OSchip
/
llvm-project
mirror of https://github.com/THU-DSP-LAB/llvm-project.git
13
0
Fork 6
Code Issues Packages Projects Releases Wiki Activity

[AVX-512] Fix a couple test cases to not pass an undef mask to gather intrinsic. This could break if any future optimizations taken advantage of the undef. 

llvm-svn: 292585
This commit is contained in:
Craig Topper 2017-01-20 07:12:30 +00:00
parent e48d7d5554
commit ae78b5dcff
1 changed files with 14 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

20
llvm/test/CodeGen/X86/masked_gather_scatter.ll
View File

@ -675,6 +675,7 @@ define <16 x float> @test12(float* %base, <16 x i32> %ind) {
define <16 x float> @test13(float* %base, <16 x i32> %ind) {
; KNL_64-LABEL: test13:
; KNL_64: # BB#0:
; KNL_64-NEXT: kxnorw %k0, %k0, %k1
; KNL_64-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1}
; KNL_64-NEXT: vmovaps %zmm1, %zmm0
; KNL_64-NEXT: retq
@ -682,12 +683,14 @@ define <16 x float> @test13(float* %base, <16 x i32> %ind) {
; KNL_32-LABEL: test13:
; KNL_32: # BB#0:
; KNL_32-NEXT: movl {{[0-9]+}}(%esp), %eax
; KNL_32-NEXT: kxnorw %k0, %k0, %k1
; KNL_32-NEXT: vgatherdps (%eax,%zmm0,4), %zmm1 {%k1}
; KNL_32-NEXT: vmovaps %zmm1, %zmm0
; KNL_32-NEXT: retl
;
; SKX-LABEL: test13:
; SKX: # BB#0:
; SKX-NEXT: kxnorw %k0, %k0, %k1
; SKX-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm1 {%k1}
; SKX-NEXT: vmovaps %zmm1, %zmm0
; SKX-NEXT: retq
@ -695,6 +698,7 @@ define <16 x float> @test13(float* %base, <16 x i32> %ind) {
; SKX_32-LABEL: test13:
; SKX_32: # BB#0:
; SKX_32-NEXT: movl {{[0-9]+}}(%esp), %eax
; SKX_32-NEXT: kxnorw %k0, %k0, %k1
; SKX_32-NEXT: vgatherdps (%eax,%zmm0,4), %zmm1 {%k1}
; SKX_32-NEXT: vmovaps %zmm1, %zmm0
; SKX_32-NEXT: retl
@ -702,7 +706,7 @@ define <16 x float> @test13(float* %base, <16 x i32> %ind) {
%sext_ind = sext <16 x i32> %ind to <16 x i64>
%gep.random = getelementptr float, float *%base, <16 x i64> %sext_ind
%res = call <16 x float> @llvm.masked.gather.v16f32(<16 x float*> %gep.random, i32 4, <16 x i1> undef, <16 x float> undef)
%res = call <16 x float> @llvm.masked.gather.v16f32(<16 x float*> %gep.random, i32 4, <16 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <16 x float> undef)
ret <16 x float>%res
}
@ -718,8 +722,9 @@ define <16 x float> @test14(float* %base, i32 %ind, <16 x float*> %vec) {
; KNL_64-NEXT: vpmovsxdq %ymm1, %zmm1
; KNL_64-NEXT: vpsllq $2, %zmm1, %zmm1
; KNL_64-NEXT: vpaddq %zmm1, %zmm0, %zmm0
; KNL_64-NEXT: kshiftrw $8, %k0, %k1
; KNL_64-NEXT: vgatherqps (,%zmm0), %ymm1 {%k1}
; KNL_64-NEXT: kxnorw %k0, %k0, %k1
; KNL_64-NEXT: kshiftrw $8, %k1, %k2
; KNL_64-NEXT: vgatherqps (,%zmm0), %ymm1 {%k2}
; KNL_64-NEXT: vgatherqps (,%zmm0), %ymm2 {%k1}
; KNL_64-NEXT: vinsertf64x4 $1, %ymm1, %zmm2, %zmm0
; KNL_64-NEXT: retq
@ -731,6 +736,7 @@ define <16 x float> @test14(float* %base, i32 %ind, <16 x float*> %vec) {
; KNL_32-NEXT: vpbroadcastd %xmm0, %zmm0
; KNL_32-NEXT: vpslld $2, {{[0-9]+}}(%esp){1to16}, %zmm1
; KNL_32-NEXT: vpaddd %zmm1, %zmm0, %zmm1
; KNL_32-NEXT: kxnorw %k0, %k0, %k1
; KNL_32-NEXT: vgatherdps (,%zmm1), %zmm0 {%k1}
; KNL_32-NEXT: retl
;
@ -743,8 +749,9 @@ define <16 x float> @test14(float* %base, i32 %ind, <16 x float*> %vec) {
; SKX-NEXT: vpmovsxdq %ymm1, %zmm1
; SKX-NEXT: vpsllq $2, %zmm1, %zmm1
; SKX-NEXT: vpaddq %zmm1, %zmm0, %zmm0
; SKX-NEXT: kshiftrw $8, %k0, %k1
; SKX-NEXT: vgatherqps (,%zmm0), %ymm1 {%k1}
; SKX-NEXT: kxnorw %k0, %k0, %k1
; SKX-NEXT: kshiftrw $8, %k1, %k2
; SKX-NEXT: vgatherqps (,%zmm0), %ymm1 {%k2}
; SKX-NEXT: vgatherqps (,%zmm0), %ymm2 {%k1}
; SKX-NEXT: vinsertf32x8 $1, %ymm1, %zmm2, %zmm0
; SKX-NEXT: retq
@ -756,6 +763,7 @@ define <16 x float> @test14(float* %base, i32 %ind, <16 x float*> %vec) {
; SKX_32-NEXT: vpbroadcastd %xmm0, %zmm0
; SKX_32-NEXT: vpslld $2, {{[0-9]+}}(%esp){1to16}, %zmm1
; SKX_32-NEXT: vpaddd %zmm1, %zmm0, %zmm1
; SKX_32-NEXT: kxnorw %k0, %k0, %k1
; SKX_32-NEXT: vgatherdps (,%zmm1), %zmm0 {%k1}
; SKX_32-NEXT: retl
@ -764,7 +772,7 @@ define <16 x float> @test14(float* %base, i32 %ind, <16 x float*> %vec) {
%gep.random = getelementptr float, <16 x float*> %broadcast.splat, i32 %ind
%res = call <16 x float> @llvm.masked.gather.v16f32(<16 x float*> %gep.random, i32 4, <16 x i1> undef, <16 x float> undef)
%res = call <16 x float> @llvm.masked.gather.v16f32(<16 x float*> %gep.random, i32 4, <16 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>, <16 x float> undef)
ret <16 x float>%res
}