; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=AVX --check-prefix=AVX1
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=AVX --check-prefix=AVX2
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f | FileCheck %s --check-prefix=AVX512 --check-prefix=AVX512F
; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512vl | FileCheck %s --check-prefix=AVX512 --check-prefix=AVX512VL
; Ideally, the shuffles should be lowered to code with the same quality as the truncates.
define void @shuffle_v32i8_to_v16i8(<32 x i8>* %L, <16 x i8>* %S) nounwind {
-; AVX-LABEL: shuffle_v32i8_to_v16i8:
-; AVX: # BB#0:
-; AVX-NEXT: vmovdqa (%rdi), %ymm0
-; AVX-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX-NEXT: vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX-NEXT: vpshufb %xmm2, %xmm1, %xmm1
-; AVX-NEXT: vpshufb %xmm2, %xmm0, %xmm0
-; AVX-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX-NEXT: vmovdqa %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: shuffle_v32i8_to_v16i8:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovdqa (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
+; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT: vmovdqa %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v32i8_to_v16i8:
+; AVX2: # BB#0:
+; AVX2-NEXT: vmovdqa (%rdi), %ymm0
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
+; AVX2-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX2-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT: vmovdqa %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: shuffle_v32i8_to_v16i8:
; AVX512F: # BB#0:
}
define void @trunc_v16i16_to_v16i8(<32 x i8>* %L, <16 x i8>* %S) nounwind {
-; AVX-LABEL: trunc_v16i16_to_v16i8:
-; AVX: # BB#0:
-; AVX-NEXT: vmovdqa (%rdi), %ymm0
-; AVX-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX-NEXT: vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
-; AVX-NEXT: vpshufb %xmm2, %xmm1, %xmm1
-; AVX-NEXT: vpshufb %xmm2, %xmm0, %xmm0
-; AVX-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX-NEXT: vmovdqa %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: trunc_v16i16_to_v16i8:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovdqa (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
+; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT: vmovdqa %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: trunc_v16i16_to_v16i8:
+; AVX2: # BB#0:
+; AVX2-NEXT: vmovdqa (%rdi), %ymm0
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm2 = <0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u>
+; AVX2-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX2-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT: vmovdqa %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: trunc_v16i16_to_v16i8:
; AVX512F: # BB#0:
}
define void @shuffle_v16i16_to_v8i16(<16 x i16>* %L, <8 x i16>* %S) nounwind {
-; AVX-LABEL: shuffle_v16i16_to_v8i16:
-; AVX: # BB#0:
-; AVX-NEXT: vmovdqa (%rdi), %ymm0
-; AVX-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
-; AVX-NEXT: vpshufb %xmm2, %xmm1, %xmm1
-; AVX-NEXT: vpshufb %xmm2, %xmm0, %xmm0
-; AVX-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; AVX-NEXT: vmovdqa %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: shuffle_v16i16_to_v8i16:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovdqa (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
+; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT: vmovdqa %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v16i16_to_v8i16:
+; AVX2: # BB#0:
+; AVX2-NEXT: vmovdqa (%rdi), %ymm0
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
+; AVX2-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX2-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX2-NEXT: vmovdqa %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: shuffle_v16i16_to_v8i16:
; AVX512F: # BB#0:
}
define void @trunc_v8i32_to_v8i16(<16 x i16>* %L, <8 x i16>* %S) nounwind {
-; AVX-LABEL: trunc_v8i32_to_v8i16:
-; AVX: # BB#0:
-; AVX-NEXT: vmovdqa (%rdi), %ymm0
-; AVX-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
-; AVX-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX-NEXT: vmovdqa %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: trunc_v8i32_to_v8i16:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovdqa (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
+; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX1-NEXT: vmovdqa %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: trunc_v8i32_to_v8i16:
+; AVX2: # BB#0:
+; AVX2-NEXT: vmovdqa (%rdi), %ymm0
+; AVX2-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
+; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; AVX2-NEXT: vmovdqa %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: trunc_v8i32_to_v8i16:
; AVX512F: # BB#0:
}
define void @trunc_v4i64_to_v4i32(<8 x i32>* %L, <4 x i32>* %S) nounwind {
-; AVX-LABEL: trunc_v4i64_to_v4i32:
-; AVX: # BB#0:
-; AVX-NEXT: vpshufd {{.*#+}} ymm0 = mem[0,2,2,3,4,6,6,7]
-; AVX-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX-NEXT: vmovdqa %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: trunc_v4i64_to_v4i32:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovaps (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vshufps {{.*#+}} xmm0 = xmm0[0,2],xmm1[0,2]
+; AVX1-NEXT: vmovaps %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: trunc_v4i64_to_v4i32:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpshufd {{.*#+}} ymm0 = mem[0,2,2,3,4,6,6,7]
+; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; AVX2-NEXT: vmovdqa %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: trunc_v4i64_to_v4i32:
; AVX512F: # BB#0:
}
define void @shuffle_v32i8_to_v8i8(<32 x i8>* %L, <8 x i8>* %S) nounwind {
-; AVX-LABEL: shuffle_v32i8_to_v8i8:
-; AVX: # BB#0:
-; AVX-NEXT: vmovdqa (%rdi), %ymm0
-; AVX-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX-NEXT: vmovdqa {{.*#+}} xmm2 = <0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u>
-; AVX-NEXT: vpshufb %xmm2, %xmm1, %xmm1
-; AVX-NEXT: vpshufb %xmm2, %xmm0, %xmm0
-; AVX-NEXT: vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
-; AVX-NEXT: vmovq %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: shuffle_v32i8_to_v8i8:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovdqa (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = <0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT: vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; AVX1-NEXT: vmovq %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v32i8_to_v8i8:
+; AVX2: # BB#0:
+; AVX2-NEXT: vmovdqa (%rdi), %ymm0
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm2 = <0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX2-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT: vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; AVX2-NEXT: vmovq %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: shuffle_v32i8_to_v8i8:
; AVX512F: # BB#0:
}
define void @trunc_v8i32_to_v8i8(<32 x i8>* %L, <8 x i8>* %S) nounwind {
-; AVX-LABEL: trunc_v8i32_to_v8i8:
-; AVX: # BB#0:
-; AVX-NEXT: vmovdqa (%rdi), %ymm0
-; AVX-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
-; AVX-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u]
-; AVX-NEXT: vmovq %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: trunc_v8i32_to_v8i8:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovdqa (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = <0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT: vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; AVX1-NEXT: vmovq %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: trunc_v8i32_to_v8i8:
+; AVX2: # BB#0:
+; AVX2-NEXT: vmovdqa (%rdi), %ymm0
+; AVX2-NEXT: vpshufb {{.*#+}} ymm0 = ymm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
+; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; AVX2-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,2,4,6,8,10,12,14,u,u,u,u,u,u,u,u]
+; AVX2-NEXT: vmovq %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: trunc_v8i32_to_v8i8:
; AVX512F: # BB#0:
}
define void @shuffle_v16i16_to_v4i16(<16 x i16>* %L, <4 x i16>* %S) nounwind {
-; AVX-LABEL: shuffle_v16i16_to_v4i16:
-; AVX: # BB#0:
-; AVX-NEXT: vmovdqa (%rdi), %ymm0
-; AVX-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
-; AVX-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,2,2,3,4,5,6,7]
-; AVX-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
-; AVX-NEXT: vpshuflw {{.*#+}} xmm0 = xmm0[0,2,2,3,4,5,6,7]
-; AVX-NEXT: vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
-; AVX-NEXT: vmovq %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: shuffle_v16i16_to_v4i16:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovdqa (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
+; AVX1-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,2,2,3,4,5,6,7]
+; AVX1-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
+; AVX1-NEXT: vpshuflw {{.*#+}} xmm0 = xmm0[0,2,2,3,4,5,6,7]
+; AVX1-NEXT: vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; AVX1-NEXT: vmovq %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v16i16_to_v4i16:
+; AVX2: # BB#0:
+; AVX2-NEXT: vmovdqa (%rdi), %ymm0
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm1[0,2,2,3]
+; AVX2-NEXT: vpshuflw {{.*#+}} xmm1 = xmm1[0,2,2,3,4,5,6,7]
+; AVX2-NEXT: vpshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
+; AVX2-NEXT: vpshuflw {{.*#+}} xmm0 = xmm0[0,2,2,3,4,5,6,7]
+; AVX2-NEXT: vpunpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
+; AVX2-NEXT: vmovq %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: shuffle_v16i16_to_v4i16:
; AVX512F: # BB#0:
}
define void @trunc_v4i64_to_v4i16(<16 x i16>* %L, <4 x i16>* %S) nounwind {
-; AVX-LABEL: trunc_v4i64_to_v4i16:
-; AVX: # BB#0:
-; AVX-NEXT: vpshufd {{.*#+}} ymm0 = mem[0,2,2,3,4,6,6,7]
-; AVX-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
-; AVX-NEXT: vmovq %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: trunc_v4i64_to_v4i16:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovaps (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vshufps {{.*#+}} xmm0 = xmm0[0,2],xmm1[0,2]
+; AVX1-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
+; AVX1-NEXT: vmovq %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: trunc_v4i64_to_v4i16:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpshufd {{.*#+}} ymm0 = mem[0,2,2,3,4,6,6,7]
+; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; AVX2-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15]
+; AVX2-NEXT: vmovq %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: trunc_v4i64_to_v4i16:
; AVX512F: # BB#0:
}
define void @shuffle_v32i8_to_v4i8(<32 x i8>* %L, <4 x i8>* %S) nounwind {
-; AVX-LABEL: shuffle_v32i8_to_v4i8:
-; AVX: # BB#0:
-; AVX-NEXT: vmovdqa (%rdi), %ymm0
-; AVX-NEXT: vextracti128 $1, %ymm0, %xmm1
-; AVX-NEXT: vmovdqa {{.*#+}} xmm2 = <0,8,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
-; AVX-NEXT: vpshufb %xmm2, %xmm1, %xmm1
-; AVX-NEXT: vpshufb %xmm2, %xmm0, %xmm0
-; AVX-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
-; AVX-NEXT: vmovd %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: shuffle_v32i8_to_v4i8:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovdqa (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vmovdqa {{.*#+}} xmm2 = <0,8,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX1-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX1-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
+; AVX1-NEXT: vmovd %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v32i8_to_v4i8:
+; AVX2: # BB#0:
+; AVX2-NEXT: vmovdqa (%rdi), %ymm0
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vmovdqa {{.*#+}} xmm2 = <0,8,u,u,u,u,u,u,u,u,u,u,u,u,u,u>
+; AVX2-NEXT: vpshufb %xmm2, %xmm1, %xmm1
+; AVX2-NEXT: vpshufb %xmm2, %xmm0, %xmm0
+; AVX2-NEXT: vpunpcklwd {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3]
+; AVX2-NEXT: vmovd %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: shuffle_v32i8_to_v4i8:
; AVX512F: # BB#0:
}
define void @trunc_v4i64_to_v4i8(<32 x i8>* %L, <4 x i8>* %S) nounwind {
-; AVX-LABEL: trunc_v4i64_to_v4i8:
-; AVX: # BB#0:
-; AVX-NEXT: vpshufd {{.*#+}} ymm0 = mem[0,2,2,3,4,6,6,7]
-; AVX-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
-; AVX-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u]
-; AVX-NEXT: vmovd %xmm0, (%rsi)
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-LABEL: trunc_v4i64_to_v4i8:
+; AVX1: # BB#0:
+; AVX1-NEXT: vmovaps (%rdi), %ymm0
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-NEXT: vshufps {{.*#+}} xmm0 = xmm0[0,2],xmm1[0,2]
+; AVX1-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u]
+; AVX1-NEXT: vmovd %xmm0, (%rsi)
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: trunc_v4i64_to_v4i8:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpshufd {{.*#+}} ymm0 = mem[0,2,2,3,4,6,6,7]
+; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; AVX2-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,4,8,12,u,u,u,u,u,u,u,u,u,u,u,u]
+; AVX2-NEXT: vmovd %xmm0, (%rsi)
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512F-LABEL: trunc_v4i64_to_v4i8:
; AVX512F: # BB#0: