[VectorCombine] foldShuffleOfBinops - prefer same cost fold if it reduces instruction count #120216
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
…uction count We don't fold "shuffle (binop), (binop)" -> "binop (shuffle), (shuffle)" if the old/new costs are equal, but we can relax this if either new shuffle will constant fold as it will reduce instruction count.
|
@llvm/pr-subscribers-vectorizers @llvm/pr-subscribers-llvm-transforms Author: Simon Pilgrim (RKSimon) ChangesWe don't fold "shuffle (binop), (binop)" -> "binop (shuffle), (shuffle)" if the old/new costs are equal, but we can relax this if either new shuffle will constant fold as it will reduce instruction count. Full diff: https://github.com/llvm/llvm-project/pull/120216.diff 2 Files Affected:
diff --git a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
index db77d6c955792c..791006c48b5ddf 100644
--- a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
+++ b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
@@ -1713,7 +1713,12 @@ bool VectorCombine::foldShuffleOfBinops(Instruction &I) {
LLVM_DEBUG(dbgs() << "Found a shuffle feeding two binops: " << I
<< "\n OldCost: " << OldCost << " vs NewCost: " << NewCost
<< "\n");
- if (NewCost >= OldCost)
+
+ // If either shuffle will constant fold away, then fold for the same cost as
+ // we will reduce the instruction count.
+ bool ReducedInstCount = (isa<Constant>(X) && isa<Constant>(Z)) ||
+ (isa<Constant>(Y) && isa<Constant>(W));
+ if (ReducedInstCount ? (NewCost > OldCost) : (NewCost >= OldCost))
return false;
Value *Shuf0 = Builder.CreateShuffleVector(X, Z, NewMask0);
diff --git a/llvm/test/Transforms/VectorCombine/X86/concat-boolmasks.ll b/llvm/test/Transforms/VectorCombine/X86/concat-boolmasks.ll
index b158d4a3676b37..057d9af314ba3b 100644
--- a/llvm/test/Transforms/VectorCombine/X86/concat-boolmasks.ll
+++ b/llvm/test/Transforms/VectorCombine/X86/concat-boolmasks.ll
@@ -80,30 +80,13 @@ define i64 @movmsk_i64_v8i32_v4i32(<4 x i32> %v0, <4 x i32> %v1) {
}
define i64 @movmsk_i64_v64i8_v16i8(<16 x i8> %v0, <16 x i8> %v1, <16 x i8> %v2, <16 x i8> %v3) {
-; SSE-LABEL: @movmsk_i64_v64i8_v16i8(
-; SSE-NEXT: [[TMP1:%.*]] = shufflevector <16 x i8> [[V3:%.*]], <16 x i8> [[V2:%.*]], <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
-; SSE-NEXT: [[TMP2:%.*]] = shufflevector <16 x i8> [[V1:%.*]], <16 x i8> [[V0:%.*]], <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
-; SSE-NEXT: [[TMP3:%.*]] = shufflevector <32 x i8> [[TMP1]], <32 x i8> [[TMP2]], <64 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60, i32 61, i32 62, i32 63>
-; SSE-NEXT: [[TMP4:%.*]] = icmp slt <64 x i8> [[TMP3]], zeroinitializer
-; SSE-NEXT: [[OR:%.*]] = bitcast <64 x i1> [[TMP4]] to i64
-; SSE-NEXT: ret i64 [[OR]]
-;
-; AVX2-LABEL: @movmsk_i64_v64i8_v16i8(
-; AVX2-NEXT: [[TMP1:%.*]] = shufflevector <16 x i8> [[V1:%.*]], <16 x i8> [[V0:%.*]], <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
-; AVX2-NEXT: [[TMP2:%.*]] = icmp slt <32 x i8> [[TMP1]], zeroinitializer
-; AVX2-NEXT: [[TMP3:%.*]] = shufflevector <16 x i8> [[V3:%.*]], <16 x i8> [[V2:%.*]], <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
-; AVX2-NEXT: [[TMP4:%.*]] = icmp slt <32 x i8> [[TMP3]], zeroinitializer
-; AVX2-NEXT: [[TMP5:%.*]] = shufflevector <32 x i1> [[TMP4]], <32 x i1> [[TMP2]], <64 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60, i32 61, i32 62, i32 63>
-; AVX2-NEXT: [[OR:%.*]] = bitcast <64 x i1> [[TMP5]] to i64
-; AVX2-NEXT: ret i64 [[OR]]
-;
-; AVX512-LABEL: @movmsk_i64_v64i8_v16i8(
-; AVX512-NEXT: [[TMP1:%.*]] = shufflevector <16 x i8> [[V1:%.*]], <16 x i8> [[V0:%.*]], <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
-; AVX512-NEXT: [[TMP2:%.*]] = shufflevector <16 x i8> [[V3:%.*]], <16 x i8> [[V2:%.*]], <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
-; AVX512-NEXT: [[TMP3:%.*]] = shufflevector <32 x i8> [[TMP2]], <32 x i8> [[TMP1]], <64 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60, i32 61, i32 62, i32 63>
-; AVX512-NEXT: [[TMP4:%.*]] = icmp slt <64 x i8> [[TMP3]], zeroinitializer
-; AVX512-NEXT: [[OR:%.*]] = bitcast <64 x i1> [[TMP4]] to i64
-; AVX512-NEXT: ret i64 [[OR]]
+; CHECK-LABEL: @movmsk_i64_v64i8_v16i8(
+; CHECK-NEXT: [[TMP1:%.*]] = shufflevector <16 x i8> [[V1:%.*]], <16 x i8> [[V0:%.*]], <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
+; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <16 x i8> [[V3:%.*]], <16 x i8> [[V2:%.*]], <32 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31>
+; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <32 x i8> [[TMP2]], <32 x i8> [[TMP1]], <64 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19, i32 20, i32 21, i32 22, i32 23, i32 24, i32 25, i32 26, i32 27, i32 28, i32 29, i32 30, i32 31, i32 32, i32 33, i32 34, i32 35, i32 36, i32 37, i32 38, i32 39, i32 40, i32 41, i32 42, i32 43, i32 44, i32 45, i32 46, i32 47, i32 48, i32 49, i32 50, i32 51, i32 52, i32 53, i32 54, i32 55, i32 56, i32 57, i32 58, i32 59, i32 60, i32 61, i32 62, i32 63>
+; CHECK-NEXT: [[TMP4:%.*]] = icmp slt <64 x i8> [[TMP3]], zeroinitializer
+; CHECK-NEXT: [[OR:%.*]] = bitcast <64 x i1> [[TMP4]] to i64
+; CHECK-NEXT: ret i64 [[OR]]
;
%c0 = icmp slt <16 x i8> %v0, zeroinitializer
%c1 = icmp slt <16 x i8> %v1, zeroinitializer
|
davemgreen
added a commit
to davemgreen/llvm-project
that referenced
this pull request
Dec 19, 2024
This fixes some regressions from recent changes to vector combine in llvm#120216. It allows shuffleToIdentity to look through fp casts as other casts, and makes sure mismatching vector types in splats and casts do not block the transform, as only the lanes should matter.
davemgreen
added a commit
that referenced
this pull request
Dec 20, 2024
This fixes some regressions from recent changes to vector combine in #120216. It allows shuffleToIdentity to look through fp casts as other casts, and makes sure mismatching vector types in splats and casts do not block the transform, as only the lanes should matter.
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.
We don't fold "shuffle (binop), (binop)" -> "binop (shuffle), (shuffle)" if the old/new costs are equal, but we can relax this if either new shuffle will constant fold as it will reduce instruction count.