From ff98682c9c60b57e09937abfd3436520232a9bdc Mon Sep 17 00:00:00 2001 From: Sanjay Patel Date: Wed, 11 Apr 2018 15:57:18 +0000 Subject: [InstCombine] limit X - (cast(-Y) --> X + cast(Y) with hasOneUse() llvm-svn: 329821 --- llvm/test/Transforms/InstCombine/fsub.ll | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) (limited to 'llvm/test/Transforms') diff --git a/llvm/test/Transforms/InstCombine/fsub.ll b/llvm/test/Transforms/InstCombine/fsub.ll index 78aebc022b2..9ce73084f79 100644 --- a/llvm/test/Transforms/InstCombine/fsub.ll +++ b/llvm/test/Transforms/InstCombine/fsub.ll @@ -189,14 +189,13 @@ define double @neg_ext_op1_fast(float %a, double %b) { ret double %t3 } -; FIXME: Extra use should prevent the transform. +; Extra use should prevent the transform. define float @neg_ext_op1_extra_use(half %a, float %b) { ; CHECK-LABEL: @neg_ext_op1_extra_use( ; CHECK-NEXT: [[T1:%.*]] = fsub half 0xH8000, [[A:%.*]] ; CHECK-NEXT: [[T2:%.*]] = fpext half [[T1]] to float -; CHECK-NEXT: [[TMP1:%.*]] = fpext half [[A]] to float -; CHECK-NEXT: [[T3:%.*]] = fadd float [[TMP1]], [[B:%.*]] +; CHECK-NEXT: [[T3:%.*]] = fsub float [[B:%.*]], [[T2]] ; CHECK-NEXT: call void @use(float [[T2]]) ; CHECK-NEXT: ret float [[T3]] ; @@ -207,8 +206,8 @@ define float @neg_ext_op1_extra_use(half %a, float %b) { ret float %t3 } -; One-use fptrunc is always hoisted above fneg, so the corresponding -; multi-use bug for fptrunc isn't visible with a fold starting from +; One-use fptrunc is always hoisted above fneg, so the corresponding +; multi-use bug for fptrunc isn't visible with a fold starting from ; the last fsub. define float @neg_trunc_op1_extra_use(double %a, float %b) { @@ -226,15 +225,13 @@ define float @neg_trunc_op1_extra_use(double %a, float %b) { ret float %t3 } -; FIXME: But the bug is visible when both preceding values have other uses. ; Extra uses should prevent the transform. define float @neg_trunc_op1_extra_uses(double %a, float %b) { ; CHECK-LABEL: @neg_trunc_op1_extra_uses( ; CHECK-NEXT: [[T1:%.*]] = fsub double -0.000000e+00, [[A:%.*]] ; CHECK-NEXT: [[T2:%.*]] = fptrunc double [[T1]] to float -; CHECK-NEXT: [[TMP1:%.*]] = fptrunc double [[A]] to float -; CHECK-NEXT: [[T3:%.*]] = fadd float [[TMP1]], [[B:%.*]] +; CHECK-NEXT: [[T3:%.*]] = fsub float [[B:%.*]], [[T2]] ; CHECK-NEXT: call void @use2(float [[T2]], double [[T1]]) ; CHECK-NEXT: ret float [[T3]] ; -- cgit v1.2.3