Revert "InstCombine: Fold ldexp with constant exponent to fmul" #171895

arsenm · 2025-12-11T20:40:33Z

Reverts #171731

Fails on a libc test

…)" This reverts commit 5eb2ec2.

llvmbot · 2025-12-11T20:41:06Z

@llvm/pr-subscribers-llvm-transforms

Author: Matt Arsenault (arsenm)

Changes

Reverts llvm/llvm-project#171731

Fails on a libc test

Full diff: https://github.com/llvm/llvm-project/pull/171895.diff

3 Files Affected:

(modified) llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp (-13)
(modified) llvm/test/Transforms/InstCombine/fold-select-fmul-if-zero.ll (+8-2)
(modified) llvm/test/Transforms/InstCombine/ldexp.ll (+26-36)

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
index b498bafb3caaa..85602a5a7575a 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
@@ -3089,19 +3089,6 @@ Instruction *InstCombinerImpl::visitCallInst(CallInst &CI) {
     // exponent. Also could broaden sign check to cover == 0 case.
     Value *Src = II->getArgOperand(0);
     Value *Exp = II->getArgOperand(1);
-
-    uint64_t ConstExp;
-    if (match(Exp, m_ConstantInt(ConstExp))) {
-      // ldexp(x, K) -> fmul x, 2^K
-      const fltSemantics &FPTy =
-          Src->getType()->getScalarType()->getFltSemantics();
-      Constant *FPConst =
-          ConstantFP::get(Src->getType(), scalbn(APFloat::getOne(FPTy),
-                                                 static_cast<int>(ConstExp),
-                                                 APFloat::rmNearestTiesToEven));
-      return BinaryOperator::CreateFMulFMF(Src, FPConst, II);
-    }
-
     Value *InnerSrc;
     Value *InnerExp;
     if (match(Src, m_OneUse(m_Intrinsic<Intrinsic::ldexp>(
diff --git a/llvm/test/Transforms/InstCombine/fold-select-fmul-if-zero.ll b/llvm/test/Transforms/InstCombine/fold-select-fmul-if-zero.ll
index 4495b0f26042b..1ba7005f99e3d 100644
--- a/llvm/test/Transforms/InstCombine/fold-select-fmul-if-zero.ll
+++ b/llvm/test/Transforms/InstCombine/fold-select-fmul-if-zero.ll
@@ -49,7 +49,10 @@ define float @fmul_by_32_if_0_oeq_zero_f32(float %x) {
 
 define float @ldexp_by_5_if_0_oeq_zero_f32(float %x) {
 ; CHECK-LABEL: @ldexp_by_5_if_0_oeq_zero_f32(
-; CHECK-NEXT:    ret float [[X:%.*]]
+; CHECK-NEXT:    [[X_IS_ZERO:%.*]] = fcmp oeq float [[X:%.*]], 0.000000e+00
+; CHECK-NEXT:    [[SCALED_X:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 5)
+; CHECK-NEXT:    [[SCALED_IF_DENORMAL:%.*]] = select i1 [[X_IS_ZERO]], float [[SCALED_X]], float [[X]]
+; CHECK-NEXT:    ret float [[SCALED_IF_DENORMAL]]
 ;
   %x.is.zero = fcmp oeq float %x, 0.0
   %scaled.x = call float @llvm.ldexp.f32.i32(float %x, i32 5)
@@ -59,7 +62,10 @@ define float @ldexp_by_5_if_0_oeq_zero_f32(float %x) {
 
