; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
-; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE41
-; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
-; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
-; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512BW
-; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512VL
+; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=SSE,SSE2
+; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefixes=SSE,SSE41
+; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-SLOW
+; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+avx,+fast-hops | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-FAST
+; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=AVX,AVX2
+; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=AVX512,AVX512BW
+; RUN: llc < %s -x86-experimental-vector-widening-legalization -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefixes=AVX512,AVX512VL
;
; vXi64
; SSE-NEXT: movd %xmm1, %eax
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v2i32:
-; AVX: # %bb.0:
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovd %xmm0, %eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v2i32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v2i32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v2i32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovd %xmm0, %eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v2i32:
; AVX512: # %bb.0:
; SSE-NEXT: movd %xmm0, %eax
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v4i32:
-; AVX: # %bb.0:
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovd %xmm0, %eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4i32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4i32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4i32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX2-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovd %xmm0, %eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4i32:
; AVX512: # %bb.0:
; SSE-NEXT: movd %xmm0, %eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v8i32:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8i32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8i32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v8i32:
; AVX2: # %bb.0:
; SSE-NEXT: movd %xmm1, %eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v16i32:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
-; AVX1-NEXT: vpaddd %xmm2, %xmm3, %xmm2
-; AVX1-NEXT: vpaddd %xmm2, %xmm1, %xmm1
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16i32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm2
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-SLOW-NEXT: vpaddd %xmm2, %xmm3, %xmm2
+; AVX1-SLOW-NEXT: vpaddd %xmm2, %xmm1, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16i32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm2
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-FAST-NEXT: vpaddd %xmm2, %xmm3, %xmm2
+; AVX1-FAST-NEXT: vpaddd %xmm2, %xmm1, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v16i32:
; AVX2: # %bb.0:
; SSE-NEXT: movd %xmm1, %eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v32i32:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vpaddd %xmm3, %xmm1, %xmm4
-; AVX1-NEXT: vextractf128 $1, %ymm3, %xmm3
-; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
-; AVX1-NEXT: vpaddd %xmm3, %xmm1, %xmm1
-; AVX1-NEXT: vextractf128 $1, %ymm2, %xmm3
-; AVX1-NEXT: vpaddd %xmm1, %xmm3, %xmm1
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
-; AVX1-NEXT: vpaddd %xmm1, %xmm3, %xmm1
-; AVX1-NEXT: vpaddd %xmm4, %xmm2, %xmm2
-; AVX1-NEXT: vpaddd %xmm1, %xmm2, %xmm1
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v32i32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpaddd %xmm3, %xmm1, %xmm4
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm3, %xmm3
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm3, %xmm1, %xmm1
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm2, %xmm3
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm3, %xmm1
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm3, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm4, %xmm2, %xmm2
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm2, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v32i32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpaddd %xmm3, %xmm1, %xmm4
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm3, %xmm3
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm3, %xmm1, %xmm1
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm2, %xmm3
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm3, %xmm1
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm3, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm4, %xmm2, %xmm2
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm2, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v32i32:
; AVX2: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v2i16:
-; AVX: # %bb.0:
-; AVX-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovd %xmm0, %eax
-; AVX-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v2i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v2i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v2i16:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovd %xmm0, %eax
+; AVX2-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v2i16:
; AVX512: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v4i16:
-; AVX: # %bb.0:
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovd %xmm0, %eax
-; AVX-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4i16:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovd %xmm0, %eax
+; AVX2-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4i16:
; AVX512: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v8i16:
-; AVX: # %bb.0:
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovd %xmm0, %eax
-; AVX-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v8i16:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovd %xmm0, %eax
+; AVX2-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v8i16:
