AMDGPU: Simplify demanded vector elts of readfirstlane sources #128646
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
This was referenced Feb 25, 2025
|
@llvm/pr-subscribers-llvm-transforms @llvm/pr-subscribers-backend-amdgpu Author: Matt Arsenault (arsenm) ChangesStub implementation of simplifyDemandedVectorEltsIntrinsic for Full diff: https://github.com/llvm/llvm-project/pull/128646.diff 2 Files Affected:
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp b/llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp
index ebc00e59584ac..617974713d6f0 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp
@@ -1544,6 +1544,10 @@ std::optional<Value *> GCNTTIImpl::simplifyDemandedVectorEltsIntrinsic(
std::function<void(Instruction *, unsigned, APInt, APInt &)>
SimplifyAndSetOp) const {
switch (II.getIntrinsicID()) {
+ case Intrinsic::amdgcn_readfirstlane:
+ // TODO: For a vector extract, should reduce the intrinsic call type.
+ SimplifyAndSetOp(&II, 0, DemandedElts, UndefElts);
+ return std::nullopt;
case Intrinsic::amdgcn_raw_buffer_load:
case Intrinsic::amdgcn_raw_ptr_buffer_load:
case Intrinsic::amdgcn_raw_buffer_load_format:
diff --git a/llvm/test/Transforms/InstCombine/AMDGPU/simplify-demanded-vector-elts-lane-intrinsics.ll b/llvm/test/Transforms/InstCombine/AMDGPU/simplify-demanded-vector-elts-lane-intrinsics.ll
index 83d9d0d032ed1..836c739048411 100644
--- a/llvm/test/Transforms/InstCombine/AMDGPU/simplify-demanded-vector-elts-lane-intrinsics.ll
+++ b/llvm/test/Transforms/InstCombine/AMDGPU/simplify-demanded-vector-elts-lane-intrinsics.ll
@@ -306,10 +306,9 @@ define <2 x i16> @extract_elt13_v4i16readfirstlane(<4 x i16> %src) {
define <2 x i32> @extract_elt13_v4i32_readfirstlane_source_simplify0(i32 %src0, i32 %src2) {
; CHECK-LABEL: define <2 x i32> @extract_elt13_v4i32_readfirstlane_source_simplify0(
; CHECK-SAME: i32 [[SRC0:%.*]], i32 [[SRC2:%.*]]) #[[ATTR0]] {
-; CHECK-NEXT: [[INS_0:%.*]] = insertelement <4 x i32> poison, i32 [[SRC0]], i64 0
-; CHECK-NEXT: [[INS_1:%.*]] = shufflevector <4 x i32> [[INS_0]], <4 x i32> poison, <4 x i32> <i32 0, i32 0, i32 poison, i32 poison>
+; CHECK-NEXT: [[INS_1:%.*]] = insertelement <4 x i32> poison, i32 [[SRC0]], i64 1
; CHECK-NEXT: [[VEC:%.*]] = call <4 x i32> @llvm.amdgcn.readfirstlane.v4i32(<4 x i32> [[INS_1]])
-; CHECK-NEXT: [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[VEC]], <4 x i32> poison, <2 x i32> <i32 1, i32 3>
+; CHECK-NEXT: [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[VEC]], <4 x i32> poison, <2 x i32> <i32 1, i32 poison>
; CHECK-NEXT: ret <2 x i32> [[SHUFFLE]]
;
%ins.0 = insertelement <4 x i32> poison, i32 %src0, i32 0
@@ -338,11 +337,7 @@ define < 2 x i32> @extract_elt13_v4i32_readfirstlane_source_simplify1(i32 %src0,
define < 2 x i32> @extract_elt13_v4i32_readfirstlane_source_simplify2(i32 %src0, i32 %src2) {
; CHECK-LABEL: define <2 x i32> @extract_elt13_v4i32_readfirstlane_source_simplify2(
; CHECK-SAME: i32 [[SRC0:%.*]], i32 [[SRC2:%.*]]) #[[ATTR0]] {
-; CHECK-NEXT: [[INS_0:%.*]] = insertelement <4 x i32> poison, i32 [[SRC0]], i64 0
-; CHECK-NEXT: [[INS_1:%.*]] = shufflevector <4 x i32> [[INS_0]], <4 x i32> poison, <4 x i32> <i32 0, i32 poison, i32 0, i32 poison>
-; CHECK-NEXT: [[VEC:%.*]] = call <4 x i32> @llvm.amdgcn.readfirstlane.v4i32(<4 x i32> [[INS_1]])
-; CHECK-NEXT: [[SHUFFLE:%.*]] = shufflevector <4 x i32> [[VEC]], <4 x i32> poison, <2 x i32> <i32 1, i32 3>
-; CHECK-NEXT: ret <2 x i32> [[SHUFFLE]]
+; CHECK-NEXT: ret <2 x i32> poison
;
%ins.0 = insertelement <4 x i32> poison, i32 %src0, i32 0
%ins.1 = insertelement <4 x i32> %ins.0, i32 %src0, i32 2
|
arsenm
force-pushed
the
users/arsenm/amdgpu/add-baseline-tests-simplify-demanded-vector-elts-readfirstlane
branch
from
fe33057 to
859a8cb
Compare
Base automatically changed from
users/arsenm/amdgpu/add-baseline-tests-simplify-demanded-vector-elts-readfirstlane
to
main
February 28, 2025 05:57
Stub implementation of simplifyDemandedVectorEltsIntrinsic for readfirstlane.
arsenm
force-pushed
the
users/arsenm/amdgpu/simplify-demanded-vector-elts-readfirstlane-src
branch
from
ca0b3a4 to
e32caff
Compare
arsenm
deleted the
users/arsenm/amdgpu/simplify-demanded-vector-elts-readfirstlane-src
branch
cheezeburglar
pushed a commit
to cheezeburglar/llvm-project
that referenced
this pull request
Feb 28, 2025
…128646) Stub implementation of simplifyDemandedVectorEltsIntrinsic for readfirstlane.
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.
Stub implementation of simplifyDemandedVectorEltsIntrinsic for
readfirstlane.