diff options
Diffstat (limited to 'llvm/test/CodeGen/X86/vector-rotate-128.ll')
-rw-r--r-- | llvm/test/CodeGen/X86/vector-rotate-128.ll | 45 |
1 files changed, 15 insertions, 30 deletions
diff --git a/llvm/test/CodeGen/X86/vector-rotate-128.ll b/llvm/test/CodeGen/X86/vector-rotate-128.ll index 04378ee2ee0..5fafecae23d 100644 --- a/llvm/test/CodeGen/X86/vector-rotate-128.ll +++ b/llvm/test/CodeGen/X86/vector-rotate-128.ll @@ -476,7 +476,7 @@ define <8 x i16> @var_rotate_v8i16(<8 x i16> %a, <8 x i16> %b) nounwind { ; ; AVX512VL-LABEL: var_rotate_v8i16: ; AVX512VL: # BB#0: -; AVX512VL-NEXT: vmovdqu {{.*#+}} xmm2 = [16,16,16,16,16,16,16,16] +; AVX512VL-NEXT: vmovdqa {{.*#+}} xmm2 = [16,16,16,16,16,16,16,16] ; AVX512VL-NEXT: vpsubw %xmm1, %xmm2, %xmm2 ; AVX512VL-NEXT: vpsllvw %xmm1, %xmm0, %xmm1 ; AVX512VL-NEXT: vpsrlvw %xmm2, %xmm0, %xmm0 @@ -701,35 +701,20 @@ define <16 x i8> @var_rotate_v16i8(<16 x i8> %a, <16 x i8> %b) nounwind { ; AVX-NEXT: vpor %xmm0, %xmm1, %xmm0 ; AVX-NEXT: retq ; -; AVX512BW-LABEL: var_rotate_v16i8: -; AVX512BW: # BB#0: -; AVX512BW-NEXT: vmovdqa {{.*#+}} xmm2 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8] -; AVX512BW-NEXT: vpsubb %xmm1, %xmm2, %xmm2 -; AVX512BW-NEXT: vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero -; AVX512BW-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero -; AVX512BW-NEXT: vpsllvd %zmm1, %zmm0, %zmm1 -; AVX512BW-NEXT: vpmovdb %zmm1, %xmm1 -; AVX512BW-NEXT: vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero -; AVX512BW-NEXT: vpsrlvd %zmm2, %zmm0, %zmm0 -; AVX512BW-NEXT: vpmovdb %zmm0, %xmm0 -; AVX512BW-NEXT: vpor %xmm0, %xmm1, %xmm0 -; AVX512BW-NEXT: vzeroupper -; AVX512BW-NEXT: retq -; -; AVX512VL-LABEL: var_rotate_v16i8: -; AVX512VL: # BB#0: -; AVX512VL-NEXT: vmovdqu {{.*#+}} xmm2 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8] -; AVX512VL-NEXT: vpsubb %xmm1, %xmm2, %xmm2 -; AVX512VL-NEXT: vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero -; AVX512VL-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero -; AVX512VL-NEXT: vpsllvd %zmm1, %zmm0, %zmm1 -; AVX512VL-NEXT: vpmovdb %zmm1, %xmm1 -; AVX512VL-NEXT: vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero -; AVX512VL-NEXT: vpsrlvd %zmm2, %zmm0, %zmm0 -; AVX512VL-NEXT: vpmovdb %zmm0, %xmm0 -; AVX512VL-NEXT: vpor %xmm0, %xmm1, %xmm0 -; AVX512VL-NEXT: vzeroupper -; AVX512VL-NEXT: retq +; AVX512-LABEL: var_rotate_v16i8: +; AVX512: # BB#0: +; AVX512-NEXT: vmovdqa {{.*#+}} xmm2 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8] +; AVX512-NEXT: vpsubb %xmm1, %xmm2, %xmm2 +; AVX512-NEXT: vpmovzxbd {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero,xmm1[8],zero,zero,zero,xmm1[9],zero,zero,zero,xmm1[10],zero,zero,zero,xmm1[11],zero,zero,zero,xmm1[12],zero,zero,zero,xmm1[13],zero,zero,zero,xmm1[14],zero,zero,zero,xmm1[15],zero,zero,zero +; AVX512-NEXT: vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero +; AVX512-NEXT: vpsllvd %zmm1, %zmm0, %zmm1 +; AVX512-NEXT: vpmovdb %zmm1, %xmm1 +; AVX512-NEXT: vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero +; AVX512-NEXT: vpsrlvd %zmm2, %zmm0, %zmm0 +; AVX512-NEXT: vpmovdb %zmm0, %xmm0 +; AVX512-NEXT: vpor %xmm0, %xmm1, %xmm0 +; AVX512-NEXT: vzeroupper +; AVX512-NEXT: retq ; ; XOP-LABEL: var_rotate_v16i8: ; XOP: # BB#0: |