[InlineCost] Don't add the soft float function call cost for the fneg idiom, fsub...

Summary: Fneg can be implemented with an xor rather than a function call so we don't need to add the function call overhead. This was pointed out in D62699

Reviewers: efriedma, cameron.mcinally

Reviewed By: efriedma

Subscribers: javed.absar, eraman, hiraditya, haicheng, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D62747

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@362304 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Analysis/InlineCost.cpp b/lib/Analysis/InlineCost.cpp
index ced30d6e3b91cf5838d2a27a5514e6ad1daaecaa..a332a439007163b744f59714d0e3a6a771173d9d 100644 (file)
--- a/lib/Analysis/InlineCost.cpp
+++ b/lib/Analysis/InlineCost.cpp
@@ -35,6 +35,7 @@
 #include "llvm/IR/InstVisitor.h"
 #include "llvm/IR/IntrinsicInst.h"
 #include "llvm/IR/Operator.h"
+#include "llvm/IR/PatternMatch.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/raw_ostream.h"
 
@@ -1095,9 +1096,11 @@ bool CallAnalyzer::visitBinaryOperator(BinaryOperator &I) {
 
   // If the instruction is floating point, and the target says this operation
   // is expensive, this may eventually become a library call. Treat the cost
-  // as such.
+  // as such. Unless it's fneg which can be implemented with an xor.
+  using namespace llvm::PatternMatch;
   if (I.getType()->isFloatingPointTy() &&
-      TTI.getFPOpCost(I.getType()) == TargetTransformInfo::TCC_Expensive)
+      TTI.getFPOpCost(I.getType()) == TargetTransformInfo::TCC_Expensive &&
+      !match(&I, m_FNeg(m_Value())))
     addCost(InlineConstants::CallPenalty);
 
   return false;

diff --git a/test/Transforms/Inline/ARM/inline-fp.ll b/test/Transforms/Inline/ARM/inline-fp.ll
index fdc066c9ba685ea33739303998df0066c277540a..1d74dfd15b187a790bf10129e315e197bf73ca4d 100644 (file)
--- a/test/Transforms/Inline/ARM/inline-fp.ll
+++ b/test/Transforms/Inline/ARM/inline-fp.ll
@@ -12,6 +12,8 @@
 ; NOFP-DAG: double not inlined into test_double because too costly to inline (cost=125, threshold=75)
 ; NOFP-DAG: single_force_soft not inlined into test_single_force_soft because too costly to inline (cost=125, threshold=75)
 ; NOFP-DAG: single_force_soft not inlined into test_single_force_soft because too costly to inline (cost=125, threshold=75)
+; NOFP-DAG: single_force_soft_fneg not inlined into test_single_force_soft_fneg because too costly to inline (cost=100, threshold=75)
+; NOFP-DAG: single_force_soft_fneg not inlined into test_single_force_soft_fneg because too costly to inline (cost=100, threshold=75)
 
 ; FULLFP-DAG: single inlined into test_single with (cost=0, threshold=75)
 ; FULLFP-DAG: single inlined into test_single with (cost=-15000, threshold=75)
@@ -21,6 +23,8 @@
 ; FULLFP-DAG: double inlined into test_double with (cost=-15000, threshold=75)
 ; FULLFP-DAG: single_force_soft not inlined into test_single_force_soft because too costly to inline (cost=125, threshold=75)
 ; FULLFP-DAG: single_force_soft not inlined into test_single_force_soft because too costly to inline (cost=125, threshold=75)
+; FULLFP-DAG: single_force_soft_fneg not inlined into test_single_force_soft_fneg because too costly to inline (cost=100, threshold=75)
+; FULLFP-DAG: single_force_soft_fneg not inlined into test_single_force_soft_fneg because too costly to inline (cost=100, threshold=75)
 
 ; SINGLEFP-DAG: single inlined into test_single with (cost=0, threshold=75)
 ; SINGLEFP-DAG: single inlined into test_single with (cost=-15000, threshold=75)
@@ -30,6 +34,8 @@
 ; SINGLEFP-DAG: double not inlined into test_double because too costly to inline (cost=125, threshold=75)
 ; SINGLEFP-DAG: single_force_soft not inlined into test_single_force_soft because too costly to inline (cost=125, threshold=75)
 ; SINGLEFP-DAG: single_force_soft not inlined into test_single_force_soft because too costly to inline (cost=125, threshold=75)
+; SINGLEFP-DAG: single_force_soft_fneg not inlined into test_single_force_soft_fneg because too costly to inline (cost=100, threshold=75)
+; SINGLEFP-DAG: single_force_soft_fneg not inlined into test_single_force_soft_fneg because too costly to inline (cost=100, threshold=75)
 
 define i32 @test_single(i32 %a, i8 %b, i32 %c, i8 %d) #0 {
   %call = call float @single(i32 %a, i8 zeroext %b)
@@ -55,6 +61,12 @@ define i32 @test_single_force_soft(i32 %a, i8 %b, i32 %c, i8 %d) #1 {
   ret i32 0
 }
 
+define i32 @test_single_force_soft_fneg(i32 %a, i8 %b, i32 %c, i8 %d) #1 {
+  %call = call float @single_force_soft_fneg(i32 %a, i8 zeroext %b) #1
+  %call2 = call float @single_force_soft_fneg(i32 %c, i8 zeroext %d) #1
+  ret i32 0
+}
+
 define internal float @single(i32 %response, i8 zeroext %value1) #0 {
 entry:
   %conv = zext i8 %value1 to i32
@@ -106,6 +118,19 @@ entry:
   ret float %div
 }
 
+define internal float @single_force_soft_fneg(i32 %response, i8 zeroext %value1) #1 {
+entry:
+  %conv = zext i8 %value1 to i32
+  %sub = add nsw i32 %conv, -1
+  %conv1 = sitofp i32 %sub to float
+  %0 = tail call float @llvm.pow.f32(float 0x3FF028F5C0000000, float %conv1)
+  %mul = fsub float -0.0, %0
+  %conv2 = sitofp i32 %response to float
+  %sub3 = fsub float %conv2, %mul
+  %div = fdiv float %sub3, %mul
+  ret float %div
+}
+
 declare float @llvm.pow.f32(float, float) optsize minsize
 declare double @llvm.pow.f64(double, double) optsize minsize