shylie/llvm-project
1
0
Fork 0
Code Issues Pull Requests Actions 6 Packages Projects Releases Wiki Activity
llvm-project/llvm/test/Transforms/SLPVectorizer/AArch64/getelementptr.ll
Florian Hahn 02d3738be9
[AArch64,TTI] Remove RealUse check for vector insert/extract costs. (#146526) 
getVectorInstrCostHelper would return costs of zero for vector
inserts/extracts that move data between GPR and vector registers, if
there was no 'real' use, i.e. there was no corresponding existing
instruction.

This meant that passes like LoopVectorize and SLPVectorizer, which
likely are the main users of the interface, would understimate the cost
of insert/extracts that move data between GPR and vector registers,
which has non-trivial costs.

The patch removes the special case and only returns costs of zero for
lane 0 if it there is no need to transfer between integer and vector
registers.

This impacts a number of SLP test, and most of them look like general
improvements.I think the change should make things more accurate for any
AArch64 target, but if not it could also just be Apple CPU specific.

I am seeing +2% end-to-end improvements on SLP-heavy workloads.

PR: https://github.com/llvm/llvm-project/pull/146526
2025-07-15 15:19:27 +01:00

223 lines
11 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -S -mtriple=aarch64--linux-gnu -passes=slp-vectorizer -slp-threshold=-12 -pass-remarks-output=%t < %s | FileCheck %s
; RUN: cat %t | FileCheck -check-prefix=YAML %s
; RUN: opt -S -mtriple=aarch64--linux-gnu -passes=slp-vectorizer -slp-threshold=-12 -pass-remarks-output=%t < %s | FileCheck %s
; RUN: cat %t | FileCheck -check-prefix=YAML %s
; These tests check that we remove from consideration pairs of seed
; getelementptrs when they are known to have a constant difference. Such pairs
; are likely not good candidates for vectorization since one can be computed
; from the other. We use an unprofitable threshold to force vectorization.
;
; int getelementptr(int *g, int n, int w, int x, int y, int z) {
; int sum = 0;
; for (int i = 0; i < n ; ++i) {
; sum += g[2*i + w]; sum += g[2*i + x];
; sum += g[2*i + y]; sum += g[2*i + z];
; }
; return sum;
; }
;
; YAML-LABEL: Function: getelementptr_4x32
; YAML: --- !Passed
; YAML-NEXT: Pass: slp-vectorizer
; YAML-NEXT: Name: VectorizedList
; YAML-NEXT: Function: getelementptr_4x32
; YAML-NEXT: Args:
; YAML-NEXT: - String: 'SLP vectorized with cost '
; YAML-NEXT: - Cost: '6'
; YAML-NEXT: - String: ' and with tree size '
; YAML-NEXT: - TreeSize: '3'
; YAML: --- !Passed
; YAML-NEXT: Pass: slp-vectorizer
; YAML-NEXT: Name: VectorizedList
; YAML-NEXT: Function: getelementptr_4x32
; YAML-NEXT: Args:
; YAML-NEXT: - String: 'SLP vectorized with cost '
; YAML-NEXT: - Cost: '8'
; YAML-NEXT: - String: ' and with tree size '
; YAML-NEXT: - TreeSize: '3'
define i32 @getelementptr_4x32(ptr nocapture readonly %g, i32 %n, i32 %x, i32 %y, i32 %z) {
