fix a missing shuffle pattern, PR9009. Patch by Artiom Myaskouvskey!

[oota-llvm.git] / test / CodeGen / X86 / sse41.ll

diff --git a/test/CodeGen/X86/sse41.ll b/test/CodeGen/X86/sse41.ll
index a35de02051ee3825dc9425e7e0d344b98c77d57c..2ac4cb435a751ba6b2d4151e8fbd1ade992deaab 100644 (file)
--- a/test/CodeGen/X86/sse41.ll
+++ b/test/CodeGen/X86/sse41.ll
@@ -1,5 +1,5 @@
-; RUN: llvm-as < %s | llc -mtriple=i686-apple-darwin9 -mattr=sse41 | FileCheck %s -check-prefix=X32
-; RUN: llvm-as < %s | llc -mtriple=x86_64-apple-darwin9 -mattr=sse41 | FileCheck %s -check-prefix=X64
+; RUN: llc < %s -mtriple=i686-apple-darwin9 -mattr=sse41 | FileCheck %s -check-prefix=X32
+; RUN: llc < %s -mtriple=x86_64-apple-darwin9 -mattr=sse41 | FileCheck %s -check-prefix=X64
 
 @g16 = external global i16
 
@@ -123,11 +123,11 @@ define float @ext_1(<4 x float> %v) nounwind {
 
 ; X32: _ext_1:
 ; X32:   pshufd        $3, %xmm0, %xmm0
-; X32:   addss LCPI8_0, %xmm0
+; X32:   addss LCPI7_0, %xmm0
 
 ; X64: _ext_1:
 ; X64:   pshufd        $3, %xmm0, %xmm0
-; X64:   addss LCPI8_0(%rip), %xmm0
+; X64:   addss LCPI7_0(%rip), %xmm0
 }
 define float @ext_2(<4 x float> %v) nounwind {
   %s = extractelement <4 x float> %v, i32 3
@@ -150,4 +150,102 @@ define i32 @ext_3(<4 x i32> %v) nounwind {
 ; X64:   pextrd        $3, %xmm0, %eax
 }
 
+define <4 x float> @insertps_1(<4 x float> %t1, <4 x float> %t2) nounwind {
+        %tmp1 = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %t1, <4 x float> %t2, i32 1) nounwind readnone
+        ret <4 x float> %tmp1
+; X32: _insertps_1:
+; X32:    insertps  $1, %xmm1, %xmm0
+
+; X64: _insertps_1:
+; X64:    insertps  $1, %xmm1, %xmm0
+}
+
+declare <4 x float> @llvm.x86.sse41.insertps(<4 x float>, <4 x float>, i32) nounwind readnone
+
+define <4 x float> @insertps_2(<4 x float> %t1, float %t2) nounwind {
+        %tmp1 = insertelement <4 x float> %t1, float %t2, i32 0
+        ret <4 x float> %tmp1
+; X32: _insertps_2:
+; X32:    insertps  $0, 4(%esp), %xmm0
+
+; X64: _insertps_2:
+; X64:    insertps  $0, %xmm1, %xmm0        
+}
+
+define <4 x float> @insertps_3(<4 x float> %t1, <4 x float> %t2) nounwind {
+        %tmp2 = extractelement <4 x float> %t2, i32 0
+        %tmp1 = insertelement <4 x float> %t1, float %tmp2, i32 0
+        ret <4 x float> %tmp1
+; X32: _insertps_3:
+; X32:    insertps  $0, %xmm1, %xmm0        
+
+; X64: _insertps_3:
+; X64:    insertps  $0, %xmm1, %xmm0        
+}
+
+define i32 @ptestz_1(<4 x float> %t1, <4 x float> %t2) nounwind {
+        %tmp1 = call i32 @llvm.x86.sse41.ptestz(<4 x float> %t1, <4 x float> %t2) nounwind readnone
+        ret i32 %tmp1
+; X32: _ptestz_1:
+; X32:    ptest        %xmm1, %xmm0
+; X32:    sete %al
+
+; X64: _ptestz_1:
+; X64:    ptest        %xmm1, %xmm0
+; X64:    sete %al
+}
+
+define i32 @ptestz_2(<4 x float> %t1, <4 x float> %t2) nounwind {
+        %tmp1 = call i32 @llvm.x86.sse41.ptestc(<4 x float> %t1, <4 x float> %t2) nounwind readnone
+        ret i32 %tmp1
+; X32: _ptestz_2:
+; X32:    ptest        %xmm1, %xmm0
+; X32:    sbbl %eax
+
+; X64: _ptestz_2:
+; X64:    ptest        %xmm1, %xmm0
+; X64:    sbbl %eax
+}
+
+define i32 @ptestz_3(<4 x float> %t1, <4 x float> %t2) nounwind {
+        %tmp1 = call i32 @llvm.x86.sse41.ptestnzc(<4 x float> %t1, <4 x float> %t2) nounwind readnone
+        ret i32 %tmp1
+; X32: _ptestz_3:
+; X32:    ptest        %xmm1, %xmm0
+; X32:    seta %al
+
+; X64: _ptestz_3:
+; X64:    ptest        %xmm1, %xmm0
+; X64:    seta %al
+}
+
+
+declare i32 @llvm.x86.sse41.ptestz(<4 x float>, <4 x float>) nounwind readnone
+declare i32 @llvm.x86.sse41.ptestc(<4 x float>, <4 x float>) nounwind readnone
+declare i32 @llvm.x86.sse41.ptestnzc(<4 x float>, <4 x float>) nounwind readnone
+
+; This used to compile to insertps $0  + insertps $16.  insertps $0 is always
+; pointless.
+define <2 x float> @buildvector(<2 x float> %A, <2 x float> %B) nounwind  {
+entry:
+  %tmp7 = extractelement <2 x float> %A, i32 0
+  %tmp5 = extractelement <2 x float> %A, i32 1
+  %tmp3 = extractelement <2 x float> %B, i32 0
+  %tmp1 = extractelement <2 x float> %B, i32 1
+  %add.r = fadd float %tmp7, %tmp3
+  %add.i = fadd float %tmp5, %tmp1
+  %tmp11 = insertelement <2 x float> undef, float %add.r, i32 0
+  %tmp9 = insertelement <2 x float> %tmp11, float %add.i, i32 1
+  ret <2 x float> %tmp9
+; X32: buildvector:
+; X32-NOT: insertps $0
+; X32: insertps $16
+; X32-NOT: insertps $0
+; X32: ret
+; X64: buildvector:
+; X64-NOT: insertps $0
+; X64: insertps $16
+; X64-NOT: insertps $0
+; X64: ret
+}