsic::vector_interleave2:
+  case Intrinsic::vector_interleave3:
+  case Intrinsic::vector_interleave4:
+  case Intrinsic::vector_interleave5:
+  case Intrinsic::vector_interleave6:
+  case Intrinsic::vector_interleave7:
+  case Intrinsic::vector_interleave8: {
     unsigned NumElements =
         cast<FixedVectorType>(Operands[0]->getType())->getNumElements();
+    unsigned NumOperands = Operands.size();
     for (unsigned I = 0; I < NumElements; ++I) {
-      Constant *Elt0 = Operands[0]->getAggregateElement(I);
-      Constant *Elt1 = Operands[1]->getAggregateElement(I);
-      if (!Elt0 || !Elt1)
-        return nullptr;
-      Result[2 * I] = Elt0;
-      Result[2 * I + 1] = Elt1;
+      for (unsigned J = 0; J < NumOperands; ++J) {
+        Constant *Elt = Operands[J]->getAggregateElement(I);
+        if (!Elt)
+          return nullptr;
+        Result[NumOperands * I + J] = Elt;
+      }
     }
     return ConstantVector::get(Result);
   }
diff --git a/llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll b/llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll
index 14543f339db5d..848f0d17ff373 100644
--- a/llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll
+++ b/llvm/test/Transforms/InstSimplify/ConstProp/vector-calls.ll
@@ -51,6 +51,54 @@ define <8 x i32> @fold_vector_interleave2() {
   ret <8 x i32> %1
 }
 
+define <12 x i32> @fold_vector_interleave3() {
+; CHECK-LABEL: define <12 x i32> @fold_vector_interleave3() {
+; CHECK-NEXT:    ret <12 x i32> <i32 1, i32 5, i32 9, i32 2, i32 6, i32 10, i32 3, i32 7, i32 11, i32 4, i32 8, i32 12>
+;
+  %1 = call <12 x i32> @llvm.vector.interleave3.v12i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>)
+  ret <12 x i32> %1
+}
+
+define <16 x i32> @fold_vector_interleave4() {
+; CHECK-LABEL: define <16 x i32> @fold_vector_interleave4() {
+; CHECK-NEXT:    ret <16 x i32> <i32 1, i32 5, i32 9, i32 13, i32 2, i32 6, i32 10, i32 14, i32 3, i32 7, i32 11, i32 15, i32 4, i32 8, i32 12, i32 16>
+;
+  %1 = call <16 x i32> @llvm.vector.interleave4.v16i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>)
+  ret <16 x i32> %1
+}
+
+define <20 x i32> @fold_vector_interleave5() {
+; CHECK-LABEL: define <20 x i32> @fold_vector_interleave5() {
+; CHECK-NEXT:    ret <20 x i32> <i32 1, i32 5, i32 9, i32 13, i32 17, i32 2, i32 6, i32 10, i32 14, i32 18, i32 3, i32 7, i32 11, i32 15, i32 19, i32 4, i32 8, i32 12, i32 16, i32 20>
+;
+  %1 = call <20 x i32> @llvm.vector.interleave5.v20i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>, <4 x i32> <i32 17, i32 18, i32 19, i32 20>)
+  ret <20 x i32> %1
+}
+
+define <24 x i32> @fold_vector_interleave6() {
+; CHECK-LABEL: define <24 x i32> @fold_vector_interleave6() {
+; CHECK-NEXT:    ret <24 x i32> <i32 1, i32 5, i32 9, i32 13, i32 17, i32 21, i32 2, i32 6, i32 10, i32 14, i32 18, i32 22, i32 3, i32 7, i32 11, i32 15, i32 19, i32 23, i32 4, i32 8, i32 12, i32 16, i32 20, i32 24>
+;
+  %1 = call <24 x i32> @llvm.vector.interleave6.v24i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>, <4 x i32> <i32 17, i32 18, i32 19, i32 20>, <4 x i32> <i32 21, i32 22, i32 23, i32 24>)
+  ret <24 x i32> %1
+}
+
+define <28 x i32> @fold_vector_interleave7() {
+; CHECK-LABEL: define <28 x i32> @fold_vector_interleave7() {
+; CHECK-NEXT:    ret <28 x i32> <i32 1, i32 5, i32 9, i32 13, i32 17, i32 21, i32 25, i32 2, i32 6, i32 10, i32 14, i32 18, i32 22, i32 26, i32 3, i32 7, i32 11, i32 15, i32 19, i32 23, i32 27, i32 4, i32 8, i32 12, i32 16, i32 20, i32 24, i32 28>
+;
+  %1 = call <28 x i32> @llvm.vector.interleave7.v28i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>, <4 x i32> <i32 17, i32 18, i32 19, i32 20>, <4 x i32> <i32 21, i32 22, i32 23, i32 24>, <4 x i32> <i32 25, i32 26, i32 27, i32 28>)
+  ret <28 x i32> %1
+}
+
+define <32 x i32> @fold_vector_interleave8() {
+; CHECK-LABEL: define <32 x i32> @fold_vector_interleave8() {
+; CHECK-NEXT:    ret <32 x i32> <i32 1, i32 5, i32 9, i32 13, i32 17, i32 21, i32 25, i32 29, i32 2, i32 6, i32 10, i32 14, i32 18, i32 22, i32 26, i32 30, i32 3, i32 7, i32 11, i32 15, i32 19, i32 23, i32 27, i32 31, i32 4, i32 8, i32 12, i32 16, i32 20, i32 24, i32 28, i32 32>
+;
+  %1 = call <32 x i32> @llvm.vector.interleave8.v32i32(<4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8>, <4 x i32> <i32 9, i32 10, i32 11, i32 12>, <4 x i32> <i32 13, i32 14, i32 15, i32 16>, <4 x i32> <i32 17, i32 18, i32 19, i32 20>, <4 x i32> <i32 21, i32 22, i32 23, i32 24>, <4 x i32> <i32 25, i32 26, i32 27, i32 28>, <4 x i32> <i32 29, i32 30, i32 31, i32 32>)
+  ret <32 x i32> %1
+}
+
 define {<4 x i32>, <4 x i32>} @fold_vector_deinterleave2() {
 ; CHECK-LABEL: define { <4 x i32>, <4 x i32> } @fold_vector_deinterleave2() {
 ; CHECK-NEXT:    ret { <4 x i32>, <4 x i32> } { <4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32> <i32 5, i32 6, i32 7, i32 8> }

…to interleave3-8.

github-actions · 2025-11-18T03:02:40Z

🐧 Linux x64 Test Results

186267 tests passed
4848 tests skipped

preames

LGTM

mshockwave

LGTM

mshockwave · 2025-11-18T18:47:55Z

llvm/lib/Analysis/ConstantFolding.cpp

+  case Intrinsic::vector_interleave6:
+  case Intrinsic::vector_interleave7:
+  case Intrinsic::vector_interleave8: {
    unsigned NumElements =


I forgot whether we have this discussions before already, but do we have a scalable vector version of this folding (for splat specifically)?

We don't have one right now. I plan to add it.

llvm-ci · 2025-11-18T20:31:38Z

LLVM Buildbot has detected a new failure on builder amdgpu-offload-rhel-9-cmake-build-only running on rocm-docker-rhel-9 while building llvm at step 2 "update-annotated-scripts".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/205/builds/28308

Here is the relevant piece of the build log for the reference

Step 2 (update-annotated-scripts) failure: update (failure)
git version 2.43.5
error: RPC failed; HTTP 500 curl 22 The requested URL returned error: 500
fatal: expected flush after ref listing
error: RPC failed; HTTP 500 curl 22 The requested URL returned error: 500
fatal: expected flush after ref listing

llvm-ci · 2025-11-18T20:41:28Z

LLVM Buildbot has detected a new failure on builder amdgpu-offload-ubuntu-22-cmake-build-only running on rocm-docker-ubu-22 while building llvm at step 2 "update-annotated-scripts".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/203/builds/29517

Here is the relevant piece of the build log for the reference

Step 2 (update-annotated-scripts) failure: update (failure)
git version 2.34.1
remote: Internal Server Error
fatal: unable to access 'https://github.com/llvm/llvm-zorg.git/': The requested URL returned error: 500
remote: Internal Server Error
fatal: unable to access 'https://github.com/llvm/llvm-zorg.git/': The requested URL returned error: 500

topperc requested review from mshockwave, nikic and npanchen November 18, 2025 02:12

llvmbot added llvm:instcombine Covers the InstCombine, InstSimplify and AggressiveInstCombine passes llvm:analysis Includes value tracking, cost tables and constant folding llvm:transforms labels Nov 18, 2025

topperc and others added 2 commits November 17, 2025 18:15

Pre-commit tests

f40e646

[ConstantFolding] Generalize constant folding for vector_interleave2 …

5626571

…to interleave3-8.

topperc force-pushed the pr/interleave-fold branch from b6e000a to 5626571 Compare November 18, 2025 02:15

preames approved these changes Nov 18, 2025

View reviewed changes

mshockwave approved these changes Nov 18, 2025

View reviewed changes

npanchen approved these changes Nov 18, 2025

View reviewed changes

topperc merged commit 4ab2423 into llvm:main Nov 18, 2025
10 checks passed

topperc deleted the pr/interleave-fold branch November 18, 2025 19:13

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[ConstantFolding] Generalize constant folding for vector_interleave2 to interleave3-8. #168473

[ConstantFolding] Generalize constant folding for vector_interleave2 to interleave3-8. #168473

Uh oh!

topperc commented Nov 18, 2025

Uh oh!

llvmbot commented Nov 18, 2025

Uh oh!

llvmbot commented Nov 18, 2025

Uh oh!

github-actions bot commented Nov 18, 2025

Uh oh!

preames left a comment

Uh oh!

mshockwave left a comment

Uh oh!

mshockwave Nov 18, 2025

Uh oh!

topperc Nov 18, 2025

Uh oh!

Uh oh!

llvm-ci commented Nov 18, 2025

Uh oh!

llvm-ci commented Nov 18, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

[ConstantFolding] Generalize constant folding for vector_interleave2 to interleave3-8. #168473

[ConstantFolding] Generalize constant folding for vector_interleave2 to interleave3-8. #168473

Uh oh!

Conversation

topperc commented Nov 18, 2025

Uh oh!

llvmbot commented Nov 18, 2025

Uh oh!

llvmbot commented Nov 18, 2025

Uh oh!

github-actions bot commented Nov 18, 2025

🐧 Linux x64 Test Results

Uh oh!

preames left a comment

Choose a reason for hiding this comment

Uh oh!

mshockwave left a comment

Choose a reason for hiding this comment

Uh oh!

mshockwave Nov 18, 2025

Choose a reason for hiding this comment

Uh oh!

topperc Nov 18, 2025

Choose a reason for hiding this comment

Uh oh!

Uh oh!

llvm-ci commented Nov 18, 2025

Uh oh!

llvm-ci commented Nov 18, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants