caio/clang-p2996
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
05727d94de75e2e98e6ae37c36aeb4f6ca6b2aed
clang-p2996/llvm/test/Transforms/LoopVectorize/runtime-check.ll
Florian Hahn 68469a80cb [LV] Disable runtime unrolling for vectorized loops. 
This patch adds metadata to disable runtime unrolling to the vectorized
loop. If runtime unrolling/interleaving is considered profitable, LV
will interleave the loop directly. There should be no need to perform
runtime unrolling at a later stage.

Note that we already add metadata to disable runtime unrolling to the
scalar loop after vectorization.

The additional unrolling unnecessarily increases code size and compile
time. In addition to that we have several bug reports of unncessary
runtime unrolling for vectorized loops, e.g. PR40961

Compile-time improvements:

  NewPM-O3: -1.04%
  NewPM-ReleaseThinLTO: -0.59%
  NewPM-ReleaseLTO-g: -0.97%

https://llvm-compile-time-tracker.com/compare.php?from=ce1be13a868d0f8afa367975558c1a6175cce33a&to=78bc2e67f22e9e10e61cdb6cdac4bb857d95eb1b&stat=instructions:u

Fixes #40306.

Reviewed By: lebedev.ri, nikic

Differential Revision: https://reviews.llvm.org/D115261
2023-01-06 10:56:17 +00:00

399 lines
23 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -aa-pipeline=basic-aa -passes=loop-vectorize,dce,instcombine -force-vector-interleave=1 -force-vector-width=4 -S | FileCheck %s
; RUN: opt < %s -aa-pipeline= -passes=loop-vectorize -S -pass-remarks-analysis='loop-vectorize' 2>&1 | FileCheck %s -check-prefix=FORCED_OPTSIZE
target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
; Make sure we vectorize this loop:
; int foo(float *a, float *b, int n) {
; for (int i=0; i<n; ++i)
; a[i] = b[i] * 3;
; }
define i32 @foo(float* nocapture %a, float* nocapture %b, i32 %n) nounwind uwtable ssp {
; CHECK-LABEL: @foo(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[B2:%.*]] = ptrtoint float* [[B:%.*]] to i64, !dbg [[DBG4:![0-9]+]]
; CHECK-NEXT: [[A1:%.*]] = ptrtoint float* [[A:%.*]] to i64, !dbg [[DBG4]]
; CHECK-NEXT: [[CMP6:%.*]] = icmp sgt i32 [[N:%.*]], 0, !dbg [[DBG4]]
; CHECK-NEXT: br i1 [[CMP6]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END:%.*]], !dbg [[DBG4]]
; CHECK: for.body.preheader:
; CHECK-NEXT: [[TMP0:%.*]] = add i32 [[N]], -1, !dbg [[DBG9:![0-9]+]]
; CHECK-NEXT: [[TMP1:%.*]] = zext i32 [[TMP0]] to i64, !dbg [[DBG9]]
; CHECK-NEXT: [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1, !dbg [[DBG9]]
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i32 [[N]], 4, !dbg [[DBG9]]
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_MEMCHECK:%.*]], !dbg [[DBG9]]
; CHECK: vector.memcheck:
; CHECK-NEXT: [[TMP3:%.*]] = sub i64 [[A1]], [[B2]], !dbg [[DBG9]]
; CHECK-NEXT: [[DIFF_CHECK:%.*]] = icmp ult i64 [[TMP3]], 16, !dbg [[DBG9]]
; CHECK-NEXT: br i1 [[DIFF_CHECK]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]], !dbg [[DBG9]]
; CHECK: vector.ph:
; CHECK-NEXT: [[N_VEC:%.*]] = and i64 [[TMP2]], -4, !dbg [[DBG9]]
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]], !dbg [[DBG9]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ], !dbg [[DBG9]]
; CHECK-NEXT: [[TMP4:%.*]] = getelementptr inbounds float, float* [[B]], i64 [[INDEX]], !dbg [[DBG9]]
; CHECK-NEXT: [[TMP5:%.*]] = bitcast float* [[TMP4]] to <4 x float>*, !dbg [[DBG9]]
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <4 x float>, <4 x float>* [[TMP5]], align 4, !dbg [[DBG9]]
; CHECK-NEXT: [[TMP6:%.*]] = fmul <4 x float> [[WIDE_LOAD]], <float 3.000000e+00, float 3.000000e+00, float 3.000000e+00, float 3.000000e+00>, !dbg [[DBG9]]
; CHECK-NEXT: [[TMP7:%.*]] = getelementptr inbounds float, float* [[A]], i64 [[INDEX]], !dbg [[DBG9]]
; CHECK-NEXT: [[TMP8:%.*]] = bitcast float* [[TMP7]] to <4 x float>*, !dbg [[DBG9]]
; CHECK-NEXT: store <4 x float> [[TMP6]], <4 x float>* [[TMP8]], align 4, !dbg [[DBG9]]
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4, !dbg [[DBG9]]
; CHECK-NEXT: [[TMP9:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]], !dbg [[DBG9]]
; CHECK-NEXT: br i1 [[TMP9]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !dbg [[DBG9]], !llvm.loop [[LOOP10:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]], !dbg [[DBG9]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[FOR_END_LOOPEXIT:%.*]], label [[SCALAR_PH]], !dbg [[DBG9]]
; CHECK: scalar.ph:
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[FOR_BODY_PREHEADER]] ], [ 0, [[VECTOR_MEMCHECK]] ], !dbg [[DBG9]]
; CHECK-NEXT: br label [[FOR_BODY:%.*]], !dbg [[DBG9]]
; CHECK: for.body:
; CHECK-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], !dbg [[DBG9]]
; CHECK-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds float, float* [[B]], i64 [[INDVARS_IV]], !dbg [[DBG9]]
; CHECK-NEXT: [[TMP10:%.*]] = load float, float* [[ARRAYIDX]], align 4, !dbg [[DBG9]]
; CHECK-NEXT: [[MUL:%.*]] = fmul float [[TMP10]], 3.000000e+00, !dbg [[DBG9]]
; CHECK-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds float, float* [[A]], i64 [[INDVARS_IV]], !dbg [[DBG9]]
; CHECK-NEXT: store float [[MUL]], float* [[ARRAYIDX2]], align 4, !dbg [[DBG9]]
; CHECK-NEXT: [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 1, !dbg [[DBG9]]
; CHECK-NEXT: [[LFTR_WIDEIV:%.*]] = trunc i64 [[INDVARS_IV_NEXT]] to i32, !dbg [[DBG9]]
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[LFTR_WIDEIV]], [[N]], !dbg [[DBG9]]
; CHECK-NEXT: br i1 [[EXITCOND]], label [[FOR_END_LOOPEXIT]], label [[FOR_BODY]], !dbg [[DBG9]], !llvm.loop [[LOOP13:![0-9]+]]
; CHECK: for.end.loopexit:
; CHECK-NEXT: br label [[FOR_END]], !dbg [[DBG14:![0-9]+]]
; CHECK: for.end:
; CHECK-NEXT: ret i32 undef, !dbg [[DBG14]]
;
; FORCED_OPTSIZE-LABEL: @foo(
; FORCED_OPTSIZE-NEXT: entry:
; FORCED_OPTSIZE-NEXT: [[CMP6:%.*]] = icmp sgt i32 [[N:%.*]], 0, !dbg [[DBG4:![0-9]+]]
; FORCED_OPTSIZE-NEXT: br i1 [[CMP6]], label [[FOR_BODY_PREHEADER:%.*]], label [[FOR_END:%.*]], !dbg [[DBG4]]
; FORCED_OPTSIZE: for.body.preheader:
; FORCED_OPTSIZE-NEXT: br label [[FOR_BODY:%.*]], !dbg [[DBG9:![0-9]+]]
; FORCED_OPTSIZE: for.body:
; FORCED_OPTSIZE-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ], [ 0, [[FOR_BODY_PREHEADER]] ], !dbg [[DBG9]]
; FORCED_OPTSIZE-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds float, float* [[B:%.*]], i64 [[INDVARS_IV]], !dbg [[DBG9]]
; FORCED_OPTSIZE-NEXT: [[TMP0:%.*]] = load float, float* [[ARRAYIDX]], align 4, !dbg [[DBG9]]
; FORCED_OPTSIZE-NEXT: [[MUL:%.*]] = fmul float [[TMP0]], 3.000000e+00, !dbg [[DBG9]]
; FORCED_OPTSIZE-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds float, float* [[A:%.*]], i64 [[INDVARS_IV]], !dbg [[DBG9]]
; FORCED_OPTSIZE-NEXT: store float [[MUL]], float* [[ARRAYIDX2]], align 4, !dbg [[DBG9]]
; FORCED_OPTSIZE-NEXT: [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 1, !dbg [[DBG9]]
; FORCED_OPTSIZE-NEXT: [[LFTR_WIDEIV:%.*]] = trunc i64 [[INDVARS_IV_NEXT]] to i32, !dbg [[DBG9]]
; FORCED_OPTSIZE-NEXT: [[EXITCOND:%.*]] = icmp eq i32 [[LFTR_WIDEIV]], [[N]], !dbg [[DBG9]]
; FORCED_OPTSIZE-NEXT: br i1 [[EXITCOND]], label [[FOR_END_LOOPEXIT:%.*]], label [[FOR_BODY]], !dbg [[DBG9]]
; FORCED_OPTSIZE: for.end.loopexit:
; FORCED_OPTSIZE-NEXT: br label [[FOR_END]], !dbg [[DBG10:![0-9]+]]
; FORCED_OPTSIZE: for.end:
; FORCED_OPTSIZE-NEXT: ret i32 undef, !dbg [[DBG10]]
;
entry:
%cmp6 = icmp sgt i32 %n, 0, !dbg !6
br i1 %cmp6, label %for.body, label %for.end, !dbg !6
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ], !dbg !7
%arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv, !dbg !7
%0 = load float, float* %arrayidx, align 4, !dbg !7
%mul = fmul float %0, 3.000000e+00, !dbg !7
%arrayidx2 = getelementptr inbounds float, float* %a, i64 %indvars.iv, !dbg !7
store float %mul, float* %arrayidx2, align 4, !dbg !7
%indvars.iv.next = add i64 %indvars.iv, 1, !dbg !7
%lftr.wideiv = trunc i64 %indvars.iv.next to i32, !dbg !7
%exitcond = icmp eq i32 %lftr.wideiv, %n, !dbg !7
br i1 %exitcond, label %for.end, label %for.body, !dbg !7
for.end: ; preds = %for.body, %entry
ret i32 undef, !dbg !8
}
; Make sure that we try to vectorize loops with a runtime check if the
; dependency check fails.