 define <2 x float> @ldexp_by_5_if_0_oeq_zero_v2f32(<2 x float> %x) {
 ; CHECK-LABEL: @ldexp_by_5_if_0_oeq_zero_v2f32(
-; CHECK-NEXT:    ret <2 x float> [[SCALED_IF_DENORMAL:%.*]]
+; CHECK-NEXT:    [[X_IS_ZERO:%.*]] = fcmp oeq <2 x float> [[X:%.*]], zeroinitializer
+; CHECK-NEXT:    [[SCALED_X:%.*]] = call <2 x float> @llvm.ldexp.v2f32.v2i32(<2 x float> [[X]], <2 x i32> splat (i32 5))
+; CHECK-NEXT:    [[SCALED_IF_DENORMAL:%.*]] = select <2 x i1> [[X_IS_ZERO]], <2 x float> [[SCALED_X]], <2 x float> [[X]]
+; CHECK-NEXT:    ret <2 x float> [[SCALED_IF_DENORMAL]]
 ;
   %x.is.zero = fcmp oeq <2 x float> %x, zeroinitializer
   %scaled.x = call <2 x float> @llvm.ldexp.v2f32.v2i32(<2 x float> %x, <2 x i32> <i32 5, i32 5>)
diff --git a/llvm/test/Transforms/InstCombine/ldexp.ll b/llvm/test/Transforms/InstCombine/ldexp.ll
index 43e80b1a1e588..8908d476b4a2c 100644
--- a/llvm/test/Transforms/InstCombine/ldexp.ll
+++ b/llvm/test/Transforms/InstCombine/ldexp.ll
@@ -444,7 +444,7 @@ define float @ldexp_ldexp_different_exp_type(float %x, i32 %a, i64 %b) {
 define float @ldexp_ldexp_constants(float %x) {
 ; CHECK-LABEL: define float @ldexp_ldexp_constants
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP1:%.*]] = fmul reassoc float [[X]], 0x41F0000000000000
+; CHECK-NEXT:    [[LDEXP1:%.*]] = call reassoc float @llvm.ldexp.f32.i32(float [[X]], i32 32)
 ; CHECK-NEXT:    ret float [[LDEXP1]]
 ;
   %ldexp0 = call reassoc float @llvm.ldexp.f32.i32(float %x, i32 8)
@@ -455,7 +455,7 @@ define float @ldexp_ldexp_constants(float %x) {
 define float @ldexp_ldexp_constants_nsz(float %x) {
 ; CHECK-LABEL: define float @ldexp_ldexp_constants_nsz
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP1:%.*]] = fmul reassoc nsz float [[X]], 0x41F0000000000000
+; CHECK-NEXT:    [[LDEXP1:%.*]] = call reassoc nsz float @llvm.ldexp.f32.i32(float [[X]], i32 32)
 ; CHECK-NEXT:    ret float [[LDEXP1]]
 ;
   %ldexp0 = call reassoc nsz float @llvm.ldexp.f32.i32(float %x, i32 8)
@@ -466,7 +466,7 @@ define float @ldexp_ldexp_constants_nsz(float %x) {
 define float @ldexp_ldexp_constants_nsz0(float %x) {
 ; CHECK-LABEL: define float @ldexp_ldexp_constants_nsz0
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP1:%.*]] = fmul reassoc float [[X]], 0x41F0000000000000
+; CHECK-NEXT:    [[LDEXP1:%.*]] = call reassoc nsz float @llvm.ldexp.f32.i32(float [[X]], i32 32)
 ; CHECK-NEXT:    ret float [[LDEXP1]]
 ;
   %ldexp0 = call reassoc nsz float @llvm.ldexp.f32.i32(float %x, i32 8)
@@ -477,7 +477,7 @@ define float @ldexp_ldexp_constants_nsz0(float %x) {
 define float @ldexp_ldexp_constants_nsz1(float %x) {
 ; CHECK-LABEL: define float @ldexp_ldexp_constants_nsz1
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP1:%.*]] = fmul reassoc nsz float [[X]], 0x41F0000000000000
+; CHECK-NEXT:    [[LDEXP1:%.*]] = call reassoc nsz float @llvm.ldexp.f32.i32(float [[X]], i32 32)
 ; CHECK-NEXT:    ret float [[LDEXP1]]
 ;
   %ldexp0 = call reassoc float @llvm.ldexp.f32.i32(float %x, i32 8)
@@ -641,7 +641,7 @@ define float @ldexp_ldexp_0(float %x, i32 %y) {
 define float @ldexp_neg150(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg150
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0.000000e+00
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -150)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -150)
@@ -651,7 +651,7 @@ define float @ldexp_neg150(float %x) {
 define float @ldexp_neg149(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg149
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x36A0000000000000
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -149)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -149)
@@ -661,7 +661,7 @@ define float @ldexp_neg149(float %x) {
 define float @ldexp_neg148(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg148
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x36B0000000000000
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -148)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -148)
@@ -671,7 +671,7 @@ define float @ldexp_neg148(float %x) {
 define float @ldexp_neg127(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg127
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x3800000000000000
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -127)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -127)
@@ -681,7 +681,7 @@ define float @ldexp_neg127(float %x) {
 define float @ldexp_neg126(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg126
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x3810000000000000
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -126)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -126)
@@ -691,7 +691,7 @@ define float @ldexp_neg126(float %x) {
 define float @ldexp_neg125(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg125
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x3820000000000000
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -125)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -125)
@@ -701,7 +701,7 @@ define float @ldexp_neg125(float %x) {
 define float @ldexp_neg16(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg16
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x3EF0000000000000
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -16)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -16)
@@ -711,7 +711,7 @@ define float @ldexp_neg16(float %x) {
 define float @ldexp_neg8(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg8
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 3.906250e-03
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -8)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -8)
@@ -721,7 +721,7 @@ define float @ldexp_neg8(float %x) {
 define float @ldexp_neg4(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg4
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 6.250000e-02
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -4)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -4)
@@ -731,7 +731,7 @@ define float @ldexp_neg4(float %x) {
 define float @ldexp_neg2(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg2
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 2.500000e-01
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -2)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -2)
@@ -741,7 +741,7 @@ define float @ldexp_neg2(float %x) {
 define float @ldexp_neg1(float %x) {
 ; CHECK-LABEL: define float @ldexp_neg1
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 5.000000e-01
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 -1)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 -1)
@@ -760,7 +760,7 @@ define float @ldexp_0(float %x) {
 define float @ldexp_1(float %x) {
 ; CHECK-LABEL: define float @ldexp_1
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 2.000000e+00
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 1)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 1)
@@ -770,7 +770,7 @@ define float @ldexp_1(float %x) {
 define float @ldexp_2(float %x) {
 ; CHECK-LABEL: define float @ldexp_2
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 4.000000e+00
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 2)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 2)
@@ -780,7 +780,7 @@ define float @ldexp_2(float %x) {
 define float @ldexp_3(float %x) {
 ; CHECK-LABEL: define float @ldexp_3
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 8.000000e+00
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 3)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 3)
@@ -790,7 +790,7 @@ define float @ldexp_3(float %x) {
 define float @ldexp_10(float %x) {
 ; CHECK-LABEL: define float @ldexp_10
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 1.024000e+03
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 10)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 10)
@@ -800,7 +800,7 @@ define float @ldexp_10(float %x) {
 define float @ldexp_125(float %x) {
 ; CHECK-LABEL: define float @ldexp_125
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x47C0000000000000
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 125)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 125)
@@ -810,7 +810,7 @@ define float @ldexp_125(float %x) {
 define float @ldexp_126(float %x) {
 ; CHECK-LABEL: define float @ldexp_126
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x47D0000000000000
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 126)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 126)
@@ -820,27 +820,17 @@ define float @ldexp_126(float %x) {
 define float @ldexp_127(float %x) {
 ; CHECK-LABEL: define float @ldexp_127
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x47E0000000000000
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 127)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 127)
   ret float %ldexp
 }
 
