CodeGen peephole: fold redundant phys reg copies

[oota-llvm.git] / test / CodeGen / X86 / sse2-intrinsics-x86.ll

diff --git a/test/CodeGen/X86/sse2-intrinsics-x86.ll b/test/CodeGen/X86/sse2-intrinsics-x86.ll
index a6b0b90bd806cb3d413cd21eccc59eadc96fe4f4..53132a167fb83ad67504f5f20a1d8f7edf571c6c 100644 (file)
--- a/test/CodeGen/X86/sse2-intrinsics-x86.ll
+++ b/test/CodeGen/X86/sse2-intrinsics-x86.ll
@@ -1,4 +1,5 @@
-; RUN: llc < %s -mtriple=i386-apple-darwin -mattr=-avx,+sse2 | FileCheck %s
+; RUN: llc < %s -mtriple=i386-apple-darwin -mattr=-avx,+sse2 | FileCheck %s --check-prefix=CHECK --check-prefix=SSE
+; RUN: llc < %s -mtriple=i386-apple-darwin -mcpu=knl | FileCheck %s
 
 define <2 x double> @test_x86_sse2_add_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: addsd
@@ -142,6 +143,7 @@ declare i32 @llvm.x86.sse2.cvtsd2si(<2 x double>) nounwind readnone
 
 define <4 x float> @test_x86_sse2_cvtsd2ss(<4 x float> %a0, <2 x double> %a1) {
   ; CHECK: cvtsd2ss 
+  ; SSE-NOT: cvtsd2ss %xmm{{[0-9]+}}, %xmm{{[0-9]+}}, %xmm{{[0-9]+}} 
   %res = call <4 x float> @llvm.x86.sse2.cvtsd2ss(<4 x float> %a0, <2 x double> %a1) ; <<4 x float>> [#uses=1]
   ret <4 x float> %res
 }
@@ -409,22 +411,6 @@ define <4 x i32> @test_x86_sse2_psll_d(<4 x i32> %a0, <4 x i32> %a1) {
 declare <4 x i32> @llvm.x86.sse2.psll.d(<4 x i32>, <4 x i32>) nounwind readnone
 
 
-define <2 x i64> @test_x86_sse2_psll_dq(<2 x i64> %a0) {
-  ; CHECK: pslldq
-  %res = call <2 x i64> @llvm.x86.sse2.psll.dq(<2 x i64> %a0, i32 7) ; <<2 x i64>> [#uses=1]
-  ret <2 x i64> %res
-}
-declare <2 x i64> @llvm.x86.sse2.psll.dq(<2 x i64>, i32) nounwind readnone
-
-
-define <2 x i64> @test_x86_sse2_psll_dq_bs(<2 x i64> %a0) {
-  ; CHECK: pslldq
-  %res = call <2 x i64> @llvm.x86.sse2.psll.dq.bs(<2 x i64> %a0, i32 7) ; <<2 x i64>> [#uses=1]
-  ret <2 x i64> %res
-}
-declare <2 x i64> @llvm.x86.sse2.psll.dq.bs(<2 x i64>, i32) nounwind readnone
-
-
 define <2 x i64> @test_x86_sse2_psll_q(<2 x i64> %a0, <2 x i64> %a1) {
   ; CHECK: psllq
   %res = call <2 x i64> @llvm.x86.sse2.psll.q(<2 x i64> %a0, <2 x i64> %a1) ; <<2 x i64>> [#uses=1]
@@ -505,22 +491,6 @@ define <4 x i32> @test_x86_sse2_psrl_d(<4 x i32> %a0, <4 x i32> %a1) {
 declare <4 x i32> @llvm.x86.sse2.psrl.d(<4 x i32>, <4 x i32>) nounwind readnone
 
 
-define <2 x i64> @test_x86_sse2_psrl_dq(<2 x i64> %a0) {
-  ; CHECK: psrldq
-  %res = call <2 x i64> @llvm.x86.sse2.psrl.dq(<2 x i64> %a0, i32 7) ; <<2 x i64>> [#uses=1]
-  ret <2 x i64> %res
-}
-declare <2 x i64> @llvm.x86.sse2.psrl.dq(<2 x i64>, i32) nounwind readnone
-
-
-define <2 x i64> @test_x86_sse2_psrl_dq_bs(<2 x i64> %a0) {
-  ; CHECK: psrldq
-  %res = call <2 x i64> @llvm.x86.sse2.psrl.dq.bs(<2 x i64> %a0, i32 7) ; <<2 x i64>> [#uses=1]
-  ret <2 x i64> %res
-}
-declare <2 x i64> @llvm.x86.sse2.psrl.dq.bs(<2 x i64>, i32) nounwind readnone
-
-
 define <2 x i64> @test_x86_sse2_psrl_q(<2 x i64> %a0, <2 x i64> %a1) {
   ; CHECK: psrlq
   %res = call <2 x i64> @llvm.x86.sse2.psrl.q(<2 x i64> %a0, <2 x i64> %a1) ; <<2 x i64>> [#uses=1]
@@ -612,7 +582,7 @@ declare <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double>) nounwind readnone
 define void @test_x86_sse2_storel_dq(i8* %a0, <4 x i32> %a1) {
   ; CHECK: test_x86_sse2_storel_dq
   ; CHECK: movl
-  ; CHECK: movq
+  ; CHECK: movlps
   call void @llvm.x86.sse2.storel.dq(i8* %a0, <4 x i32> %a1)
   ret void
 }
@@ -709,3 +679,37 @@ define i32 @test_x86_sse2_ucomineq_sd(<2 x double> %a0, <2 x double> %a1) {
   ret i32 %res
 }
 declare i32 @llvm.x86.sse2.ucomineq.sd(<2 x double>, <2 x double>) nounwind readnone
+
+define void @test_x86_sse2_pause() {
+  ; CHECK: pause
+  tail call void @llvm.x86.sse2.pause()
+  ret void 
+}
+declare void @llvm.x86.sse2.pause() nounwind
+
+define <4 x i32> @test_x86_sse2_pshuf_d(<4 x i32> %a) {
+; CHECK-LABEL: test_x86_sse2_pshuf_d:
+; CHECK: pshufd $27
+entry:
+   %res = call <4 x i32> @llvm.x86.sse2.pshuf.d(<4 x i32> %a, i8 27) nounwind readnone
+   ret <4 x i32> %res
+}
+declare <4 x i32> @llvm.x86.sse2.pshuf.d(<4 x i32>, i8) nounwind readnone
+
+define <8 x i16> @test_x86_sse2_pshufl_w(<8 x i16> %a) {
+; CHECK-LABEL: test_x86_sse2_pshufl_w:
+; CHECK: pshuflw $27
+entry:
+   %res = call <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16> %a, i8 27) nounwind readnone
+   ret <8 x i16> %res
+}
+declare <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16>, i8) nounwind readnone
+
+define <8 x i16> @test_x86_sse2_pshufh_w(<8 x i16> %a) {
+; CHECK-LABEL: test_x86_sse2_pshufh_w:
+; CHECK: pshufhw $27
+entry:
+   %res = call <8 x i16> @llvm.x86.sse2.pshufh.w(<8 x i16> %a, i8 27) nounwind readnone
+   ret <8 x i16> %res
+}
+declare <8 x i16> @llvm.x86.sse2.pshufh.w(<8 x i16>, i8) nounwind readnone