[llvm] r265493 - [SLPVectorizer] Vectorize libcalls of sqrt

[David Majnemer via llvm-commits]

Author: majnemer
Date: Tue Apr  5 19:14:59 2016
New Revision: 265493

URL: http://llvm.org/viewvc/llvm-project?rev=265493&view=rev
Log:
[SLPVectorizer] Vectorize libcalls of sqrt

We didn't realize that we could transform the libcall into a vectorized
intrinsic.

Modified:
    llvm/trunk/lib/Analysis/VectorUtils.cpp
    llvm/trunk/test/Transforms/LoopVectorize/X86/veclib-calls.ll
    llvm/trunk/test/Transforms/SLPVectorizer/X86/call.ll

Modified: llvm/trunk/lib/Analysis/VectorUtils.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/Analysis/VectorUtils.cpp?rev=265493&r1=265492&r2=265493&view=diff
==============================================================================
--- llvm/trunk/lib/Analysis/VectorUtils.cpp (original)
+++ llvm/trunk/lib/Analysis/VectorUtils.cpp Tue Apr  5 19:14:59 2016
@@ -220,6 +220,10 @@ Intrinsic::ID llvm::getIntrinsicIDForCal
   case LibFunc::powf:
   case LibFunc::powl:
     return checkBinaryFloatSignature(*CI, Intrinsic::pow);
+  case LibFunc::sqrt:
+  case LibFunc::sqrtf:
+  case LibFunc::sqrtl:
+    return checkUnaryFloatSignature(*CI, Intrinsic::sqrt);
   }
 
   return Intrinsic::not_intrinsic;

Modified: llvm/trunk/test/Transforms/LoopVectorize/X86/veclib-calls.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/LoopVectorize/X86/veclib-calls.ll?rev=265493&r1=265492&r2=265493&view=diff
==============================================================================
--- llvm/trunk/test/Transforms/LoopVectorize/X86/veclib-calls.ll (original)
+++ llvm/trunk/test/Transforms/LoopVectorize/X86/veclib-calls.ll Tue Apr  5 19:14:59 2016
@@ -3,31 +3,6 @@
 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"
 target triple = "x86_64-unknown-linux-gnu"
 
-;CHECK-LABEL: @sqrt_f32(
-;CHECK: vsqrtf{{.*}}<4 x float>
-;CHECK: ret void
-declare float @sqrtf(float) nounwind readnone
-define void @sqrt_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
-entry:
-  %cmp6 = icmp sgt i32 %n, 0
-  br i1 %cmp6, label %for.body, label %for.end
-
-for.body:                                         ; preds = %entry, %for.body
-  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
-  %0 = load float, float* %arrayidx, align 4
-  %call = tail call float @sqrtf(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
-  store float %call, float* %arrayidx2, align 4
-  %indvars.iv.next = add i64 %indvars.iv, 1
-  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
-  %exitcond = icmp eq i32 %lftr.wideiv, %n
-  br i1 %exitcond, label %for.end, label %for.body
-
-for.end:                                          ; preds = %for.body, %entry
-  ret void
-}
-
 ;CHECK-LABEL: @exp_f32(
 ;CHECK: vexpf{{.*}}<4 x float>
 ;CHECK: ret void
@@ -160,6 +135,7 @@ for.end:
 ;CHECK-LABEL: @sqrt_f32_nobuiltin(
 ;CHECK-NOT: vsqrtf{{.*}}<4 x float>
 ;CHECK: ret void
+declare float @sqrtf(float) nounwind readnone
 define void @sqrt_f32_nobuiltin(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
 entry:
   %cmp6 = icmp sgt i32 %n, 0

Modified: llvm/trunk/test/Transforms/SLPVectorizer/X86/call.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Transforms/SLPVectorizer/X86/call.ll?rev=265493&r1=265492&r2=265493&view=diff
==============================================================================
--- llvm/trunk/test/Transforms/SLPVectorizer/X86/call.ll (original)
+++ llvm/trunk/test/Transforms/SLPVectorizer/X86/call.ll Tue Apr  5 19:14:59 2016
@@ -7,6 +7,7 @@ declare double @sin(double)
 declare double @cos(double)
 declare double @pow(double, double)
 declare double @exp2(double)
+declare double @sqrt(double)
 declare i64 @round(i64)
 
 
@@ -92,6 +93,28 @@ entry:
   store double %call, double* %c, align 8
   %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %call5, double* %arrayidx5, align 8
+  ret void
+}
+
+
+; CHECK: sqrt_libm
+; CHECK: call <2 x double> @llvm.sqrt.v2f64
+; CHECK: ret void
+define void @sqrt_libm(double* %a, double* %b, double* %c) {
+entry:
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
+  %mul = fmul double %i0, %i1
+  %call = tail call double @sqrt(double %mul) nounwind readnone
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
+  %mul5 = fmul double %i3, %i4
+  %call5 = tail call double @sqrt(double %mul5) nounwind readnone
+  store double %call, double* %c, align 8
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
+  store double %call5, double* %arrayidx5, align 8
   ret void
 }