shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/llvm/test/Transforms/SLPVectorizer/X86/powof2div.ll
Philip Reames 7d6e8f2a96 [slp] Delete dead scalar instructions feeding vectorized instructions 
If we vectorize a e.g. store, we leave around a bunch of getelementptrs for the individual scalar stores which we removed. We can go ahead and delete them as well.

This is purely for test output quality and readability. It should have no effect in any sane pipeline.

Differential Revision: https://reviews.llvm.org/D122493
2022-03-28 20:10:13 -07:00

131 lines
6.5 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -basic-aa -slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -mattr=+sse2 | FileCheck %s --check-prefixes=CHECK,SSE
; RUN: opt < %s -basic-aa -slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx | FileCheck %s --check-prefixes=CHECK,AVX
; RUN: opt < %s -basic-aa -slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx2 | FileCheck %s --check-prefixes=CHECK,AVX
define void @powof2div_uniform(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* noalias nocapture readonly %c){
; CHECK-LABEL: @powof2div_uniform(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = bitcast i32* [[B:%.*]] to <4 x i32>*
; CHECK-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* [[TMP0]], align 4
; CHECK-NEXT: [[TMP2:%.*]] = bitcast i32* [[C:%.*]] to <4 x i32>*
; CHECK-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[TMP2]], align 4
; CHECK-NEXT: [[TMP4:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP1]]
; CHECK-NEXT: [[TMP5:%.*]] = sdiv <4 x i32> [[TMP4]], <i32 2, i32 2, i32 2, i32 2>
; CHECK-NEXT: [[TMP6:%.*]] = bitcast i32* [[A:%.*]] to <4 x i32>*
; CHECK-NEXT: store <4 x i32> [[TMP5]], <4 x i32>* [[TMP6]], align 4
; CHECK-NEXT: ret void
;
entry:
%0 = load i32, i32* %b, align 4
%1 = load i32, i32* %c, align 4
%add = add nsw i32 %1, %0
%div = sdiv i32 %add, 2
store i32 %div, i32* %a, align 4
%arrayidx3 = getelementptr inbounds i32, i32* %b, i64 1
%2 = load i32, i32* %arrayidx3, align 4
%arrayidx4 = getelementptr inbounds i32, i32* %c, i64 1
%3 = load i32, i32* %arrayidx4, align 4
%add5 = add nsw i32 %3, %2
%div6 = sdiv i32 %add5, 2
%arrayidx7 = getelementptr inbounds i32, i32* %a, i64 1
store i32 %div6, i32* %arrayidx7, align 4
%arrayidx8 = getelementptr inbounds i32, i32* %b, i64 2
%4 = load i32, i32* %arrayidx8, align 4
%arrayidx9 = getelementptr inbounds i32, i32* %c, i64 2
%5 = load i32, i32* %arrayidx9, align 4
%add10 = add nsw i32 %5, %4
%div11 = sdiv i32 %add10, 2
%arrayidx12 = getelementptr inbounds i32, i32* %a, i64 2
store i32 %div11, i32* %arrayidx12, align 4
%arrayidx13 = getelementptr inbounds i32, i32* %b, i64 3
%6 = load i32, i32* %arrayidx13, align 4
%arrayidx14 = getelementptr inbounds i32, i32* %c, i64 3
%7 = load i32, i32* %arrayidx14, align 4
%add15 = add nsw i32 %7, %6
%div16 = sdiv i32 %add15, 2
%arrayidx17 = getelementptr inbounds i32, i32* %a, i64 3
store i32 %div16, i32* %arrayidx17, align 4
ret void
}
define void @powof2div_nonuniform(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* noalias nocapture readonly %c){
; SSE-LABEL: @powof2div_nonuniform(
; SSE-NEXT: entry:
; SSE-NEXT: [[TMP0:%.*]] = load i32, i32* [[B:%.*]], align 4
; SSE-NEXT: [[TMP1:%.*]] = load i32, i32* [[C:%.*]], align 4
; SSE-NEXT: [[ADD:%.*]] = add nsw i32 [[TMP1]], [[TMP0]]
; SSE-NEXT: [[DIV:%.*]] = sdiv i32 [[ADD]], 2
; SSE-NEXT: store i32 [[DIV]], i32* [[A:%.*]], align 4
