CodeGen peephole: fold redundant phys reg copies

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

diff --git a/test/CodeGen/X86/sse-domains.ll b/test/CodeGen/X86/sse-domains.ll
index d26d32287e03c8a4a79286d0151dc9d20ce51089..8016a246fa66e72e764dc91cc9f2a97c9c201414 100644 (file)
--- a/test/CodeGen/X86/sse-domains.ll
+++ b/test/CodeGen/X86/sse-domains.ll
@@ -1,4 +1,4 @@
-; RUN: llc < %s | FileCheck %s
+; RUN: llc < %s -mcpu=nehalem | FileCheck %s
 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-apple-macosx10.7"
 
@@ -33,9 +33,9 @@ while.body:
   %x.02 = phi <4 x i32> [ %add, %while.body ], [ zeroinitializer, %entry ]
   %dec = add nsw i32 %n.addr.03, -1
   %and = and <4 x i32> %x.02, <i32 127, i32 127, i32 127, i32 127>
-  %incdec.ptr = getelementptr inbounds <4 x i32>* %p.addr.04, i64 1
+  %incdec.ptr = getelementptr inbounds <4 x i32>, <4 x i32>* %p.addr.04, i64 1
   store <4 x i32> %and, <4 x i32>* %p.addr.04, align 16
-  %0 = load <4 x i32>* %incdec.ptr, align 16
+  %0 = load <4 x i32>, <4 x i32>* %incdec.ptr, align 16
   %add = shl <4 x i32> %0, <i32 1, i32 1, i32 1, i32 1>
   %tobool = icmp eq i32 %dec, 0
   br i1 %tobool, label %while.end, label %while.body