; Those test cases exerce the latter.
define void @isel_crash_16b(i8* %cV_R.addr) {
-; X32-AVX2-LABEL: isel_crash_16b:
-; X32-AVX2: ## BB#0: ## %eintry
-; X32-AVX2-NEXT: subl $60, %esp
-; X32-AVX2-NEXT: Lcfi0:
-; X32-AVX2-NEXT: .cfi_def_cfa_offset 64
-; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
-; X32-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X32-AVX2-NEXT: vmovaps %xmm0, (%esp)
-; X32-AVX2-NEXT: vpbroadcastb (%eax), %xmm1
-; X32-AVX2-NEXT: vmovaps %xmm0, {{[0-9]+}}(%esp)
-; X32-AVX2-NEXT: vmovdqa %xmm1, {{[0-9]+}}(%esp)
-; X32-AVX2-NEXT: addl $60, %esp
-; X32-AVX2-NEXT: retl
-;
-; X64-AVX2-LABEL: isel_crash_16b:
-; X64-AVX2: ## BB#0: ## %eintry
-; X64-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X64-AVX2-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
-; X64-AVX2-NEXT: movb (%rdi), %al
-; X64-AVX2-NEXT: vmovd %eax, %xmm1
-; X64-AVX2-NEXT: vpbroadcastb %xmm1, %xmm1
-; X64-AVX2-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
-; X64-AVX2-NEXT: vmovdqa %xmm1, -{{[0-9]+}}(%rsp)
-; X64-AVX2-NEXT: retq
-;
-; X32-AVX512VL-LABEL: isel_crash_16b:
-; X32-AVX512VL: ## BB#0: ## %eintry
-; X32-AVX512VL-NEXT: subl $60, %esp
-; X32-AVX512VL-NEXT: Lcfi0:
-; X32-AVX512VL-NEXT: .cfi_def_cfa_offset 64
-; X32-AVX512VL-NEXT: movl {{[0-9]+}}(%esp), %eax
-; X32-AVX512VL-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X32-AVX512VL-NEXT: vmovaps %xmm0, (%esp)
-; X32-AVX512VL-NEXT: vpbroadcastb (%eax), %xmm1
-; X32-AVX512VL-NEXT: vmovaps %xmm0, {{[0-9]+}}(%esp)
-; X32-AVX512VL-NEXT: vmovdqa %xmm1, {{[0-9]+}}(%esp)
-; X32-AVX512VL-NEXT: addl $60, %esp
-; X32-AVX512VL-NEXT: retl
+; X32-LABEL: isel_crash_16b:
+; X32: ## BB#0: ## %eintry
+; X32-NEXT: subl $60, %esp
+; X32-NEXT: Lcfi0:
+; X32-NEXT: .cfi_def_cfa_offset 64
+; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
+; X32-NEXT: vmovaps %xmm0, (%esp)
+; X32-NEXT: vpbroadcastb (%eax), %xmm1
+; X32-NEXT: vmovaps %xmm0, {{[0-9]+}}(%esp)
+; X32-NEXT: vmovdqa %xmm1, {{[0-9]+}}(%esp)
+; X32-NEXT: addl $60, %esp
+; X32-NEXT: retl
;
-; X64-AVX512VL-LABEL: isel_crash_16b:
-; X64-AVX512VL: ## BB#0: ## %eintry
-; X64-AVX512VL-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X64-AVX512VL-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
-; X64-AVX512VL-NEXT: movb (%rdi), %al
-; X64-AVX512VL-NEXT: vmovd %eax, %xmm1
-; X64-AVX512VL-NEXT: vpbroadcastb %xmm1, %xmm1
-; X64-AVX512VL-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
-; X64-AVX512VL-NEXT: vmovdqa %xmm1, -{{[0-9]+}}(%rsp)
-; X64-AVX512VL-NEXT: retq
+; X64-LABEL: isel_crash_16b:
+; X64: ## BB#0: ## %eintry
+; X64-NEXT: vxorps %xmm0, %xmm0, %xmm0
+; X64-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
+; X64-NEXT: movb (%rdi), %al
+; X64-NEXT: vmovd %eax, %xmm1
+; X64-NEXT: vpbroadcastb %xmm1, %xmm1
+; X64-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
+; X64-NEXT: vmovdqa %xmm1, -{{[0-9]+}}(%rsp)
+; X64-NEXT: retq
eintry:
%__a.addr.i = alloca <2 x i64>, align 16
%__b.addr.i = alloca <2 x i64>, align 16
}
define void @isel_crash_8w(i16* %cV_R.addr) {
-; X32-AVX2-LABEL: isel_crash_8w:
-; X32-AVX2: ## BB#0: ## %entry
-; X32-AVX2-NEXT: subl $60, %esp
-; X32-AVX2-NEXT: Lcfi4:
-; X32-AVX2-NEXT: .cfi_def_cfa_offset 64
-; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
-; X32-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X32-AVX2-NEXT: vmovaps %xmm0, (%esp)
-; X32-AVX2-NEXT: vpbroadcastw (%eax), %xmm1
-; X32-AVX2-NEXT: vmovaps %xmm0, {{[0-9]+}}(%esp)
-; X32-AVX2-NEXT: vmovdqa %xmm1, {{[0-9]+}}(%esp)
-; X32-AVX2-NEXT: addl $60, %esp
-; X32-AVX2-NEXT: retl
-;
-; X64-AVX2-LABEL: isel_crash_8w:
-; X64-AVX2: ## BB#0: ## %entry
-; X64-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X64-AVX2-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
-; X64-AVX2-NEXT: movw (%rdi), %ax
-; X64-AVX2-NEXT: vmovd %eax, %xmm1
-; X64-AVX2-NEXT: vpbroadcastw %xmm1, %xmm1
-; X64-AVX2-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
-; X64-AVX2-NEXT: vmovdqa %xmm1, -{{[0-9]+}}(%rsp)
-; X64-AVX2-NEXT: retq
-;
-; X32-AVX512VL-LABEL: isel_crash_8w:
-; X32-AVX512VL: ## BB#0: ## %entry
-; X32-AVX512VL-NEXT: subl $60, %esp
-; X32-AVX512VL-NEXT: Lcfi4:
-; X32-AVX512VL-NEXT: .cfi_def_cfa_offset 64
-; X32-AVX512VL-NEXT: movl {{[0-9]+}}(%esp), %eax
-; X32-AVX512VL-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X32-AVX512VL-NEXT: vmovaps %xmm0, (%esp)
-; X32-AVX512VL-NEXT: vpbroadcastw (%eax), %xmm1
-; X32-AVX512VL-NEXT: vmovaps %xmm0, {{[0-9]+}}(%esp)
-; X32-AVX512VL-NEXT: vmovdqa %xmm1, {{[0-9]+}}(%esp)
-; X32-AVX512VL-NEXT: addl $60, %esp
-; X32-AVX512VL-NEXT: retl
+; X32-LABEL: isel_crash_8w:
+; X32: ## BB#0: ## %entry
+; X32-NEXT: subl $60, %esp
+; X32-NEXT: Lcfi4:
+; X32-NEXT: .cfi_def_cfa_offset 64
+; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
+; X32-NEXT: vmovaps %xmm0, (%esp)
+; X32-NEXT: vpbroadcastw (%eax), %xmm1
+; X32-NEXT: vmovaps %xmm0, {{[0-9]+}}(%esp)
+; X32-NEXT: vmovdqa %xmm1, {{[0-9]+}}(%esp)
+; X32-NEXT: addl $60, %esp
+; X32-NEXT: retl
;
-; X64-AVX512VL-LABEL: isel_crash_8w:
-; X64-AVX512VL: ## BB#0: ## %entry
-; X64-AVX512VL-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X64-AVX512VL-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
-; X64-AVX512VL-NEXT: movw (%rdi), %ax
-; X64-AVX512VL-NEXT: vmovd %eax, %xmm1
-; X64-AVX512VL-NEXT: vpbroadcastw %xmm1, %xmm1
-; X64-AVX512VL-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
-; X64-AVX512VL-NEXT: vmovdqa %xmm1, -{{[0-9]+}}(%rsp)
-; X64-AVX512VL-NEXT: retq
+; X64-LABEL: isel_crash_8w:
+; X64: ## BB#0: ## %entry
+; X64-NEXT: vxorps %xmm0, %xmm0, %xmm0
+; X64-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
+; X64-NEXT: movw (%rdi), %ax
+; X64-NEXT: vmovd %eax, %xmm1
+; X64-NEXT: vpbroadcastw %xmm1, %xmm1
+; X64-NEXT: vmovaps %xmm0, -{{[0-9]+}}(%rsp)
+; X64-NEXT: vmovdqa %xmm1, -{{[0-9]+}}(%rsp)
+; X64-NEXT: retq
entry:
%__a.addr.i = alloca <2 x i64>, align 16
%__b.addr.i = alloca <2 x i64>, align 16
}
define void @isel_crash_2q(i64* %cV_R.addr) {
-; X32-AVX2-LABEL: isel_crash_2q:
-; X32-AVX2: ## BB#0: ## %entry
-; X32-AVX2-NEXT: subl $60, %esp
-; X32-AVX2-NEXT: Lcfi12:
-; X32-AVX2-NEXT: .cfi_def_cfa_offset 64
-; X32-AVX2-NEXT: movl {{[0-9]+}}(%esp), %eax
-; X32-AVX2-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X32-AVX2-NEXT: vmovaps %xmm0, (%esp)
-; X32-AVX2-NEXT: movl (%eax), %ecx
-; X32-AVX2-NEXT: movl 4(%eax), %eax
-; X32-AVX2-NEXT: vmovd %ecx, %xmm1
-; X32-AVX2-NEXT: vpinsrd $1, %eax, %xmm1, %xmm1
-; X32-AVX2-NEXT: vpinsrd $2, %ecx, %xmm1, %xmm1
-; X32-AVX2-NEXT: vpinsrd $3, %eax, %xmm1, %xmm1
-; X32-AVX2-NEXT: vmovaps %xmm0, {{[0-9]+}}(%esp)
-; X32-AVX2-NEXT: vmovdqa %xmm1, {{[0-9]+}}(%esp)
-; X32-AVX2-NEXT: addl $60, %esp
-; X32-AVX2-NEXT: retl
+; X32-LABEL: isel_crash_2q:
+; X32: ## BB#0: ## %entry
+; X32-NEXT: subl $60, %esp
+; X32-NEXT: Lcfi12:
+; X32-NEXT: .cfi_def_cfa_offset 64
+; X32-NEXT: movl {{[0-9]+}}(%esp), %eax
+; X32-NEXT: vxorps %xmm0, %xmm0, %xmm0
+; X32-NEXT: vmovaps %xmm0, (%esp)
+; X32-NEXT: movl (%eax), %ecx
+; X32-NEXT: movl 4(%eax), %eax
+; X32-NEXT: vmovd %ecx, %xmm1
+; X32-NEXT: vpinsrd $1, %eax, %xmm1, %xmm1
+; X32-NEXT: vpinsrd $2, %ecx, %xmm1, %xmm1
+; X32-NEXT: vpinsrd $3, %eax, %xmm1, %xmm1
+; X32-NEXT: vmovaps %xmm0, {{[0-9]+}}(%esp)
+; X32-NEXT: vmovdqa %xmm1, {{[0-9]+}}(%esp)
+; X32-NEXT: addl $60, %esp
+; X32-NEXT: retl
;
; X64-AVX2-LABEL: isel_crash_2q:
; X64-AVX2: ## BB#0: ## %entry
; X64-AVX2-NEXT: vmovdqa %xmm1, -{{[0-9]+}}(%rsp)
; X64-AVX2-NEXT: retq
;
-; X32-AVX512VL-LABEL: isel_crash_2q:
-; X32-AVX512VL: ## BB#0: ## %entry
-; X32-AVX512VL-NEXT: subl $60, %esp
-; X32-AVX512VL-NEXT: Lcfi12:
-; X32-AVX512VL-NEXT: .cfi_def_cfa_offset 64
-; X32-AVX512VL-NEXT: movl {{[0-9]+}}(%esp), %eax
-; X32-AVX512VL-NEXT: vxorps %xmm0, %xmm0, %xmm0
-; X32-AVX512VL-NEXT: vmovaps %xmm0, (%esp)
-; X32-AVX512VL-NEXT: movl (%eax), %ecx
-; X32-AVX512VL-NEXT: movl 4(%eax), %eax
-; X32-AVX512VL-NEXT: vmovd %ecx, %xmm1
-; X32-AVX512VL-NEXT: vpinsrd $1, %eax, %xmm1, %xmm1
-; X32-AVX512VL-NEXT: vpinsrd $2, %ecx, %xmm1, %xmm1
-; X32-AVX512VL-NEXT: vpinsrd $3, %eax, %xmm1, %xmm1
-; X32-AVX512VL-NEXT: vmovaps %xmm0, {{[0-9]+}}(%esp)
-; X32-AVX512VL-NEXT: vmovdqa %xmm1, {{[0-9]+}}(%esp)
-; X32-AVX512VL-NEXT: addl $60, %esp
-; X32-AVX512VL-NEXT: retl
-;
; X64-AVX512VL-LABEL: isel_crash_2q:
; X64-AVX512VL: ## BB#0: ## %entry
; X64-AVX512VL-NEXT: vxorps %xmm0, %xmm0, %xmm0
@gb4 = global <8 x i64> zeroinitializer, align 8
define void @fallback_broadcast_v4i64_to_v8i64(<4 x i64> %a, <8 x i64> %b) {
+; X32-AVX1-LABEL: fallback_broadcast_v4i64_to_v8i64:
+; X32-AVX1: ## BB#0: ## %entry
+; X32-AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
+; X32-AVX1-NEXT: vmovaps {{.*#+}} ymm4 = [1,0,2,0,3,0,4,0]
+; X32-AVX1-NEXT: vextractf128 $1, %ymm4, %xmm5
+; X32-AVX1-NEXT: vpaddq %xmm5, %xmm3, %xmm3
+; X32-AVX1-NEXT: vpaddq %xmm4, %xmm0, %xmm0
+; X32-AVX1-NEXT: vinsertf128 $1, %xmm3, %ymm0, %ymm0
+; X32-AVX1-NEXT: vextractf128 $1, %ymm2, %xmm3
+; X32-AVX1-NEXT: vpaddq %xmm5, %xmm3, %xmm3
+; X32-AVX1-NEXT: vpaddq %xmm4, %xmm2, %xmm2
+; X32-AVX1-NEXT: vinsertf128 $1, %xmm3, %ymm2, %ymm2
+; X32-AVX1-NEXT: vextractf128 $1, %ymm1, %xmm3
+; X32-AVX1-NEXT: vpaddq %xmm5, %xmm3, %xmm3
+; X32-AVX1-NEXT: vpaddq %xmm4, %xmm1, %xmm1
+; X32-AVX1-NEXT: vinsertf128 $1, %xmm3, %ymm1, %ymm1
+; X32-AVX1-NEXT: vandps %ymm4, %ymm1, %ymm1
+; X32-AVX1-NEXT: vandps %ymm4, %ymm2, %ymm2
+; X32-AVX1-NEXT: vmovups %ymm0, _ga4
+; X32-AVX1-NEXT: vmovups %ymm2, _gb4+32
+; X32-AVX1-NEXT: vmovups %ymm1, _gb4
+; X32-AVX1-NEXT: vzeroupper
+; X32-AVX1-NEXT: retl
+;
+; X32-AVX2-LABEL: fallback_broadcast_v4i64_to_v8i64:
+; X32-AVX2: ## BB#0: ## %entry
+; X32-AVX2-NEXT: vmovdqa {{.*#+}} ymm3 = [1,0,2,0,3,0,4,0]
+; X32-AVX2-NEXT: vpaddq %ymm3, %ymm0, %ymm0
+; X32-AVX2-NEXT: vpaddq %ymm3, %ymm2, %ymm2
+; X32-AVX2-NEXT: vpaddq %ymm3, %ymm1, %ymm1
+; X32-AVX2-NEXT: vpand %ymm3, %ymm1, %ymm1
+; X32-AVX2-NEXT: vpand %ymm3, %ymm2, %ymm2
+; X32-AVX2-NEXT: vmovdqu %ymm0, _ga4
+; X32-AVX2-NEXT: vmovdqu %ymm2, _gb4+32
+; X32-AVX2-NEXT: vmovdqu %ymm1, _gb4
+; X32-AVX2-NEXT: vzeroupper
+; X32-AVX2-NEXT: retl
+;
; X32-AVX512-LABEL: fallback_broadcast_v4i64_to_v8i64:
; X32-AVX512: ## BB#0: ## %entry
; X32-AVX512-NEXT: vpaddq LCPI26_0, %ymm0, %ymm0
; X32-AVX512-NEXT: vmovdqu64 %zmm1, _gb4
; X32-AVX512-NEXT: retl
;
+; X64-AVX1-LABEL: fallback_broadcast_v4i64_to_v8i64:
+; X64-AVX1: ## BB#0: ## %entry
+; X64-AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
+; X64-AVX1-NEXT: vmovdqa {{.*#+}} xmm4 = [3,4]
+; X64-AVX1-NEXT: vpaddq %xmm4, %xmm3, %xmm3
+; X64-AVX1-NEXT: vmovdqa {{.*#+}} xmm5 = [1,2]
+; X64-AVX1-NEXT: vpaddq %xmm5, %xmm0, %xmm0
+; X64-AVX1-NEXT: vinsertf128 $1, %xmm3, %ymm0, %ymm0
+; X64-AVX1-NEXT: vmovaps {{.*#+}} ymm3 = [1,2,3,4]
+; X64-AVX1-NEXT: vextractf128 $1, %ymm2, %xmm6
+; X64-AVX1-NEXT: vpaddq %xmm4, %xmm6, %xmm6
+; X64-AVX1-NEXT: vpaddq %xmm5, %xmm2, %xmm2
+; X64-AVX1-NEXT: vinsertf128 $1, %xmm6, %ymm2, %ymm2
+; X64-AVX1-NEXT: vextractf128 $1, %ymm1, %xmm6
+; X64-AVX1-NEXT: vpaddq %xmm4, %xmm6, %xmm4
+; X64-AVX1-NEXT: vpaddq %xmm5, %xmm1, %xmm1
+; X64-AVX1-NEXT: vinsertf128 $1, %xmm4, %ymm1, %ymm1
+; X64-AVX1-NEXT: vandps %ymm3, %ymm1, %ymm1
+; X64-AVX1-NEXT: vandps %ymm3, %ymm2, %ymm2
+; X64-AVX1-NEXT: vmovups %ymm0, {{.*}}(%rip)
+; X64-AVX1-NEXT: vmovups %ymm2, _gb4+{{.*}}(%rip)
+; X64-AVX1-NEXT: vmovups %ymm1, {{.*}}(%rip)
+; X64-AVX1-NEXT: vzeroupper
+; X64-AVX1-NEXT: retq
+;
+; X64-AVX2-LABEL: fallback_broadcast_v4i64_to_v8i64:
+; X64-AVX2: ## BB#0: ## %entry
+; X64-AVX2-NEXT: vmovdqa {{.*#+}} ymm3 = [1,2,3,4]
+; X64-AVX2-NEXT: vpaddq %ymm3, %ymm0, %ymm0
+; X64-AVX2-NEXT: vpaddq %ymm3, %ymm2, %ymm2
+; X64-AVX2-NEXT: vpaddq %ymm3, %ymm1, %ymm1
+; X64-AVX2-NEXT: vpand %ymm3, %ymm1, %ymm1
+; X64-AVX2-NEXT: vpand %ymm3, %ymm2, %ymm2
+; X64-AVX2-NEXT: vmovdqu %ymm0, {{.*}}(%rip)
+; X64-AVX2-NEXT: vmovdqu %ymm2, _gb4+{{.*}}(%rip)
+; X64-AVX2-NEXT: vmovdqu %ymm1, {{.*}}(%rip)
+; X64-AVX2-NEXT: vzeroupper
+; X64-AVX2-NEXT: retq
+;
; X64-AVX512-LABEL: fallback_broadcast_v4i64_to_v8i64:
; X64-AVX512: ## BB#0: ## %entry
; X64-AVX512-NEXT: vmovdqa {{.*#+}} ymm2 = [1,2,3,4]
@gb2 = global <8 x double> zeroinitializer, align 8
define void @fallback_broadcast_v4f64_to_v8f64(<4 x double> %a, <8 x double> %b) {
+; X32-AVX-LABEL: fallback_broadcast_v4f64_to_v8f64:
+; X32-AVX: ## BB#0: ## %entry
+; X32-AVX-NEXT: vmovapd {{.*#+}} ymm3 = [1.000000e+00,2.000000e+00,3.000000e+00,4.000000e+00]
+; X32-AVX-NEXT: vaddpd %ymm3, %ymm0, %ymm0
+; X32-AVX-NEXT: vaddpd %ymm3, %ymm2, %ymm2
+; X32-AVX-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; X32-AVX-NEXT: vdivpd %ymm3, %ymm1, %ymm1
+; X32-AVX-NEXT: vdivpd %ymm3, %ymm2, %ymm2
+; X32-AVX-NEXT: vmovupd %ymm0, _ga2
+; X32-AVX-NEXT: vmovupd %ymm2, _gb2+32
+; X32-AVX-NEXT: vmovupd %ymm1, _gb2
+; X32-AVX-NEXT: vzeroupper
+; X32-AVX-NEXT: retl
+;
; X32-AVX512-LABEL: fallback_broadcast_v4f64_to_v8f64:
; X32-AVX512: ## BB#0: ## %entry
; X32-AVX512-NEXT: vmovapd {{.*#+}} ymm2 = [1.000000e+00,2.000000e+00,3.000000e+00,4.000000e+00]
; X32-AVX512-NEXT: vmovupd %zmm1, _gb2
; X32-AVX512-NEXT: retl
;
+; X64-AVX-LABEL: fallback_broadcast_v4f64_to_v8f64:
+; X64-AVX: ## BB#0: ## %entry
+; X64-AVX-NEXT: vmovapd {{.*#+}} ymm3 = [1.000000e+00,2.000000e+00,3.000000e+00,4.000000e+00]
+; X64-AVX-NEXT: vaddpd %ymm3, %ymm0, %ymm0
+; X64-AVX-NEXT: vaddpd %ymm3, %ymm2, %ymm2
+; X64-AVX-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; X64-AVX-NEXT: vdivpd %ymm3, %ymm1, %ymm1
+; X64-AVX-NEXT: vdivpd %ymm3, %ymm2, %ymm2
+; X64-AVX-NEXT: vmovupd %ymm0, {{.*}}(%rip)
+; X64-AVX-NEXT: vmovupd %ymm2, _gb2+{{.*}}(%rip)
+; X64-AVX-NEXT: vmovupd %ymm1, {{.*}}(%rip)
+; X64-AVX-NEXT: vzeroupper
+; X64-AVX-NEXT: retq
+;
; X64-AVX512-LABEL: fallback_broadcast_v4f64_to_v8f64:
; X64-AVX512: ## BB#0: ## %entry
; X64-AVX512-NEXT: vmovapd {{.*#+}} ymm2 = [1.000000e+00,2.000000e+00,3.000000e+00,4.000000e+00]
; AVX1-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1,2],ymm0[3,4,5,6,7]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_00040000:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,4,0,0,0,0]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_00040000:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,4,0,0,0,0]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_00040000:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,4,0,0,0,0]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 0, i32 4, i32 0, i32 0, i32 0, i32 0>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,0,1,0,4,4,4,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_00500000:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,5,0,0,0,0,0]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_00500000:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,5,0,0,0,0,0]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_00500000:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,5,0,0,0,0,0]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 5, i32 0, i32 0, i32 0, i32 0, i32 0>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,2,0,0,4,4,4,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_06000000:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,6,0,0,0,0,0,0]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_06000000:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,6,0,0,0,0,0,0]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_06000000:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,6,0,0,0,0,0,0]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 6, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_00112233:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,1,1,2,2,3,3]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_00112233:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,1,1,2,2,3,3]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_00112233:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,1,1,2,2,3,3]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 1, i32 1, i32 2, i32 2, i32 3, i32 3>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,0,0,1,5,4,4,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_00015444:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,1,5,4,4,4]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_00015444:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,1,5,4,4,4]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_00015444:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,0,1,5,4,4,4]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 0, i32 1, i32 5, i32 4, i32 4, i32 4>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,0,2,0,4,6,4,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_00204644:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,2,0,4,6,4,4]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_00204644:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,2,0,4,6,4,4]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_00204644:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,2,0,4,6,4,4]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 2, i32 0, i32 4, i32 6, i32 4, i32 4>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,3,0,0,4,4,7,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_03004474:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,3,0,0,4,4,7,4]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_03004474:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,3,0,0,4,4,7,4]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_03004474:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,3,0,0,4,4,7,4]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 3, i32 0, i32 0, i32 4, i32 4, i32 7, i32 4>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[1,0,0,0,4,4,4,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_10004444:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [1,0,0,0,4,4,4,4]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_10004444:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [1,0,0,0,4,4,4,4]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_10004444:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [1,0,0,0,4,4,4,4]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 1, i32 0, i32 0, i32 0, i32 4, i32 4, i32 4, i32 4>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[2,2,0,0,6,4,4,6]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_22006446:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [2,2,0,0,6,4,4,6]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_22006446:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [2,2,0,0,6,4,4,6]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_22006446:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [2,2,0,0,6,4,4,6]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 2, i32 2, i32 0, i32 0, i32 6, i32 4, i32 4, i32 6>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[3,3,3,0,7,4,7,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_33307474:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [3,3,3,0,7,4,7,4]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_33307474:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [3,3,3,0,7,4,7,4]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_33307474:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [3,3,3,0,7,4,7,4]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 3, i32 3, i32 3, i32 0, i32 7, i32 4, i32 7, i32 4>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[3,2,1,0,4,5,6,7]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_32104567:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [3,2,1,0,4,5,6,7]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_32104567:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [3,2,1,0,4,5,6,7]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_32104567:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [3,2,1,0,4,5,6,7]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 3, i32 2, i32 1, i32 0, i32 4, i32 5, i32 6, i32 7>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,0,2,3,6,7,4,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_00236744:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,2,3,6,7,4,4]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_00236744:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,2,3,6,7,4,4]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_00236744:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,2,3,6,7,4,4]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 2, i32 3, i32 6, i32 7, i32 4, i32 4>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,0,2,2,6,6,4,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_00226644:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,2,2,6,6,4,4]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_00226644:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,2,2,6,6,4,4]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_00226644:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,0,2,2,6,6,4,4]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 2, i32 2, i32 6, i32 6, i32 4, i32 4>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[1,0,3,2,4,5,6,7]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_10324567:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [1,0,3,2,4,5,6,7]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_10324567:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [1,0,3,2,4,5,6,7]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_10324567:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [1,0,3,2,4,5,6,7]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 1, i32 0, i32 3, i32 2, i32 4, i32 5, i32 6, i32 7>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[1,1,3,3,4,5,6,7]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_11334567:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [1,1,3,3,4,5,6,7]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_11334567:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [1,1,3,3,4,5,6,7]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_11334567:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [1,1,3,3,4,5,6,7]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 1, i32 1, i32 3, i32 3, i32 4, i32 5, i32 6, i32 7>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,1,2,3,5,4,6,7]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_01235467:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,2,3,5,4,6,7]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_01235467:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,2,3,5,4,6,7]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_01235467:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,2,3,5,4,6,7]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 5, i32 4, i32 6, i32 7>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,1,2,3,5,4,6,6]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_01235466:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,2,3,5,4,6,6]
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_01235466:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,2,3,5,4,6,6]
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_01235466:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = [0,1,2,3,5,4,6,6]
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 5, i32 4, i32 6, i32 6>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,0,2,u,6,u,4,4]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_002u6u44:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,2,u,6,u,4,4>
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_002u6u44:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,2,u,6,u,4,4>
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_002u6u44:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,2,u,6,u,4,4>
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 2, i32 undef, i32 6, i32 undef, i32 4, i32 4>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,0,u,u,6,6,u,u]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_00uu66uu:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,u,u,6,6,u,u>
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_00uu66uu:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,u,u,6,6,u,u>
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_00uu66uu:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,0,u,u,6,6,u,u>
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 0, i32 undef, i32 undef, i32 6, i32 6, i32 undef, i32 undef>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[1,0,3,2,4,5,u,u]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_103245uu:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = <1,0,3,2,4,5,u,u>
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_103245uu:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <1,0,3,2,4,5,u,u>
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_103245uu:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <1,0,3,2,4,5,u,u>
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 1, i32 0, i32 3, i32 2, i32 4, i32 5, i32 undef, i32 undef>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[1,1,3,3,u,u,6,7]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_1133uu67:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = <1,1,3,3,u,u,6,7>
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_1133uu67:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <1,1,3,3,u,u,6,7>
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_1133uu67:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <1,1,3,3,u,u,6,7>
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 1, i32 1, i32 3, i32 3, i32 undef, i32 undef, i32 6, i32 7>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[0,u,u,3,5,4,u,u]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_0uu354uu:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = <0,u,u,3,5,4,u,u>
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_0uu354uu:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,u,u,3,5,4,u,u>
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_0uu354uu:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <0,u,u,3,5,4,u,u>
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 0, i32 undef, i32 undef, i32 3, i32 5, i32 4, i32 undef, i32 undef>
ret <8 x i32> %shuffle
}
; AVX1-NEXT: vpermilps {{.*#+}} ymm0 = ymm0[u,u,u,3,u,u,6,6]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v8i32_uuu3uu66:
-; AVX2: # BB#0:
-; AVX2-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,3,u,u,6,6>
-; AVX2-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v8i32_uuu3uu66:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,3,u,u,6,6>
-; AVX512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; AVX2OR512VL-LABEL: shuffle_v8i32_uuu3uu66:
+; AVX2OR512VL: # BB#0:
+; AVX2OR512VL-NEXT: vmovdqa {{.*#+}} ymm1 = <u,u,u,3,u,u,6,6>
+; AVX2OR512VL-NEXT: vpermd %ymm0, %ymm1, %ymm0
+; AVX2OR512VL-NEXT: retq
%shuffle = shufflevector <8 x i32> %a, <8 x i32> %b, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 3, i32 undef, i32 undef, i32 6, i32 6>
ret <8 x i32> %shuffle
}