Fix ordering of operands on lowering of atomicrmw min/max nodes on ARM.

[oota-llvm.git] / test / CodeGen / X86 / break-sse-dep.ll

diff --git a/test/CodeGen/X86/break-sse-dep.ll b/test/CodeGen/X86/break-sse-dep.ll
index 00c943f78db42c085b7ca9ba3b5ff3578d1a4bf1..4d801891da5ccb7c3dcadce910ea6af4fb5c3349 100644 (file)
--- a/test/CodeGen/X86/break-sse-dep.ll
+++ b/test/CodeGen/X86/break-sse-dep.ll
@@ -1,28 +1,62 @@
-; RUN: llc < %s -march=x86-64 -mattr=+sse2,+break-sse-dep | FileCheck %s --check-prefix=YES
-; RUN: llc < %s -march=x86-64 -mattr=+sse2,-break-sse-dep | FileCheck %s --check-prefix=NO
+; RUN: llc < %s -mtriple=x86_64-linux -mattr=+sse2 -mcpu=nehalem | FileCheck %s
+; RUN: llc < %s -mtriple=x86_64-win32 -mattr=+sse2 -mcpu=nehalem | FileCheck %s
 
 define double @t1(float* nocapture %x) nounwind readonly ssp {
 entry:
-; YES: t1:
-; YES: movss (%rdi), %xmm0
-; YES; cvtss2sd %xmm0, %xmm0
+; CHECK: t1:
+; CHECK: movss ([[A0:%rdi|%rcx]]), %xmm0
+; CHECK: cvtss2sd %xmm0, %xmm0
 
-; NO: t1:
-; NO; cvtss2sd (%rdi), %xmm0
   %0 = load float* %x, align 4
   %1 = fpext float %0 to double
   ret double %1
 }
 
-define float @t2(double* nocapture %x) nounwind readonly ssp {
+define float @t2(double* nocapture %x) nounwind readonly ssp optsize {
 entry:
-; YES: t2:
-; YES: movsd (%rdi), %xmm0
-; YES; cvtsd2ss %xmm0, %xmm0
-
-; NO: t2:
-; NO; cvtsd2ss (%rdi), %xmm0
+; CHECK: t2:
+; CHECK: cvtsd2ss ([[A0]]), %xmm0
   %0 = load double* %x, align 8
   %1 = fptrunc double %0 to float
   ret float %1
 }
+
+define float @squirtf(float* %x) nounwind {
+entry:
+; CHECK: squirtf:
+; CHECK: movss ([[A0]]), %xmm0
+; CHECK: sqrtss %xmm0, %xmm0
+  %z = load float* %x
+  %t = call float @llvm.sqrt.f32(float %z)
+  ret float %t
+}
+
+define double @squirt(double* %x) nounwind {
+entry:
+; CHECK: squirt:
+; CHECK: sqrtsd ([[A0]]), %xmm0
+  %z = load double* %x
+  %t = call double @llvm.sqrt.f64(double %z)
+  ret double %t
+}
+
+define float @squirtf_size(float* %x) nounwind optsize {
+entry:
+; CHECK: squirtf_size:
+; CHECK: sqrtss ([[A0]]), %xmm0
+  %z = load float* %x
+  %t = call float @llvm.sqrt.f32(float %z)
+  ret float %t
+}
+
+define double @squirt_size(double* %x) nounwind optsize {
+entry:
+; CHECK: squirt_size:
+; CHECK: sqrtsd ([[A0]]), %xmm0
+  %z = load double* %x
+  %t = call double @llvm.sqrt.f64(double %z)
+  ret double %t
+}
+
+declare float @llvm.sqrt.f32(float)
+declare double @llvm.sqrt.f64(double)