define void @test_runtime_check(float* %a, float %b, i64 %offset, i64 %offset2, i64 %n) {
; CHECK-LABEL: @test_runtime_check(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N:%.*]], 4
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_MEMCHECK:%.*]]
; CHECK: vector.memcheck:
; CHECK-NEXT: [[SCEVGEP:%.*]] = getelementptr float, float* [[A:%.*]], i64 [[OFFSET:%.*]]
; CHECK-NEXT: [[TMP0:%.*]] = add i64 [[N]], [[OFFSET]]
; CHECK-NEXT: [[SCEVGEP2:%.*]] = getelementptr float, float* [[A]], i64 [[TMP0]]
; CHECK-NEXT: [[SCEVGEP4:%.*]] = getelementptr float, float* [[A]], i64 [[OFFSET2:%.*]]
; CHECK-NEXT: [[TMP1:%.*]] = add i64 [[N]], [[OFFSET2]]
; CHECK-NEXT: [[SCEVGEP6:%.*]] = getelementptr float, float* [[A]], i64 [[TMP1]]
; CHECK-NEXT: [[BOUND0:%.*]] = icmp ult float* [[SCEVGEP]], [[SCEVGEP6]]
; CHECK-NEXT: [[BOUND1:%.*]] = icmp ult float* [[SCEVGEP4]], [[SCEVGEP2]]
; CHECK-NEXT: [[FOUND_CONFLICT:%.*]] = and i1 [[BOUND0]], [[BOUND1]]
; CHECK-NEXT: br i1 [[FOUND_CONFLICT]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
; CHECK-NEXT: [[N_VEC:%.*]] = and i64 [[N]], -4
; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x float> poison, float [[B:%.*]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x float> [[BROADCAST_SPLATINSERT]], <4 x float> poison, <4 x i32> zeroinitializer
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP2:%.*]] = add i64 [[INDEX]], [[OFFSET]]
; CHECK-NEXT: [[TMP3:%.*]] = getelementptr inbounds float, float* [[A]], i64 [[TMP2]]
; CHECK-NEXT: [[TMP4:%.*]] = bitcast float* [[TMP3]] to <4 x float>*
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <4 x float>, <4 x float>* [[TMP4]], align 4, !alias.scope !15, !noalias !18
; CHECK-NEXT: [[TMP5:%.*]] = add i64 [[INDEX]], [[OFFSET2]]
; CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds float, float* [[A]], i64 [[TMP5]]
; CHECK-NEXT: [[TMP7:%.*]] = bitcast float* [[TMP6]] to <4 x float>*
; CHECK-NEXT: [[WIDE_LOAD8:%.*]] = load <4 x float>, <4 x float>* [[TMP7]], align 4, !alias.scope !18
; CHECK-NEXT: [[TMP8:%.*]] = fmul fast <4 x float> [[BROADCAST_SPLAT]], [[WIDE_LOAD8]]
; CHECK-NEXT: [[TMP9:%.*]] = fadd fast <4 x float> [[WIDE_LOAD]], [[TMP8]]
; CHECK-NEXT: [[TMP10:%.*]] = bitcast float* [[TMP3]] to <4 x float>*
; CHECK-NEXT: store <4 x float> [[TMP9]], <4 x float>* [[TMP10]], align 4, !alias.scope !15, !noalias !18
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT: [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP20:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N_VEC]], [[N]]
; CHECK-NEXT: br i1 [[CMP_N]], label [[LOOPEXIT:%.*]], label [[SCALAR_PH]]
; CHECK: scalar.ph:
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_MEMCHECK]] ]
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.body:
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
; CHECK-NEXT: [[IND_SUM:%.*]] = add i64 [[IV]], [[OFFSET]]
; CHECK-NEXT: [[ARR_IDX:%.*]] = getelementptr inbounds float, float* [[A]], i64 [[IND_SUM]]
; CHECK-NEXT: [[L1:%.*]] = load float, float* [[ARR_IDX]], align 4
; CHECK-NEXT: [[IND_SUM2:%.*]] = add i64 [[IV]], [[OFFSET2]]
; CHECK-NEXT: [[ARR_IDX2:%.*]] = getelementptr inbounds float, float* [[A]], i64 [[IND_SUM2]]
; CHECK-NEXT: [[L2:%.*]] = load float, float* [[ARR_IDX2]], align 4
; CHECK-NEXT: [[M:%.*]] = fmul fast float [[L2]], [[B]]
; CHECK-NEXT: [[AD:%.*]] = fadd fast float [[L1]], [[M]]
; CHECK-NEXT: store float [[AD]], float* [[ARR_IDX]], align 4
; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i64 [[IV_NEXT]], [[N]]
; CHECK-NEXT: br i1 [[EXITCOND]], label [[LOOPEXIT]], label [[FOR_BODY]], !llvm.loop [[LOOP21:![0-9]+]]
; CHECK: loopexit:
; CHECK-NEXT: ret void
;
; FORCED_OPTSIZE-LABEL: @test_runtime_check(
; FORCED_OPTSIZE-NEXT: entry:
; FORCED_OPTSIZE-NEXT: br label [[FOR_BODY:%.*]]
; FORCED_OPTSIZE: for.body:
; FORCED_OPTSIZE-NEXT: [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
; FORCED_OPTSIZE-NEXT: [[IND_SUM:%.*]] = add i64 [[IV]], [[OFFSET:%.*]]
; FORCED_OPTSIZE-NEXT: [[ARR_IDX:%.*]] = getelementptr inbounds float, float* [[A:%.*]], i64 [[IND_SUM]]
; FORCED_OPTSIZE-NEXT: [[L1:%.*]] = load float, float* [[ARR_IDX]], align 4
; FORCED_OPTSIZE-NEXT: [[IND_SUM2:%.*]] = add i64 [[IV]], [[OFFSET2:%.*]]
; FORCED_OPTSIZE-NEXT: [[ARR_IDX2:%.*]] = getelementptr inbounds float, float* [[A]], i64 [[IND_SUM2]]
; FORCED_OPTSIZE-NEXT: [[L2:%.*]] = load float, float* [[ARR_IDX2]], align 4
; FORCED_OPTSIZE-NEXT: [[M:%.*]] = fmul fast float [[B:%.*]], [[L2]]
; FORCED_OPTSIZE-NEXT: [[AD:%.*]] = fadd fast float [[L1]], [[M]]
; FORCED_OPTSIZE-NEXT: store float [[AD]], float* [[ARR_IDX]], align 4
; FORCED_OPTSIZE-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; FORCED_OPTSIZE-NEXT: [[EXITCOND:%.*]] = icmp eq i64 [[IV_NEXT]], [[N:%.*]]
; FORCED_OPTSIZE-NEXT: br i1 [[EXITCOND]], label [[LOOPEXIT:%.*]], label [[FOR_BODY]]
; FORCED_OPTSIZE: loopexit:
; FORCED_OPTSIZE-NEXT: ret void
;
entry:
br label %for.body
for.body:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
%ind.sum = add i64 %iv, %offset
%arr.idx = getelementptr inbounds float, float* %a, i64 %ind.sum
%l1 = load float, float* %arr.idx, align 4
%ind.sum2 = add i64 %iv, %offset2
%arr.idx2 = getelementptr inbounds float, float* %a, i64 %ind.sum2
%l2 = load float, float* %arr.idx2, align 4
%m = fmul fast float %b, %l2
%ad = fadd fast float %l1, %m
store float %ad, float* %arr.idx, align 4
%iv.next = add nuw nsw i64 %iv, 1
%exitcond = icmp eq i64 %iv.next, %n
br i1 %exitcond, label %loopexit, label %for.body
loopexit:
ret void
}
; Check we do not generate runtime checks if we found a known dependence preventing
; vectorization. In this case, it is a read of c[i-1] followed by a write of c[i].
; The runtime checks would always fail.
; void test_runtime_check2(float *a, float b, unsigned offset, unsigned offset2, unsigned n, float *c) {
; for (unsigned i = 1; i < n; i++) {
; a[i+o1] += a[i+o2] + b;
; c[i] = c[i-1] + b;
; }
; }
define void @test_runtime_check2(float* %a, float %b, i64 %offset, i64 %offset2, i64 %n, float* %c) {
; CHECK-LABEL: @test_runtime_check2(
; CHECK-NEXT: entry:
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.body:
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
; CHECK-NEXT: [[IND_SUM:%.*]] = add i64 [[IV]], [[OFFSET:%.*]]
; CHECK-NEXT: [[ARR_IDX:%.*]] = getelementptr inbounds float, float* [[A:%.*]], i64 [[IND_SUM]]
; CHECK-NEXT: [[L1:%.*]] = load float, float* [[ARR_IDX]], align 4
; CHECK-NEXT: [[IND_SUM2:%.*]] = add i64 [[IV]], [[OFFSET2:%.*]]
; CHECK-NEXT: [[ARR_IDX2:%.*]] = getelementptr inbounds float, float* [[A]], i64 [[IND_SUM2]]
; CHECK-NEXT: [[L2:%.*]] = load float, float* [[ARR_IDX2]], align 4
; CHECK-NEXT: [[M:%.*]] = fmul fast float [[L2]], [[B:%.*]]
; CHECK-NEXT: [[AD:%.*]] = fadd fast float [[L1]], [[M]]
; CHECK-NEXT: store float [[AD]], float* [[ARR_IDX]], align 4
; CHECK-NEXT: [[C_IND:%.*]] = add nsw i64 [[IV]], -1
; CHECK-NEXT: [[C_IDX:%.*]] = getelementptr inbounds float, float* [[C:%.*]], i64 [[C_IND]]
; CHECK-NEXT: [[LC:%.*]] = load float, float* [[C_IDX]], align 4
; CHECK-NEXT: [[VC:%.*]] = fadd float [[LC]], 1.000000e+00
; CHECK-NEXT: [[C_IDX2:%.*]] = getelementptr inbounds float, float* [[C]], i64 [[IV]]
; CHECK-NEXT: store float [[VC]], float* [[C_IDX2]], align 4
; CHECK-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; CHECK-NEXT: [[EXITCOND:%.*]] = icmp eq i64 [[IV_NEXT]], [[N:%.*]]
; CHECK-NEXT: br i1 [[EXITCOND]], label [[LOOPEXIT:%.*]], label [[FOR_BODY]]
; CHECK: loopexit:
; CHECK-NEXT: ret void
;
; FORCED_OPTSIZE-LABEL: @test_runtime_check2(
; FORCED_OPTSIZE-NEXT: entry:
; FORCED_OPTSIZE-NEXT: br label [[FOR_BODY:%.*]]
; FORCED_OPTSIZE: for.body:
; FORCED_OPTSIZE-NEXT: [[IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
; FORCED_OPTSIZE-NEXT: [[IND_SUM:%.*]] = add i64 [[IV]], [[OFFSET:%.*]]
; FORCED_OPTSIZE-NEXT: [[ARR_IDX:%.*]] = getelementptr inbounds float, float* [[A:%.*]], i64 [[IND_SUM]]
; FORCED_OPTSIZE-NEXT: [[L1:%.*]] = load float, float* [[ARR_IDX]], align 4
; FORCED_OPTSIZE-NEXT: [[IND_SUM2:%.*]] = add i64 [[IV]], [[OFFSET2:%.*]]
; FORCED_OPTSIZE-NEXT: [[ARR_IDX2:%.*]] = getelementptr inbounds float, float* [[A]], i64 [[IND_SUM2]]
; FORCED_OPTSIZE-NEXT: [[L2:%.*]] = load float, float* [[ARR_IDX2]], align 4
; FORCED_OPTSIZE-NEXT: [[M:%.*]] = fmul fast float [[B:%.*]], [[L2]]
; FORCED_OPTSIZE-NEXT: [[AD:%.*]] = fadd fast float [[L1]], [[M]]
; FORCED_OPTSIZE-NEXT: store float [[AD]], float* [[ARR_IDX]], align 4
; FORCED_OPTSIZE-NEXT: [[C_IND:%.*]] = add i64 [[IV]], -1
; FORCED_OPTSIZE-NEXT: [[C_IDX:%.*]] = getelementptr inbounds float, float* [[C:%.*]], i64 [[C_IND]]
; FORCED_OPTSIZE-NEXT: [[LC:%.*]] = load float, float* [[C_IDX]], align 4
; FORCED_OPTSIZE-NEXT: [[VC:%.*]] = fadd float [[LC]], 1.000000e+00
; FORCED_OPTSIZE-NEXT: [[C_IDX2:%.*]] = getelementptr inbounds float, float* [[C]], i64 [[IV]]
; FORCED_OPTSIZE-NEXT: store float [[VC]], float* [[C_IDX2]], align 4
; FORCED_OPTSIZE-NEXT: [[IV_NEXT]] = add nuw nsw i64 [[IV]], 1
; FORCED_OPTSIZE-NEXT: [[EXITCOND:%.*]] = icmp eq i64 [[IV_NEXT]], [[N:%.*]]
; FORCED_OPTSIZE-NEXT: br i1 [[EXITCOND]], label [[LOOPEXIT:%.*]], label [[FOR_BODY]]
; FORCED_OPTSIZE: loopexit:
; FORCED_OPTSIZE-NEXT: ret void
;
entry:
br label %for.body
for.body:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
%ind.sum = add i64 %iv, %offset
%arr.idx = getelementptr inbounds float, float* %a, i64 %ind.sum
%l1 = load float, float* %arr.idx, align 4
%ind.sum2 = add i64 %iv, %offset2
%arr.idx2 = getelementptr inbounds float, float* %a, i64 %ind.sum2
%l2 = load float, float* %arr.idx2, align 4
%m = fmul fast float %b, %l2
%ad = fadd fast float %l1, %m
store float %ad, float* %arr.idx, align 4
%c.ind = add i64 %iv, -1
%c.idx = getelementptr inbounds float, float* %c, i64 %c.ind
%lc = load float, float* %c.idx, align 4
%vc = fadd float %lc, 1.0
%c.idx2 = getelementptr inbounds float, float* %c, i64 %iv
store float %vc, float* %c.idx2
%iv.next = add nuw nsw i64 %iv, 1
%exitcond = icmp eq i64 %iv.next, %n
br i1 %exitcond, label %loopexit, label %for.body
loopexit:
ret void
}
define dso_local void @forced_optsize(i64* noalias nocapture readonly %x_p, i64* noalias nocapture readonly %y_p, i64* noalias nocapture %z_p) minsize optsize {
; CHECK-LABEL: @forced_optsize(
; CHECK-NEXT: entry:
; CHECK-NEXT: br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[TMP0:%.*]] = getelementptr inbounds i64, i64* [[X_P:%.*]], i64 [[INDEX]]
; CHECK-NEXT: [[TMP1:%.*]] = bitcast i64* [[TMP0]] to <2 x i64>*
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <2 x i64>, <2 x i64>* [[TMP1]], align 8
; CHECK-NEXT: [[TMP2:%.*]] = getelementptr inbounds i64, i64* [[Y_P:%.*]], i64 [[INDEX]]
; CHECK-NEXT: [[TMP3:%.*]] = bitcast i64* [[TMP2]] to <2 x i64>*
; CHECK-NEXT: [[WIDE_LOAD1:%.*]] = load <2 x i64>, <2 x i64>* [[TMP3]], align 8
; CHECK-NEXT: [[TMP4:%.*]] = add nsw <2 x i64> [[WIDE_LOAD1]], [[WIDE_LOAD]]
; CHECK-NEXT: [[TMP5:%.*]] = getelementptr inbounds i64, i64* [[Z_P:%.*]], i64 [[INDEX]]
; CHECK-NEXT: [[TMP6:%.*]] = bitcast i64* [[TMP5]] to <2 x i64>*
; CHECK-NEXT: store <2 x i64> [[TMP4]], <2 x i64>* [[TMP6]], align 8
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
; CHECK-NEXT: [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], 128
; CHECK-NEXT: br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP22:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: br i1 true, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
; CHECK: scalar.ph:
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.cond.cleanup:
; CHECK-NEXT: ret void
; CHECK: for.body:
; CHECK-NEXT: br i1 poison, label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP23:![0-9]+]]
;
; FORCED_OPTSIZE-LABEL: @forced_optsize(
; FORCED_OPTSIZE-NEXT: entry:
; FORCED_OPTSIZE-NEXT: br label [[FOR_BODY:%.*]]
; FORCED_OPTSIZE: for.cond.cleanup:
; FORCED_OPTSIZE-NEXT: ret void
; FORCED_OPTSIZE: for.body:
; FORCED_OPTSIZE-NEXT: [[INDVARS_IV:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
; FORCED_OPTSIZE-NEXT: [[ARRAYIDX:%.*]] = getelementptr inbounds i64, i64* [[X_P:%.*]], i64 [[INDVARS_IV]]
; FORCED_OPTSIZE-NEXT: [[TMP0:%.*]] = load i64, i64* [[ARRAYIDX]], align 8
; FORCED_OPTSIZE-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i64, i64* [[Y_P:%.*]], i64 [[INDVARS_IV]]
; FORCED_OPTSIZE-NEXT: [[TMP1:%.*]] = load i64, i64* [[ARRAYIDX2]], align 8
; FORCED_OPTSIZE-NEXT: [[ADD:%.*]] = add nsw i64 [[TMP1]], [[TMP0]]
; FORCED_OPTSIZE-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i64, i64* [[Z_P:%.*]], i64 [[INDVARS_IV]]
; FORCED_OPTSIZE-NEXT: store i64 [[ADD]], i64* [[ARRAYIDX4]], align 8
; FORCED_OPTSIZE-NEXT: [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 1
; FORCED_OPTSIZE-NEXT: [[EXITCOND:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 128
; FORCED_OPTSIZE-NEXT: br i1 [[EXITCOND]], label [[FOR_COND_CLEANUP:%.*]], label [[FOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
;
entry:
br label %for.body
for.cond.cleanup:
ret void
for.body:
%indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
%arrayidx = getelementptr inbounds i64, i64* %x_p, i64 %indvars.iv
%0 = load i64, i64* %arrayidx, align 8
%arrayidx2 = getelementptr inbounds i64, i64* %y_p, i64 %indvars.iv
%1 = load i64, i64* %arrayidx2, align 8
%add = add nsw i64 %1, %0
%arrayidx4 = getelementptr inbounds i64, i64* %z_p, i64 %indvars.iv
store i64 %add, i64* %arrayidx4, align 8
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
%exitcond = icmp eq i64 %indvars.iv.next, 128
br i1 %exitcond, label %for.cond.cleanup, label %for.body, !llvm.loop !12
}
; CHECK: !9 = !DILocation(line: 101, column: 1, scope: !{{.*}})
!llvm.module.flags = !{!0, !1}
!llvm.dbg.cu = !{!9}
!0 = !{i32 2, !"Dwarf Version", i32 4}
!1 = !{i32 2, !"Debug Info Version", i32 3}
!2 = !{}
!3 = !DISubroutineType(types: !2)
!4 = !DIFile(filename: "test.cpp", directory: "/tmp")
!5 = distinct !DISubprogram(name: "foo", scope: !4, file: !4, line: 99, type: !3, isLocal: false, isDefinition: true, scopeLine: 100, flags: DIFlagPrototyped, isOptimized: false, unit: !9, retainedNodes: !2)
!6 = !DILocation(line: 100, column: 1, scope: !5)
!7 = !DILocation(line: 101, column: 1, scope: !5)
!8 = !DILocation(line: 102, column: 1, scope: !5)
!9 = distinct !DICompileUnit(language: DW_LANG_C99, producer: "clang",
file: !10,
isOptimized: true, flags: "-O2",
splitDebugFilename: "abc.debug", emissionKind: 2)
!10 = !DIFile(filename: "path/to/file", directory: "/path/to/dir")
!11 = !{i32 2, !"Debug Info Version", i32 3}
!12 = distinct !{!12, !13, !14}
!13 = !{!"llvm.loop.vectorize.width", i32 2}
!14 = !{!"llvm.loop.vectorize.enable", i1 true}
Reference in New Issue View Git Blame Copy Permalink