; CHECK-LABEL: @getelementptr_4x32(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[CMP31:%.*]] = icmp sgt i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[CMP31]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_COND_CLEANUP:%.*]]
; CHECK: for.body.preheader:
; CHECK-NEXT: [[TMP0:%.*]] = insertelement <2 x i32> <i32 0, i32 poison>, i32 [[X:%.*]], i32 1
; CHECK-NEXT: [[TMP1:%.*]] = insertelement <2 x i32> poison, i32 [[Y:%.*]], i32 0
; CHECK-NEXT: [[TMP2:%.*]] = insertelement <2 x i32> [[TMP1]], i32 [[Z:%.*]], i32 1
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.cond.cleanup.loopexit:
; CHECK-NEXT: [[ADD16:%.*]] = extractelement <2 x i32> [[TMP17:%.*]], i32 0
; CHECK-NEXT: br label [[FOR_COND_CLEANUP]]
; CHECK: for.cond.cleanup:
; CHECK-NEXT: [[SUM_0_LCSSA:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[ADD16]], [[FOR_COND_CLEANUP_LOOPEXIT:%.*]] ]
; CHECK-NEXT: ret i32 [[SUM_0_LCSSA]]
; CHECK: for.body:
; CHECK-NEXT: [[TMP6:%.*]] = phi <2 x i32> [ zeroinitializer, [[FOR_BODY_PREHEADER]] ], [ [[TMP17]], [[FOR_BODY]] ]
; CHECK-NEXT: [[TMP15:%.*]] = extractelement <2 x i32> [[TMP6]], i32 1
; CHECK-NEXT: [[T4:%.*]] = shl nsw i32 [[TMP15]], 1
; CHECK-NEXT: [[TMP16:%.*]] = extractelement <2 x i32> [[TMP6]], i32 0
; CHECK-NEXT: [[TMP3:%.*]] = insertelement <2 x i32> poison, i32 [[T4]], i32 0
; CHECK-NEXT: [[TMP4:%.*]] = shufflevector <2 x i32> [[TMP3]], <2 x i32> poison, <2 x i32> zeroinitializer
; CHECK-NEXT: [[TMP5:%.*]] = add nsw <2 x i32> [[TMP4]], [[TMP0]]
; CHECK-NEXT: [[TMP12:%.*]] = extractelement <2 x i32> [[TMP5]], i32 0
; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[G:%.*]], i32 [[TMP12]]
; CHECK-NEXT: [[T6:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
; CHECK-NEXT: [[ADD1:%.*]] = add nsw i32 [[T6]], [[TMP16]]
; CHECK-NEXT: [[TMP11:%.*]] = extractelement <2 x i32> [[TMP5]], i32 1
; CHECK-NEXT: [[ARRAYIDX5:%.*]] = getelementptr inbounds i32, ptr [[G]], i32 [[TMP11]]
; CHECK-NEXT: [[T8:%.*]] = load i32, ptr [[ARRAYIDX5]], align 4
; CHECK-NEXT: [[ADD6:%.*]] = add nsw i32 [[ADD1]], [[T8]]
; CHECK-NEXT: [[TMP18:%.*]] = add nsw <2 x i32> [[TMP4]], [[TMP2]]
; CHECK-NEXT: [[TMP13:%.*]] = extractelement <2 x i32> [[TMP18]], i32 0
; CHECK-NEXT: [[ARRAYIDX10:%.*]] = getelementptr inbounds i32, ptr [[G]], i32 [[TMP13]]
; CHECK-NEXT: [[T10:%.*]] = load i32, ptr [[ARRAYIDX10]], align 4
; CHECK-NEXT: [[ADD11:%.*]] = add nsw i32 [[ADD6]], [[T10]]
; CHECK-NEXT: [[TMP14:%.*]] = extractelement <2 x i32> [[TMP18]], i32 1
; CHECK-NEXT: [[ARRAYIDX15:%.*]] = getelementptr inbounds i32, ptr [[G]], i32 [[TMP14]]
; CHECK-NEXT: [[T12:%.*]] = load i32, ptr [[ARRAYIDX15]], align 4
; CHECK-NEXT: [[TMP19:%.*]] = insertelement <2 x i32> <i32 poison, i32 1>, i32 [[ADD11]], i32 0
; CHECK-NEXT: [[TMP20:%.*]] = insertelement <2 x i32> [[TMP6]], i32 [[T12]], i32 0
; CHECK-NEXT: [[TMP17]] = add nsw <2 x i32> [[TMP19]], [[TMP20]]
; CHECK-NEXT: [[INDVARS_IV_NEXT:%.*]] = extractelement <2 x i32> [[TMP17]], i32 1
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[INDVARS_IV_NEXT]], [[N]]
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP_LOOPEXIT]], label [[FOR_BODY]]
;
entry:
%cmp31 = icmp sgt i32 %n, 0
br i1 %cmp31, label %for.body.preheader, label %for.cond.cleanup
for.body.preheader:
br label %for.body
for.cond.cleanup.loopexit:
br label %for.cond.cleanup
for.cond.cleanup:
%sum.0.lcssa = phi i32 [ 0, %entry ], [ %add16, %for.cond.cleanup.loopexit ]
ret i32 %sum.0.lcssa
for.body:
%indvars.iv = phi i32 [ 0, %for.body.preheader ], [ %indvars.iv.next, %for.body ]
%sum.032 = phi i32 [ 0, %for.body.preheader ], [ %add16, %for.body ]
%t4 = shl nsw i32 %indvars.iv, 1
%t5 = add nsw i32 %t4, 0
%arrayidx = getelementptr inbounds i32, ptr %g, i32 %t5
%t6 = load i32, ptr %arrayidx, align 4
%add1 = add nsw i32 %t6, %sum.032
%t7 = add nsw i32 %t4, %x
%arrayidx5 = getelementptr inbounds i32, ptr %g, i32 %t7
%t8 = load i32, ptr %arrayidx5, align 4
%add6 = add nsw i32 %add1, %t8
%t9 = add nsw i32 %t4, %y
%arrayidx10 = getelementptr inbounds i32, ptr %g, i32 %t9
%t10 = load i32, ptr %arrayidx10, align 4
%add11 = add nsw i32 %add6, %t10
%t11 = add nsw i32 %t4, %z
%arrayidx15 = getelementptr inbounds i32, ptr %g, i32 %t11
%t12 = load i32, ptr %arrayidx15, align 4
%add16 = add nsw i32 %add11, %t12
%indvars.iv.next = add nuw nsw i32 %indvars.iv, 1
%exitcond = icmp eq i32 %indvars.iv.next , %n
br i1 %exitcond, label %for.cond.cleanup.loopexit, label %for.body
}
; YAML-LABEL: Function: getelementptr_2x32
; YAML: --- !Passed
; YAML: Pass: slp-vectorizer
; YAML: Name: VectorizedList
; YAML: Function: getelementptr_2x32
; YAML: Args:
; YAML: - String: 'SLP vectorized with cost '
; YAML: - Cost: '8'
; YAML-NEXT: - String: ' and with tree size '
; YAML-NEXT: - TreeSize: '3'
define i32 @getelementptr_2x32(ptr nocapture readonly %g, i32 %n, i32 %x, i32 %y, i32 %z) {
; CHECK-LABEL: @getelementptr_2x32(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[CMP31:%.*]] = icmp sgt i32 [[N:%.*]], 0
; CHECK-NEXT: br i1 [[CMP31]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_COND_CLEANUP:%.*]]
; CHECK: for.body.preheader:
; CHECK-NEXT: [[TMP4:%.*]] = insertelement <2 x i32> poison, i32 [[Y:%.*]], i32 0
; CHECK-NEXT: [[TMP0:%.*]] = insertelement <2 x i32> [[TMP4]], i32 [[Z:%.*]], i32 1
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.cond.cleanup.loopexit:
; CHECK-NEXT: [[OP_RDX:%.*]] = extractelement <2 x i32> [[TMP13:%.*]], i32 0
; CHECK-NEXT: br label [[FOR_COND_CLEANUP]]
; CHECK: for.cond.cleanup:
; CHECK-NEXT: [[SUM_0_LCSSA:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[OP_RDX]], [[FOR_COND_CLEANUP_LOOPEXIT:%.*]] ]
; CHECK-NEXT: ret i32 [[SUM_0_LCSSA]]
; CHECK: for.body:
; CHECK-NEXT: [[TMP6:%.*]] = phi <2 x i32> [ zeroinitializer, [[FOR_BODY_PREHEADER]] ], [ [[TMP13]], [[FOR_BODY]] ]
; CHECK-NEXT: [[TMP12:%.*]] = extractelement <2 x i32> [[TMP6]], i32 1
; CHECK-NEXT: [[T4:%.*]] = shl nsw i32 [[TMP12]], 1
; CHECK-NEXT: [[T5:%.*]] = add nsw i32 [[T4]], 0
; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i32, ptr [[G:%.*]], i32 [[T5]]
; CHECK-NEXT: [[T6:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