; SSE-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 1
; SSE-NEXT: [[TMP2:%.*]] = load i32, i32* [[ARRAYIDX3]], align 4
; SSE-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 1
; SSE-NEXT: [[TMP3:%.*]] = load i32, i32* [[ARRAYIDX4]], align 4
; SSE-NEXT: [[ADD5:%.*]] = add nsw i32 [[TMP3]], [[TMP2]]
; SSE-NEXT: [[DIV6:%.*]] = sdiv i32 [[ADD5]], 4
; SSE-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 1
; SSE-NEXT: store i32 [[DIV6]], i32* [[ARRAYIDX7]], align 4
; SSE-NEXT: [[ARRAYIDX8:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 2
; SSE-NEXT: [[TMP4:%.*]] = load i32, i32* [[ARRAYIDX8]], align 4
; SSE-NEXT: [[ARRAYIDX9:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 2
; SSE-NEXT: [[TMP5:%.*]] = load i32, i32* [[ARRAYIDX9]], align 4
; SSE-NEXT: [[ADD10:%.*]] = add nsw i32 [[TMP5]], [[TMP4]]
; SSE-NEXT: [[DIV11:%.*]] = sdiv i32 [[ADD10]], 8
; SSE-NEXT: [[ARRAYIDX12:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 2
; SSE-NEXT: store i32 [[DIV11]], i32* [[ARRAYIDX12]], align 4
; SSE-NEXT: [[ARRAYIDX13:%.*]] = getelementptr inbounds i32, i32* [[B]], i64 3
; SSE-NEXT: [[TMP6:%.*]] = load i32, i32* [[ARRAYIDX13]], align 4
; SSE-NEXT: [[ARRAYIDX14:%.*]] = getelementptr inbounds i32, i32* [[C]], i64 3
; SSE-NEXT: [[TMP7:%.*]] = load i32, i32* [[ARRAYIDX14]], align 4
; SSE-NEXT: [[ADD15:%.*]] = add nsw i32 [[TMP7]], [[TMP6]]
; SSE-NEXT: [[DIV16:%.*]] = sdiv i32 [[ADD15]], 16
; SSE-NEXT: [[ARRAYIDX17:%.*]] = getelementptr inbounds i32, i32* [[A]], i64 3
; SSE-NEXT: store i32 [[DIV16]], i32* [[ARRAYIDX17]], align 4
; SSE-NEXT: ret void
;
; AVX-LABEL: @powof2div_nonuniform(
; AVX-NEXT: entry:
; AVX-NEXT: [[TMP0:%.*]] = bitcast i32* [[B:%.*]] to <4 x i32>*
; AVX-NEXT: [[TMP1:%.*]] = load <4 x i32>, <4 x i32>* [[TMP0]], align 4
; AVX-NEXT: [[TMP2:%.*]] = bitcast i32* [[C:%.*]] to <4 x i32>*
; AVX-NEXT: [[TMP3:%.*]] = load <4 x i32>, <4 x i32>* [[TMP2]], align 4
; AVX-NEXT: [[TMP4:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP1]]
; AVX-NEXT: [[TMP5:%.*]] = sdiv <4 x i32> [[TMP4]], <i32 2, i32 4, i32 8, i32 16>
; AVX-NEXT: [[TMP6:%.*]] = bitcast i32* [[A:%.*]] to <4 x i32>*
; AVX-NEXT: store <4 x i32> [[TMP5]], <4 x i32>* [[TMP6]], align 4
; AVX-NEXT: ret void
;
entry:
%0 = load i32, i32* %b, align 4
%1 = load i32, i32* %c, align 4
%add = add nsw i32 %1, %0
%div = sdiv i32 %add, 2
store i32 %div, i32* %a, align 4
%arrayidx3 = getelementptr inbounds i32, i32* %b, i64 1
%2 = load i32, i32* %arrayidx3, align 4
%arrayidx4 = getelementptr inbounds i32, i32* %c, i64 1
%3 = load i32, i32* %arrayidx4, align 4
%add5 = add nsw i32 %3, %2
%div6 = sdiv i32 %add5, 4
%arrayidx7 = getelementptr inbounds i32, i32* %a, i64 1
store i32 %div6, i32* %arrayidx7, align 4
%arrayidx8 = getelementptr inbounds i32, i32* %b, i64 2
%4 = load i32, i32* %arrayidx8, align 4
%arrayidx9 = getelementptr inbounds i32, i32* %c, i64 2
%5 = load i32, i32* %arrayidx9, align 4
%add10 = add nsw i32 %5, %4
%div11 = sdiv i32 %add10, 8
%arrayidx12 = getelementptr inbounds i32, i32* %a, i64 2
store i32 %div11, i32* %arrayidx12, align 4
%arrayidx13 = getelementptr inbounds i32, i32* %b, i64 3
%6 = load i32, i32* %arrayidx13, align 4
%arrayidx14 = getelementptr inbounds i32, i32* %c, i64 3
%7 = load i32, i32* %arrayidx14, align 4
%add15 = add nsw i32 %7, %6
%div16 = sdiv i32 %add15, 16
%arrayidx17 = getelementptr inbounds i32, i32* %a, i64 3
store i32 %div16, i32* %arrayidx17, align 4
ret void
}
Reference in New Issue View Git Blame Copy Permalink