[llvm] r256679 - [LibCallSimplifier] propagate FMF when shrinking unary calls
Sanjay Patel via llvm-commits
llvm-commits at lists.llvm.org
Thu Dec 31 13:52:32 PST 2015
Author: spatel
Date: Thu Dec 31 15:52:31 2015
New Revision: 256679
URL: http://llvm.org/viewvc/llvm-project?rev=256679&view=rev
Log:
[LibCallSimplifier] propagate FMF when shrinking unary calls
Modified:
llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp
llvm/trunk/test/Transforms/InstCombine/double-float-shrink-1.ll
Modified: llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp?rev=256679&r1=256678&r2=256679&view=diff
==============================================================================
--- llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp (original)
+++ llvm/trunk/lib/Transforms/Utils/SimplifyLibCalls.cpp Thu Dec 31 15:52:31 2015
@@ -994,6 +994,10 @@ Value *LibCallSimplifier::optimizeUnaryD
Value *V = valueHasFloatPrecision(CI->getArgOperand(0));
if (V == nullptr)
return nullptr;
+
+ // Propagate fast-math flags from the existing call to the new call.
+ IRBuilder<>::FastMathFlagGuard Guard(B);
+ B.SetFastMathFlags(CI->getFastMathFlags());
// floor((double)floatval) -> (double)floorf(floatval)
if (Callee->isIntrinsic()) {
Modified: llvm/trunk/test/Transforms/InstCombine/double-float-shrink-1.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/InstCombine/double-float-shrink-1.ll?rev=256679&r1=256678&r2=256679&view=diff
==============================================================================
--- llvm/trunk/test/Transforms/InstCombine/double-float-shrink-1.ll (original)
+++ llvm/trunk/test/Transforms/InstCombine/double-float-shrink-1.ll Thu Dec 31 15:52:31 2015
@@ -9,15 +9,13 @@ target triple = "x86_64-unknown-linux-gn
; the -enable-double-float-shrink option.
; PR17850: http://llvm.org/bugs/show_bug.cgi?id=17850
-; FIXME: For all tests where the call shrinks, the 'fast' attribute on the call inst should be propagated to the new call.
-
define float @acos_test1(float %f) {
%conv = fpext float %f to double
%call = call fast double @acos(double %conv)
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: acos_test1
-; CHECK: call float @acosf(float %f)
+; CHECK: call fast float @acosf(float %f)
}
define double @acos_test2(float %f) {
@@ -34,7 +32,7 @@ define float @acosh_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: acosh_test1
-; CHECK: call float @acoshf(float %f)
+; CHECK: call fast float @acoshf(float %f)
}
define double @acosh_test2(float %f) {
@@ -51,7 +49,7 @@ define float @asin_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: asin_test1
-; CHECK: call float @asinf(float %f)
+; CHECK: call fast float @asinf(float %f)
}
define double @asin_test2(float %f) {
@@ -68,7 +66,7 @@ define float @asinh_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: asinh_test1
-; CHECK: call float @asinhf(float %f)
+; CHECK: call fast float @asinhf(float %f)
}
define double @asinh_test2(float %f) {
@@ -85,7 +83,7 @@ define float @atan_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: atan_test1
-; CHECK: call float @atanf(float %f)
+; CHECK: call fast float @atanf(float %f)
}
define double @atan_test2(float %f) {
@@ -102,7 +100,7 @@ define float @atanh_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: atanh_test1
-; CHECK: call float @atanhf(float %f)
+; CHECK: call fast float @atanhf(float %f)
}
define double @atanh_test2(float %f) {
@@ -119,7 +117,7 @@ define float @cbrt_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: cbrt_test1
-; CHECK: call float @cbrtf(float %f)
+; CHECK: call fast float @cbrtf(float %f)
}
define double @cbrt_test2(float %f) {
@@ -136,7 +134,7 @@ define float @exp_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: exp_test1
-; CHECK: call float @expf(float %f)
+; CHECK: call fast float @expf(float %f)
}
define double @exp_test2(float %f) {
@@ -153,7 +151,7 @@ define float @expm1_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: expm1_test1
-; CHECK: call float @expm1f(float %f)
+; CHECK: call fast float @expm1f(float %f)
}
define double @expm1_test2(float %f) {
@@ -189,7 +187,7 @@ define float @log_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: log_test1
-; CHECK: call float @logf(float %f)
+; CHECK: call fast float @logf(float %f)
}
define double @log_test2(float %f) {
@@ -206,7 +204,7 @@ define float @log10_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: log10_test1
-; CHECK: call float @log10f(float %f)
+; CHECK: call fast float @log10f(float %f)
}
define double @log10_test2(float %f) {
@@ -223,7 +221,7 @@ define float @log1p_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: log1p_test1
-; CHECK: call float @log1pf(float %f)
+; CHECK: call fast float @log1pf(float %f)
}
define double @log1p_test2(float %f) {
@@ -240,7 +238,7 @@ define float @log2_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: log2_test1
-; CHECK: call float @log2f(float %f)
+; CHECK: call fast float @log2f(float %f)
}
define double @log2_test2(float %f) {
@@ -257,7 +255,7 @@ define float @logb_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: logb_test1
-; CHECK: call float @logbf(float %f)
+; CHECK: call fast float @logbf(float %f)
}
define double @logb_test2(float %f) {
@@ -274,7 +272,7 @@ define float @sin_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: sin_test1
-; CHECK: call float @sinf(float %f)
+; CHECK: call fast float @sinf(float %f)
}
define double @sin_test2(float %f) {
@@ -325,7 +323,7 @@ define float @tan_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: tan_test1
-; CHECK: call float @tanf(float %f)
+; CHECK: call fast float @tanf(float %f)
}
define double @tan_test2(float %f) {
@@ -341,7 +339,7 @@ define float @tanh_test1(float %f) {
%conv1 = fptrunc double %call to float
ret float %conv1
; CHECK-LABEL: tanh_test1
-; CHECK: call float @tanhf(float %f)
+; CHECK: call fast float @tanhf(float %f)
}
define double @tanh_test2(float %f) {
More information about the llvm-commits
mailing list