; AVX512: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v16i16:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v16i16:
; AVX2: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v32i16:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
-; AVX1-NEXT: vpaddw %xmm2, %xmm3, %xmm2
-; AVX1-NEXT: vpaddw %xmm2, %xmm1, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v32i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm2
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-SLOW-NEXT: vpaddw %xmm2, %xmm3, %xmm2
+; AVX1-SLOW-NEXT: vpaddw %xmm2, %xmm1, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v32i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm2
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-FAST-NEXT: vpaddw %xmm2, %xmm3, %xmm2
+; AVX1-FAST-NEXT: vpaddw %xmm2, %xmm1, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v32i16:
; AVX2: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v64i16:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vpaddw %xmm3, %xmm1, %xmm4
-; AVX1-NEXT: vextractf128 $1, %ymm3, %xmm3
-; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
-; AVX1-NEXT: vpaddw %xmm3, %xmm1, %xmm1
-; AVX1-NEXT: vextractf128 $1, %ymm2, %xmm3
-; AVX1-NEXT: vpaddw %xmm1, %xmm3, %xmm1
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
-; AVX1-NEXT: vpaddw %xmm1, %xmm3, %xmm1
-; AVX1-NEXT: vpaddw %xmm4, %xmm2, %xmm2
-; AVX1-NEXT: vpaddw %xmm1, %xmm2, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v64i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpaddw %xmm3, %xmm1, %xmm4
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm3, %xmm3
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm3, %xmm1, %xmm1
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm2, %xmm3
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm3, %xmm1
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm3, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm4, %xmm2, %xmm2
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm2, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v64i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpaddw %xmm3, %xmm1, %xmm4
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm3, %xmm3
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm3, %xmm1, %xmm1
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm2, %xmm3
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm3, %xmm1
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm3, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm4, %xmm2, %xmm2
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm2, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v64i16:
; AVX2: # %bb.0:
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE41
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512BW
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512VL
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=SSE,SSE2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefixes=SSE,SSE41
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-SLOW
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fast-hops | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-FAST
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=AVX,AVX2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=AVX512,AVX512BW
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefixes=AVX512,AVX512VL
;
; vXi64
; SSE-NEXT: movd %xmm0, %eax
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v4i32:
-; AVX: # %bb.0:
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovd %xmm0, %eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4i32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4i32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4i32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX2-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovd %xmm0, %eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4i32:
; AVX512: # %bb.0:
; SSE-NEXT: movd %xmm0, %eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v8i32:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8i32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8i32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v8i32:
; AVX2: # %bb.0:
; SSE-NEXT: movd %xmm1, %eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v16i32:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
-; AVX1-NEXT: vpaddd %xmm2, %xmm3, %xmm2
-; AVX1-NEXT: vpaddd %xmm2, %xmm1, %xmm1
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16i32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm2
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-SLOW-NEXT: vpaddd %xmm2, %xmm3, %xmm2
+; AVX1-SLOW-NEXT: vpaddd %xmm2, %xmm1, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16i32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm2
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-FAST-NEXT: vpaddd %xmm2, %xmm3, %xmm2
+; AVX1-FAST-NEXT: vpaddd %xmm2, %xmm1, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v16i32:
; AVX2: # %bb.0:
; SSE-NEXT: movd %xmm1, %eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v32i32:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vpaddd %xmm3, %xmm1, %xmm4
-; AVX1-NEXT: vextractf128 $1, %ymm3, %xmm3
-; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
-; AVX1-NEXT: vpaddd %xmm3, %xmm1, %xmm1
-; AVX1-NEXT: vextractf128 $1, %ymm2, %xmm3
-; AVX1-NEXT: vpaddd %xmm1, %xmm3, %xmm1
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
-; AVX1-NEXT: vpaddd %xmm1, %xmm3, %xmm1
-; AVX1-NEXT: vpaddd %xmm4, %xmm2, %xmm2
-; AVX1-NEXT: vpaddd %xmm1, %xmm2, %xmm1
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v32i32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpaddd %xmm3, %xmm1, %xmm4
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm3, %xmm3
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm3, %xmm1, %xmm1
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm2, %xmm3
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm3, %xmm1
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm3, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm4, %xmm2, %xmm2
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm2, %xmm1
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v32i32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpaddd %xmm3, %xmm1, %xmm4
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm3, %xmm3
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm3, %xmm1, %xmm1
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm2, %xmm3
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm3, %xmm1
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm3, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm4, %xmm2, %xmm2
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm2, %xmm1
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v32i32:
; AVX2: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v4i16:
-; AVX: # %bb.0:
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovd %xmm0, %eax
-; AVX-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4i16:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX2-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovd %xmm0, %eax
+; AVX2-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4i16:
; AVX512: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v8i16:
-; AVX: # %bb.0:
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovd %xmm0, %eax
-; AVX-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v8i16:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovd %xmm0, %eax
+; AVX2-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v8i16:
; AVX512: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v16i16:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v16i16:
; AVX2: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v32i16:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
-; AVX1-NEXT: vpaddw %xmm2, %xmm3, %xmm2
-; AVX1-NEXT: vpaddw %xmm2, %xmm1, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v32i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm2
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-SLOW-NEXT: vpaddw %xmm2, %xmm3, %xmm2
+; AVX1-SLOW-NEXT: vpaddw %xmm2, %xmm1, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v32i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm2
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-FAST-NEXT: vpaddw %xmm2, %xmm3, %xmm2
+; AVX1-FAST-NEXT: vpaddw %xmm2, %xmm1, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v32i16:
; AVX2: # %bb.0:
; SSE-NEXT: # kill: def $ax killed $ax killed $eax
; SSE-NEXT: retq
;
-; AVX1-LABEL: test_v64i16:
-; AVX1: # %bb.0:
-; AVX1-NEXT: vpaddw %xmm3, %xmm1, %xmm4
-; AVX1-NEXT: vextractf128 $1, %ymm3, %xmm3
-; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
-; AVX1-NEXT: vpaddw %xmm3, %xmm1, %xmm1
-; AVX1-NEXT: vextractf128 $1, %ymm2, %xmm3
-; AVX1-NEXT: vpaddw %xmm1, %xmm3, %xmm1
-; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
-; AVX1-NEXT: vpaddw %xmm1, %xmm3, %xmm1
-; AVX1-NEXT: vpaddw %xmm4, %xmm2, %xmm2
-; AVX1-NEXT: vpaddw %xmm1, %xmm2, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX1-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX1-NEXT: vmovd %xmm0, %eax
-; AVX1-NEXT: # kill: def $ax killed $ax killed $eax
-; AVX1-NEXT: vzeroupper
-; AVX1-NEXT: retq
+; AVX1-SLOW-LABEL: test_v64i16:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpaddw %xmm3, %xmm1, %xmm4
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm3, %xmm3
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm3, %xmm1, %xmm1
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm2, %xmm3
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm3, %xmm1
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm3, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm4, %xmm2, %xmm2
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm2, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovd %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v64i16:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpaddw %xmm3, %xmm1, %xmm4
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm3, %xmm3
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm3, %xmm1, %xmm1
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm2, %xmm3
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm3, %xmm1
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm3, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm4, %xmm2, %xmm2
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm2, %xmm1
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vmovd %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $ax killed $ax killed $eax
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
;
; AVX2-LABEL: test_v64i16:
; AVX2: # %bb.0:
; SSE41-NEXT: # kill: def $al killed $al killed $eax
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v4i8:
-; AVX: # %bb.0:
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX-NEXT: vpaddd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpextrb $0, %xmm0, %eax
-; AVX-NEXT: # kill: def $al killed $al killed $eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4i8:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpextrb $0, %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $al killed $al killed $eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4i8:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpextrb $0, %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $al killed $al killed $eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4i8:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX2-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT: vpaddd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpextrb $0, %xmm0, %eax
+; AVX2-NEXT: # kill: def $al killed $al killed $eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4i8:
; AVX512: # %bb.0:
; SSE41-NEXT: # kill: def $al killed $al killed $eax
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v8i8:
-; AVX: # %bb.0:
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpsrld $16, %xmm0, %xmm1
-; AVX-NEXT: vpaddw %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpextrb $0, %xmm0, %eax
-; AVX-NEXT: # kill: def $al killed $al killed $eax
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8i8:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX1-SLOW-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpextrb $0, %xmm0, %eax
+; AVX1-SLOW-NEXT: # kill: def $al killed $al killed $eax
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8i8:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-FAST-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vphaddw %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpextrb $0, %xmm0, %eax
+; AVX1-FAST-NEXT: # kill: def $al killed $al killed $eax
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v8i8:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpsrld $16, %xmm0, %xmm1
+; AVX2-NEXT: vpaddw %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpextrb $0, %xmm0, %eax
+; AVX2-NEXT: # kill: def $al killed $al killed $eax
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v8i8:
; AVX512: # %bb.0:
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE41
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512BW
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512VL
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=SSE,SSE2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefixes=SSE,SSE41
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-SLOW
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fast-hops | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-FAST
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=AVX,AVX2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=AVX512,AVX512BW
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefixes=AVX512,AVX512VL
;
; vXf32 (accum)
; SSE41-NEXT: addss %xmm1, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v2f32:
-; AVX: # %bb.0:
-; AVX-NEXT: vmovshdup {{.*#+}} xmm0 = xmm1[1,1,3,3]
-; AVX-NEXT: vaddss %xmm0, %xmm1, %xmm0
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v2f32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm0 = xmm1[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm0, %xmm1, %xmm0
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v2f32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vhaddps %xmm1, %xmm1, %xmm0
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v2f32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm0 = xmm1[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm0, %xmm1, %xmm0
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v2f32:
; AVX512: # %bb.0:
; SSE41-NEXT: addss %xmm2, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v4f32:
-; AVX: # %bb.0:
-; AVX-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
-; AVX-NEXT: vaddps %xmm0, %xmm1, %xmm0
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4f32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
+; AVX1-SLOW-NEXT: vaddps %xmm0, %xmm1, %xmm0
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4f32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
+; AVX1-FAST-NEXT: vaddps %xmm0, %xmm1, %xmm0
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4f32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
+; AVX2-NEXT: vaddps %xmm0, %xmm1, %xmm0
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4f32:
; AVX512: # %bb.0:
; SSE41-NEXT: addss %xmm2, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v8f32:
-; AVX: # %bb.0:
-; AVX-NEXT: vextractf128 $1, %ymm1, %xmm0
-; AVX-NEXT: vaddps %xmm0, %xmm1, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8f32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm0
+; AVX1-SLOW-NEXT: vaddps %xmm0, %xmm1, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8f32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm0
+; AVX1-FAST-NEXT: vaddps %xmm0, %xmm1, %xmm0
+; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v8f32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vextractf128 $1, %ymm1, %xmm0
+; AVX2-NEXT: vaddps %xmm0, %xmm1, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v8f32:
; AVX512: # %bb.0:
; SSE41-NEXT: addss %xmm2, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v16f32:
-; AVX: # %bb.0:
-; AVX-NEXT: vaddps %ymm2, %ymm1, %ymm0
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16f32:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vaddps %ymm2, %ymm1, %ymm0
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16f32:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vaddps %ymm2, %ymm1, %ymm0
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v16f32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vaddps %ymm2, %ymm1, %ymm0
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v16f32:
; AVX512: # %bb.0:
; SSE41-NEXT: addss %xmm1, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v2f32_zero:
-; AVX: # %bb.0:
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v2f32_zero:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v2f32_zero:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v2f32_zero:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v2f32_zero:
; AVX512: # %bb.0:
; SSE41-NEXT: movaps %xmm1, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v4f32_zero:
-; AVX: # %bb.0:
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4f32_zero:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4f32_zero:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4f32_zero:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4f32_zero:
; AVX512: # %bb.0:
; SSE41-NEXT: movaps %xmm1, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v8f32_zero:
-; AVX: # %bb.0:
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8f32_zero:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8f32_zero:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v8f32_zero:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v8f32_zero:
; AVX512: # %bb.0:
; SSE41-NEXT: movaps %xmm1, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v16f32_zero:
-; AVX: # %bb.0:
-; AVX-NEXT: vaddps %ymm1, %ymm0, %ymm0
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16f32_zero:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vaddps %ymm1, %ymm0, %ymm0
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16f32_zero:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vaddps %ymm1, %ymm0, %ymm0
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v16f32_zero:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vaddps %ymm1, %ymm0, %ymm0
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v16f32_zero:
; AVX512: # %bb.0:
; SSE41-NEXT: addss %xmm1, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v2f32_undef:
-; AVX: # %bb.0:
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v2f32_undef:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v2f32_undef:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v2f32_undef:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v2f32_undef:
; AVX512: # %bb.0:
; SSE41-NEXT: movaps %xmm1, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v4f32_undef:
-; AVX: # %bb.0:
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4f32_undef:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4f32_undef:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4f32_undef:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4f32_undef:
; AVX512: # %bb.0:
; SSE41-NEXT: movaps %xmm1, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v8f32_undef:
-; AVX: # %bb.0:
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8f32_undef:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8f32_undef:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v8f32_undef:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v8f32_undef:
; AVX512: # %bb.0:
; SSE41-NEXT: movaps %xmm1, %xmm0
; SSE41-NEXT: retq
;
-; AVX-LABEL: test_v16f32_undef:
-; AVX: # %bb.0:
-; AVX-NEXT: vaddps %ymm1, %ymm0, %ymm0
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddps %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
-; AVX-NEXT: vaddss %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16f32_undef:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vaddps %ymm1, %ymm0, %ymm0
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX1-SLOW-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16f32_undef:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vaddps %ymm1, %ymm0, %ymm0
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-FAST-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddps %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v16f32_undef:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vaddps %ymm1, %ymm0, %ymm0
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vmovshdup {{.*#+}} xmm1 = xmm0[1,1,3,3]
+; AVX2-NEXT: vaddss %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v16f32_undef:
; AVX512: # %bb.0:
; SSE-NEXT: addsd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v2f64:
-; AVX: # %bb.0:
-; AVX-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
-; AVX-NEXT: vaddsd %xmm0, %xmm1, %xmm0
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v2f64:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm0, %xmm1, %xmm0
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v2f64:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vhaddpd %xmm1, %xmm1, %xmm0
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v2f64:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm0 = xmm1[1,0]
+; AVX2-NEXT: vaddsd %xmm0, %xmm1, %xmm0
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v2f64:
; AVX512: # %bb.0:
; SSE-NEXT: addsd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v4f64:
-; AVX: # %bb.0:
-; AVX-NEXT: vextractf128 $1, %ymm1, %xmm0
-; AVX-NEXT: vaddpd %xmm0, %xmm1, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4f64:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm1, %xmm0
+; AVX1-SLOW-NEXT: vaddpd %xmm0, %xmm1, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4f64:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm1, %xmm0
+; AVX1-FAST-NEXT: vaddpd %xmm0, %xmm1, %xmm0
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4f64:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vextractf128 $1, %ymm1, %xmm0
+; AVX2-NEXT: vaddpd %xmm0, %xmm1, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4f64:
; AVX512: # %bb.0:
; SSE-NEXT: addsd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v8f64:
-; AVX: # %bb.0:
-; AVX-NEXT: vaddpd %ymm2, %ymm1, %ymm0
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddpd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8f64:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vaddpd %ymm2, %ymm1, %ymm0
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8f64:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vaddpd %ymm2, %ymm1, %ymm0
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v8f64:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vaddpd %ymm2, %ymm1, %ymm0
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v8f64:
; AVX512: # %bb.0:
; SSE-NEXT: addsd %xmm4, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v16f64:
-; AVX: # %bb.0:
-; AVX-NEXT: vaddpd %ymm4, %ymm2, %ymm0
-; AVX-NEXT: vaddpd %ymm3, %ymm1, %ymm1
-; AVX-NEXT: vaddpd %ymm0, %ymm1, %ymm0
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddpd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16f64:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vaddpd %ymm4, %ymm2, %ymm0
+; AVX1-SLOW-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; AVX1-SLOW-NEXT: vaddpd %ymm0, %ymm1, %ymm0
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16f64:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vaddpd %ymm4, %ymm2, %ymm0
+; AVX1-FAST-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; AVX1-FAST-NEXT: vaddpd %ymm0, %ymm1, %ymm0
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v16f64:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vaddpd %ymm4, %ymm2, %ymm0
+; AVX2-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; AVX2-NEXT: vaddpd %ymm0, %ymm1, %ymm0
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v16f64:
; AVX512: # %bb.0:
; SSE-NEXT: movapd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v2f64_zero:
