Fix a broadcast related regression on the vector shuffle lowering.

[oota-llvm.git] / test / CodeGen / X86 / byval4.ll

diff --git a/test/CodeGen/X86/byval4.ll b/test/CodeGen/X86/byval4.ll
index 591749f768e138436374350d34385b2057406104..4711e451111274fe7a84d8d3fd21ab663151e858 100644 (file)
--- a/test/CodeGen/X86/byval4.ll
+++ b/test/CodeGen/X86/byval4.ll
@@ -1,11 +1,42 @@
-; RUN: llvm-as < %s | llc -march=x86-64 | grep rep.movsw | count 2
-; RUN: llvm-as < %s | llc -march=x86 | grep rep.movsl   | count 2
+; RUN: llc < %s -mtriple=x86_64-linux -mattr=-avx | FileCheck %s -check-prefix=X64
+; X64-NOT:     movsq
+; X64:     rep
+; X64-NOT:     rep
+; X64:     movsq
+; X64-NOT:     movsq
+; X64:     rep
+; X64-NOT:     rep
+; X64:     movsq
+; X64-NOT:     rep
+; X64-NOT:     movsq
 
-%struct.s = type { i16, i16, i16, i16, i16, i16 }
+; Win64 has not supported byval yet.
+
+; RUN: llc < %s -march=x86 -mattr=-avx | FileCheck %s -check-prefix=X32
+; X32-NOT:     movsl
+; X32:     rep
+; X32-NOT:     rep
+; X32:     movsl
+; X32-NOT:     movsl
+; X32:     rep
+; X32-NOT:     rep
+; X32:     movsl
+; X32-NOT:     rep
+; X32-NOT:     movsl
+
+%struct.s = type { i16, i16, i16, i16, i16, i16, i16, i16,
+                   i16, i16, i16, i16, i16, i16, i16, i16,
+                   i16, i16, i16, i16, i16, i16, i16, i16,
+                   i16, i16, i16, i16, i16, i16, i16, i16,
+                   i16, i16, i16, i16, i16, i16, i16, i16,
+                   i16, i16, i16, i16, i16, i16, i16, i16,
+                   i16, i16, i16, i16, i16, i16, i16, i16,
+                   i16, i16, i16, i16, i16, i16, i16, i16,
+                   i16 }
 
 
 define void @g(i16 signext  %a1, i16 signext  %a2, i16 signext  %a3,
-        i16 signext  %a4, i16 signext  %a5, i16 signext  %a6) {
+        i16 signext  %a4, i16 signext  %a5, i16 signext  %a6) nounwind {
 entry:
         %a = alloca %struct.s, align 16
         %tmp = getelementptr %struct.s* %a, i32 0, i32 0
@@ -20,8 +51,8 @@ entry:
         store i16 %a5, i16* %tmp8, align 16
         %tmp10 = getelementptr %struct.s* %a, i32 0, i32 5
         store i16 %a6, i16* %tmp10, align 16
-        call void @f( %struct.s* %a byval )
-        call void @f( %struct.s* %a byval )
+        call void @f( %struct.s* byval %a )
+        call void @f( %struct.s* byval %a )
         ret void
 }