-define float @ldexp_128(float %x) {
-; CHECK-LABEL: define float @ldexp_128
-; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 0x7FF0000000000000
-; CHECK-NEXT:    ret float [[LDEXP]]
-;
-  %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 128)
-  ret float %ldexp
-}
-
 define <2 x float> @ldexp_3_vector(<2 x float> %x) {
 ; CHECK-LABEL: define <2 x float> @ldexp_3_vector
 ; CHECK-SAME: (<2 x float> [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul <2 x float> [[X]], splat (float 8.000000e+00)
+; CHECK-NEXT:    [[LDEXP:%.*]] = call <2 x float> @llvm.ldexp.v2f32.v2i32(<2 x float> [[X]], <2 x i32> splat (i32 3))
 ; CHECK-NEXT:    ret <2 x float> [[LDEXP]]
 ;
   %ldexp = call <2 x float> @llvm.ldexp.v2f32.v2i32(<2 x float> %x, <2 x i32> <i32 3, i32 3>)
@@ -870,7 +860,7 @@ define <2 x float> @ldexp_3_4_vector(<2 x float> %x) {
 define float @ldexp_2_flags(float %x) {
 ; CHECK-LABEL: define float @ldexp_2_flags
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul nsz contract float [[X]], 4.000000e+00
+; CHECK-NEXT:    [[LDEXP:%.*]] = call nsz contract float @llvm.ldexp.f32.i32(float [[X]], i32 2)
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call contract nsz float @llvm.ldexp.f32.i32(float %x, i32 2)
@@ -880,7 +870,7 @@ define float @ldexp_2_flags(float %x) {
 define float @ldexp_metadata(float %x) {
 ; CHECK-LABEL: define float @ldexp_metadata
 ; CHECK-SAME: (float [[X:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul float [[X]], 4.000000e+00
+; CHECK-NEXT:    [[LDEXP:%.*]] = call float @llvm.ldexp.f32.i32(float [[X]], i32 2), !foo [[META2:![0-9]+]]
 ; CHECK-NEXT:    ret float [[LDEXP]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 2), !foo !2
@@ -890,7 +880,7 @@ define float @ldexp_metadata(float %x) {
 define float @ldexp_8_contractable(float %x, float %y) {
 ; CHECK-LABEL: define float @ldexp_8_contractable
 ; CHECK-SAME: (float [[X:%.*]], float [[Y:%.*]]) {
-; CHECK-NEXT:    [[LDEXP:%.*]] = fmul contract float [[X]], 4.000000e+00
+; CHECK-NEXT:    [[LDEXP:%.*]] = call contract float @llvm.ldexp.f32.i32(float [[X]], i32 2)
 ; CHECK-NEXT:    [[FADD:%.*]] = fadd contract float [[LDEXP]], [[Y]]
 ; CHECK-NEXT:    ret float [[FADD]]
 ;

llvm-ci · 2025-12-12T00:29:37Z

LLVM Buildbot has detected a new failure on builder ppc64le-mlir-rhel-clang running on ppc64le-mlir-rhel-test while building llvm at step 6 "test-build-check-mlir-build-only-check-mlir".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/129/builds/34858

Here is the relevant piece of the build log for the reference

Step 6 (test-build-check-mlir-build-only-check-mlir) failure: 1200 seconds without output running [b'ninja', b'check-mlir'], attempting to kill
...
PASS: MLIR :: Pass/ir-printing.mlir (3744 of 3755)
PASS: MLIR :: mlir-tblgen/op-error.td (3745 of 3755)
PASS: MLIR-Unit :: IR/./MLIRIRTests/0/131 (3746 of 3755)
PASS: MLIR :: mlir-runner/simple.mlir (3747 of 3755)
PASS: MLIR-Unit :: IR/./MLIRIRTests/38/131 (3748 of 3755)
PASS: MLIR-Unit :: IR/./MLIRIRTests/39/131 (3749 of 3755)
PASS: MLIR-Unit :: Interfaces/./MLIRInterfacesTests/12/22 (3750 of 3755)
PASS: MLIR-Unit :: Pass/./MLIRPassTests/11/14 (3751 of 3755)
PASS: MLIR-Unit :: Interfaces/./MLIRInterfacesTests/13/22 (3752 of 3755)
PASS: MLIR :: mlir-reduce/dce-test.mlir (3753 of 3755)
command timed out: 1200 seconds without output running [b'ninja', b'check-mlir'], attempting to kill
process killed by signal 9
program finished with exit code -1
elapsedTime=2596.023720

…1895) This reverts commit 757c5b3. Reapply with the transform skipped if the scaling overflows or underflows.

…1895) (#171977)

…m#171895) (llvm#171977)

Revert "InstCombine: Fold ldexp with constant exponent to fmul (#171731…

8eafbbd

…)" This reverts commit 5eb2ec2.

arsenm requested a review from nikic as a code owner December 11, 2025 20:40

arsenm enabled auto-merge (squash) December 11, 2025 20:40

llvmbot added llvm:instcombine Covers the InstCombine, InstSimplify and AggressiveInstCombine passes llvm:transforms labels Dec 11, 2025

arsenm merged commit 757c5b3 into main Dec 11, 2025
12 of 13 checks passed

arsenm deleted the revert-171731-users/arsenm/instcombine-fold-ldexp-constant-to-fmul-pow2 branch December 11, 2025 21:13

arsenm added a commit that referenced this pull request Dec 12, 2025

Reapply "InstCombine: Fold ldexp with constant exponent to fmul" (#17…

81e746e

…1895) This reverts commit 757c5b3. Reapply with the transform skipped if the scaling overflows or underflows.

arsenm mentioned this pull request Dec 12, 2025

Reapply "InstCombine: Fold ldexp with constant exponent to fmul" (#171895) #171977

Merged

arsenm added a commit that referenced this pull request Dec 12, 2025

Reapply "InstCombine: Fold ldexp with constant exponent to fmul" (#17…

6e47d4e

…1895) (#171977)

anonymouspc pushed a commit to anonymouspc/llvm that referenced this pull request Dec 15, 2025

Reapply "InstCombine: Fold ldexp with constant exponent to fmul" (llv…

b43b185

…m#171895) (llvm#171977)

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Revert "InstCombine: Fold ldexp with constant exponent to fmul" #171895

Revert "InstCombine: Fold ldexp with constant exponent to fmul" #171895

Uh oh!

arsenm commented Dec 11, 2025

Uh oh!

llvmbot commented Dec 11, 2025

Uh oh!

Uh oh!

llvm-ci commented Dec 12, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Revert "InstCombine: Fold ldexp with constant exponent to fmul" #171895

Revert "InstCombine: Fold ldexp with constant exponent to fmul" #171895

Uh oh!

Conversation

arsenm commented Dec 11, 2025

Uh oh!

llvmbot commented Dec 11, 2025

Uh oh!

Uh oh!

llvm-ci commented Dec 12, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants