diff options
| -rw-r--r-- | test/CodeGen/X86/sse2.ll | 21 |
1 files changed, 18 insertions, 3 deletions
diff --git a/test/CodeGen/X86/sse2.ll b/test/CodeGen/X86/sse2.ll index 88e303d..463df30 100644 --- a/test/CodeGen/X86/sse2.ll +++ b/test/CodeGen/X86/sse2.ll @@ -1,6 +1,5 @@ -; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse2 | grep movlpd -; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse2 | grep movhpd -; RUN: llvm-as < %s | llc -march=x86 -mattr=+sse2 | not grep movsd +; Tests for SSE2 and below, without SSE3+. +; RUN: llvm-as < %s | llc -march=x86 -mcpu=pentium4 | FileCheck %s define void @t1(<2 x double>* %r, <2 x double>* %A, double %B) nounwind { %tmp3 = load <2 x double>* %A, align 16 @@ -8,6 +7,14 @@ define void @t1(<2 x double>* %r, <2 x double>* %A, double %B) nounwind { %tmp9 = shufflevector <2 x double> %tmp3, <2 x double> %tmp7, <2 x i32> < i32 2, i32 1 > store <2 x double> %tmp9, <2 x double>* %r, align 16 ret void + +; CHECK: t1: +; CHECK: movl 8(%esp), %eax +; CHECK: movapd (%eax), %xmm0 +; CHECK: movlpd 12(%esp), %xmm0 +; CHECK: movl 4(%esp), %eax +; CHECK: movapd %xmm0, (%eax) +; CHECK: ret } define void @t2(<2 x double>* %r, <2 x double>* %A, double %B) nounwind { @@ -16,4 +23,12 @@ define void @t2(<2 x double>* %r, <2 x double>* %A, double %B) nounwind { %tmp9 = shufflevector <2 x double> %tmp3, <2 x double> %tmp7, <2 x i32> < i32 0, i32 2 > store <2 x double> %tmp9, <2 x double>* %r, align 16 ret void + +; CHECK: t2: +; CHECK: movl 8(%esp), %eax +; CHECK: movapd (%eax), %xmm0 +; CHECK: movhpd 12(%esp), %xmm0 +; CHECK: movl 4(%esp), %eax +; CHECK: movapd %xmm0, (%eax) +; CHECK: ret } |