; CHECK-NEXT: [[TMP5:%.*]] = extractelement <2 x i32> [[TMP6]], i32 0
; CHECK-NEXT: [[ADD1:%.*]] = add nsw i32 [[T6]], [[TMP5]]
; CHECK-NEXT: [[T7:%.*]] = add nsw i32 [[T4]], 1
; CHECK-NEXT: [[ARRAYIDX5:%.*]] = getelementptr inbounds i32, ptr [[G]], i32 [[T7]]
; CHECK-NEXT: [[T8:%.*]] = load i32, ptr [[ARRAYIDX5]], align 4
; CHECK-NEXT: [[ADD6:%.*]] = add nsw i32 [[ADD1]], [[T8]]
; CHECK-NEXT: [[TMP1:%.*]] = insertelement <2 x i32> poison, i32 [[T4]], i32 0
; CHECK-NEXT: [[TMP2:%.*]] = shufflevector <2 x i32> [[TMP1]], <2 x i32> poison, <2 x i32> zeroinitializer
; CHECK-NEXT: [[TMP3:%.*]] = add nsw <2 x i32> [[TMP2]], [[TMP0]]
; CHECK-NEXT: [[TMP9:%.*]] = extractelement <2 x i32> [[TMP3]], i32 0
; CHECK-NEXT: [[ARRAYIDX11:%.*]] = getelementptr inbounds i32, ptr [[G]], i32 [[TMP9]]
; CHECK-NEXT: [[T10:%.*]] = load i32, ptr [[ARRAYIDX11]], align 4
; CHECK-NEXT: [[ADD11:%.*]] = add nsw i32 [[ADD6]], [[T10]]
; CHECK-NEXT: [[TMP10:%.*]] = extractelement <2 x i32> [[TMP3]], i32 1
; CHECK-NEXT: [[ARRAYIDX15:%.*]] = getelementptr inbounds i32, ptr [[G]], i32 [[TMP10]]
; CHECK-NEXT: [[T12:%.*]] = load i32, ptr [[ARRAYIDX15]], align 4
; CHECK-NEXT: [[TMP11:%.*]] = insertelement <2 x i32> <i32 poison, i32 1>, i32 [[ADD11]], i32 0
; CHECK-NEXT: [[TMP14:%.*]] = insertelement <2 x i32> [[TMP6]], i32 [[T12]], i32 0
; CHECK-NEXT: [[TMP13]] = add nsw <2 x i32> [[TMP11]], [[TMP14]]
; CHECK-NEXT: [[INDVARS_IV_NEXT:%.*]] = extractelement <2 x i32> [[TMP13]], i32 1
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[INDVARS_IV_NEXT]], [[N]]
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP_LOOPEXIT]], label [[FOR_BODY]]
;
entry:
%cmp31 = icmp sgt i32 %n, 0
br i1 %cmp31, label %for.body.preheader, label %for.cond.cleanup
for.body.preheader:
br label %for.body
for.cond.cleanup.loopexit:
br label %for.cond.cleanup
for.cond.cleanup:
%sum.0.lcssa = phi i32 [ 0, %entry ], [ %add16, %for.cond.cleanup.loopexit ]
ret i32 %sum.0.lcssa
for.body:
%indvars.iv = phi i32 [ 0, %for.body.preheader ], [ %indvars.iv.next, %for.body ]
%sum.032 = phi i32 [ 0, %for.body.preheader ], [ %add16, %for.body ]
%t4 = shl nsw i32 %indvars.iv, 1
%t5 = add nsw i32 %t4, 0
%arrayidx = getelementptr inbounds i32, ptr %g, i32 %t5
%t6 = load i32, ptr %arrayidx, align 4
%add1 = add nsw i32 %t6, %sum.032
%t7 = add nsw i32 %t4, 1
%arrayidx5 = getelementptr inbounds i32, ptr %g, i32 %t7
%t8 = load i32, ptr %arrayidx5, align 4
%add6 = add nsw i32 %add1, %t8
%t9 = add nsw i32 %t4, %y
%arrayidx10 = getelementptr inbounds i32, ptr %g, i32 %t9
%t10 = load i32, ptr %arrayidx10, align 4
%add11 = add nsw i32 %add6, %t10
%t11 = add nsw i32 %t4, %z
%arrayidx15 = getelementptr inbounds i32, ptr %g, i32 %t11
%t12 = load i32, ptr %arrayidx15, align 4
%add16 = add nsw i32 %add11, %t12
%indvars.iv.next = add nuw nsw i32 %indvars.iv, 1
%exitcond = icmp eq i32 %indvars.iv.next , %n
br i1 %exitcond, label %for.cond.cleanup.loopexit, label %for.body
}
Reference in New Issue View Git Blame Copy Permalink