-; AVX: # %bb.0:
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v2f64_zero:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v2f64_zero:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v2f64_zero:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v2f64_zero:
; AVX512: # %bb.0:
; SSE-NEXT: movapd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v4f64_zero:
-; AVX: # %bb.0:
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddpd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4f64_zero:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4f64_zero:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4f64_zero:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4f64_zero:
; AVX512: # %bb.0:
; SSE-NEXT: movapd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v8f64_zero:
-; AVX: # %bb.0:
-; AVX-NEXT: vaddpd %ymm1, %ymm0, %ymm0
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddpd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8f64_zero:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8f64_zero:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v8f64_zero:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v8f64_zero:
; AVX512: # %bb.0:
; SSE-NEXT: addsd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v16f64_zero:
-; AVX: # %bb.0:
-; AVX-NEXT: vaddpd %ymm3, %ymm1, %ymm1
-; AVX-NEXT: vaddpd %ymm2, %ymm0, %ymm0
-; AVX-NEXT: vaddpd %ymm1, %ymm0, %ymm0
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddpd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16f64_zero:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; AVX1-SLOW-NEXT: vaddpd %ymm2, %ymm0, %ymm0
+; AVX1-SLOW-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16f64_zero:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; AVX1-FAST-NEXT: vaddpd %ymm2, %ymm0, %ymm0
+; AVX1-FAST-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v16f64_zero:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; AVX2-NEXT: vaddpd %ymm2, %ymm0, %ymm0
+; AVX2-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v16f64_zero:
; AVX512: # %bb.0:
; SSE-NEXT: movapd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v2f64_undef:
-; AVX: # %bb.0:
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v2f64_undef:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v2f64_undef:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v2f64_undef:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v2f64_undef:
; AVX512: # %bb.0:
; SSE-NEXT: movapd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v4f64_undef:
-; AVX: # %bb.0:
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddpd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v4f64_undef:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v4f64_undef:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v4f64_undef:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v4f64_undef:
; AVX512: # %bb.0:
; SSE-NEXT: movapd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v8f64_undef:
-; AVX: # %bb.0:
-; AVX-NEXT: vaddpd %ymm1, %ymm0, %ymm0
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddpd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v8f64_undef:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v8f64_undef:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v8f64_undef:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v8f64_undef:
; AVX512: # %bb.0:
; SSE-NEXT: addsd %xmm1, %xmm0
; SSE-NEXT: retq
;
-; AVX-LABEL: test_v16f64_undef:
-; AVX: # %bb.0:
-; AVX-NEXT: vaddpd %ymm3, %ymm1, %ymm1
-; AVX-NEXT: vaddpd %ymm2, %ymm0, %ymm0
-; AVX-NEXT: vaddpd %ymm1, %ymm0, %ymm0
-; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
-; AVX-NEXT: vaddpd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
-; AVX-NEXT: vaddsd %xmm1, %xmm0, %xmm0
-; AVX-NEXT: vzeroupper
-; AVX-NEXT: retq
+; AVX1-SLOW-LABEL: test_v16f64_undef:
+; AVX1-SLOW: # %bb.0:
+; AVX1-SLOW-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; AVX1-SLOW-NEXT: vaddpd %ymm2, %ymm0, %ymm0
+; AVX1-SLOW-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX1-SLOW-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-SLOW-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX1-SLOW-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX1-SLOW-NEXT: vzeroupper
+; AVX1-SLOW-NEXT: retq
+;
+; AVX1-FAST-LABEL: test_v16f64_undef:
+; AVX1-FAST: # %bb.0:
+; AVX1-FAST-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; AVX1-FAST-NEXT: vaddpd %ymm2, %ymm0, %ymm0
+; AVX1-FAST-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX1-FAST-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX1-FAST-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vhaddpd %xmm0, %xmm0, %xmm0
+; AVX1-FAST-NEXT: vzeroupper
+; AVX1-FAST-NEXT: retq
+;
+; AVX2-LABEL: test_v16f64_undef:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vaddpd %ymm3, %ymm1, %ymm1
+; AVX2-NEXT: vaddpd %ymm2, %ymm0, %ymm0
+; AVX2-NEXT: vaddpd %ymm1, %ymm0, %ymm0
+; AVX2-NEXT: vextractf128 $1, %ymm0, %xmm1
+; AVX2-NEXT: vaddpd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vpermilpd {{.*#+}} xmm1 = xmm0[1,0]
+; AVX2-NEXT: vaddsd %xmm1, %xmm0, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
;
; AVX512-LABEL: test_v16f64_undef:
; AVX512: # %bb.0:
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE41
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512BW
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefix=ALL --check-prefix=AVX512 --check-prefix=AVX512VL
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=SSE,SSE2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefixes=SSE,SSE41
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-SLOW
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx,+fast-hops | FileCheck %s --check-prefixes=AVX,AVX1,AVX1-FAST
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefixes=AVX,AVX2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=AVX512,AVX512BW
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw,+avx512vl | FileCheck %s --check-prefixes=AVX512,AVX512VL
;
; vXf32 (accum)