RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-23 11:13:28 +01:00
Code Issues Projects Releases Wiki Activity

[X86][SSE4A] Test SSE4A shuffle combining on SSE42 capable target as well

Browse Source 
llvm-svn: 307038
This commit is contained in:
Simon Pilgrim 2017-07-03 15:55:54 +00:00
parent e1f310fac6
commit 66be995978
1 changed files with 36 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

53
test/CodeGen/X86/vector-shuffle-combining-sse4a.ll
View File

@ -1,9 +1,10 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+ssse3,+sse4a | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE4A
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+ssse3,+sse4a | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSSE3
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+sse4.2,+sse4a | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE42
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx,+sse4a| FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx2,+sse4a | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
;
; Combine tests involving SSE41 target shuffles (BLEND,INSERTPS,MOVZX)
; Combine tests involving SSE4A target shuffles (EXTRQI,INSERTQI)
declare <16 x i8> @llvm.x86.ssse3.pshuf.b.128(<16 x i8>, <16 x i8>)
@ -25,11 +26,17 @@ define <16 x i8> @combine_extrqi_pshufb_16i8(<16 x i8> %a0) {
}
define <8 x i16> @combine_extrqi_pshufb_8i16(<8 x i16> %a0) {
; SSE-LABEL: combine_extrqi_pshufb_8i16:
; SSE: # BB#0:
; SSE-NEXT: extrq {{.*#+}} xmm0 = xmm0[2,3,4,5],zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSE-NEXT: pshufb {{.*#+}} xmm0 = xmm0[0,1],zero,zero,zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSE-NEXT: retq
; SSSE3-LABEL: combine_extrqi_pshufb_8i16:
; SSSE3: # BB#0:
; SSSE3-NEXT: extrq {{.*#+}} xmm0 = xmm0[2,3,4,5],zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSSE3-NEXT: pshufb {{.*#+}} xmm0 = xmm0[0,1],zero,zero,zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSSE3-NEXT: retq
;
; SSE42-LABEL: combine_extrqi_pshufb_8i16:
; SSE42: # BB#0:
; SSE42-NEXT: extrq {{.*#+}} xmm0 = xmm0[2,3,4,5],zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSE42-NEXT: pmovzxwq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; SSE42-NEXT: retq
;
; AVX-LABEL: combine_extrqi_pshufb_8i16:
; AVX: # BB#0:
@ -44,11 +51,17 @@ define <8 x i16> @combine_extrqi_pshufb_8i16(<8 x i16> %a0) {
}
define <16 x i8> @combine_insertqi_pshufb_16i8(<16 x i8> %a0, <16 x i8> %a1) {
; SSE-LABEL: combine_insertqi_pshufb_16i8:
; SSE: # BB#0:
; SSE-NEXT: insertq {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3,4,5,6,7,u,u,u,u,u,u,u,u]
; SSE-NEXT: pshufb {{.*#+}} xmm0 = xmm0[0,1],zero,zero,zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSE-NEXT: retq
; SSSE3-LABEL: combine_insertqi_pshufb_16i8:
; SSSE3: # BB#0:
; SSSE3-NEXT: insertq {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3,4,5,6,7,u,u,u,u,u,u,u,u]
; SSSE3-NEXT: pshufb {{.*#+}} xmm0 = xmm0[0,1],zero,zero,zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSSE3-NEXT: retq
;
; SSE42-LABEL: combine_insertqi_pshufb_16i8:
; SSE42: # BB#0:
; SSE42-NEXT: insertq {{.*#+}} xmm0 = xmm1[0,1,2],xmm0[3,4,5,6,7,u,u,u,u,u,u,u,u]
; SSE42-NEXT: pmovzxwq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; SSE42-NEXT: retq
;
; AVX-LABEL: combine_insertqi_pshufb_16i8:
; AVX: # BB#0:
@ -61,11 +74,17 @@ define <16 x i8> @combine_insertqi_pshufb_16i8(<16 x i8> %a0, <16 x i8> %a1) {
}
define <8 x i16> @combine_insertqi_pshufb_8i16(<8 x i16> %a0, <8 x i16> %a1) {
; SSE-LABEL: combine_insertqi_pshufb_8i16:
; SSE: # BB#0:
; SSE-NEXT: insertq {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3,4,5,6,7,u,u,u,u,u,u,u,u]
; SSE-NEXT: pshufb {{.*#+}} xmm0 = xmm0[0,1],zero,zero,zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSE-NEXT: retq
; SSSE3-LABEL: combine_insertqi_pshufb_8i16:
; SSSE3: # BB#0:
; SSSE3-NEXT: insertq {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3,4,5,6,7,u,u,u,u,u,u,u,u]
; SSSE3-NEXT: pshufb {{.*#+}} xmm0 = xmm0[0,1],zero,zero,zero,zero,zero,zero,xmm0[u,u,u,u,u,u,u,u]
; SSSE3-NEXT: retq
;
; SSE42-LABEL: combine_insertqi_pshufb_8i16:
; SSE42: # BB#0:
; SSE42-NEXT: insertq {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3,4,5,6,7,u,u,u,u,u,u,u,u]
; SSE42-NEXT: pmovzxwq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
; SSE42-NEXT: retq
;
; AVX-LABEL: combine_insertqi_pshufb_8i16:
; AVX: # BB#0: