test/CodeGen/X86/compress_expand.ll

; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mcpu=skylake-avx512 < %s | FileCheck %s --check-prefix=ALL --check-prefix=SKX
; RUN: llc -mcpu=knl < %s | FileCheck %s --check-prefix=ALL --check-prefix=KNL

target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
target triple = "x86_64-unknown-linux-gnu"

define <16 x float> @expandload_v16f32_const_undef(float* %base) {
; SKX-LABEL: expandload_v16f32_const_undef:
; SKX:       # %bb.0:
; SKX-NEXT:    movw $-2049, %ax # imm = 0xF7FF
; SKX-NEXT:    kmovd %eax, %k1
; SKX-NEXT:    vexpandps (%rdi), %zmm0 {%k1} {z}
; SKX-NEXT:    retq
;
; KNL-LABEL: expandload_v16f32_const_undef:
; KNL:       # %bb.0:
; KNL-NEXT:    movw $-2049, %ax # imm = 0xF7FF
; KNL-NEXT:    kmovw %eax, %k1
; KNL-NEXT:    vexpandps (%rdi), %zmm0 {%k1} {z}
; KNL-NEXT:    retq
  %res = call <16 x float> @llvm.masked.expandload.v16f32(float* %base, <16 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 true>, <16 x float> undef)
  ret <16 x float>%res
}

define <16 x float> @expandload_v16f32_const(float* %base, <16 x float> %src0) {
; SKX-LABEL: expandload_v16f32_const:
; SKX:       # %bb.0:
; SKX-NEXT:    movw $30719, %ax # imm = 0x77FF
; SKX-NEXT:    kmovd %eax, %k1
; SKX-NEXT:    vexpandps (%rdi), %zmm0 {%k1}
; SKX-NEXT:    retq
;
; KNL-LABEL: expandload_v16f32_const:
; KNL:       # %bb.0:
; KNL-NEXT:    movw $30719, %ax # imm = 0x77FF
; KNL-NEXT:    kmovw %eax, %k1
; KNL-NEXT:    vexpandps (%rdi), %zmm0 {%k1}
; KNL-NEXT:    retq
  %res = call <16 x float> @llvm.masked.expandload.v16f32(float* %base, <16 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 false>, <16 x float> %src0)
  ret <16 x float>%res
}

define <8 x double> @expandload_v8f64_v8i1(double* %base, <8 x double> %src0, <8 x i1> %mask) {
; SKX-LABEL: expandload_v8f64_v8i1:
; SKX:       # %bb.0:
; SKX-NEXT:    vpsllw $15, %xmm1, %xmm1
; SKX-NEXT:    vpmovw2m %xmm1, %k1
; SKX-NEXT:    vexpandpd (%rdi), %zmm0 {%k1}
; SKX-NEXT:    retq
;
; KNL-LABEL: expandload_v8f64_v8i1:
; KNL:       # %bb.0:
; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
; KNL-NEXT:    vpsllq $63, %zmm1, %zmm1
; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
; KNL-NEXT:    vexpandpd (%rdi), %zmm0 {%k1}
; KNL-NEXT:    retq
  %res = call <8 x double> @llvm.masked.expandload.v8f64(double* %base, <8 x i1> %mask, <8 x double> %src0)
  ret <8 x double>%res
}

define <4 x float> @expandload_v4f32_const(float* %base, <4 x float> %src0) {
; SKX-LABEL: expandload_v4f32_const:
; SKX:       # %bb.0:
; SKX-NEXT:    movb $7, %al
; SKX-NEXT:    kmovd %eax, %k1
; SKX-NEXT:    vexpandps (%rdi), %xmm0 {%k1}
; SKX-NEXT:    retq
;
; KNL-LABEL: expandload_v4f32_const:
; KNL:       # %bb.0:
; KNL-NEXT:    # kill: def $xmm0 killed $xmm0 def $zmm0
; KNL-NEXT:    movw $7, %ax
; KNL-NEXT:    kmovw %eax, %k1
; KNL-NEXT:    vexpandps (%rdi), %zmm0 {%k1}
; KNL-NEXT:    # kill: def $xmm0 killed $xmm0 killed $zmm0
; KNL-NEXT:    retq
  %res = call <4 x float> @llvm.masked.expandload.v4f32(float* %base, <4 x i1> <i1 true, i1 true, i1 true, i1 false>, <4 x float> %src0)
  ret <4 x float>%res
}

define <2 x i64> @expandload_v2i64_const(i64* %base, <2 x i64> %src0) {
; SKX-LABEL: expandload_v2i64_const:
; SKX:       # %bb.0:
; SKX-NEXT:    movb $2, %al
; SKX-NEXT:    kmovd %eax, %k1
; SKX-NEXT:    vpexpandq (%rdi), %xmm0 {%k1}
; SKX-NEXT:    retq
;
; KNL-LABEL: expandload_v2i64_const:
; KNL:       # %bb.0:
; KNL-NEXT:    # kill: def $xmm0 killed $xmm0 def $zmm0
; KNL-NEXT:    movb $2, %al
; KNL-NEXT:    kmovw %eax, %k1
; KNL-NEXT:    vpexpandq (%rdi), %zmm0 {%k1}
; KNL-NEXT:    # kill: def $xmm0 killed $xmm0 killed $zmm0
; KNL-NEXT:    retq
  %res = call <2 x i64> @llvm.masked.expandload.v2i64(i64* %base, <2 x i1> <i1 false, i1 true>, <2 x i64> %src0)
  ret <2 x i64>%res
}

declare <16 x float> @llvm.masked.expandload.v16f32(float*, <16 x i1>, <16 x float>)
declare <8 x double> @llvm.masked.expandload.v8f64(double*, <8 x i1>, <8 x double>)
declare <4 x float>  @llvm.masked.expandload.v4f32(float*, <4 x i1>, <4 x float>)
declare <2 x i64>    @llvm.masked.expandload.v2i64(i64*, <2 x i1>, <2 x i64>)

define void @compressstore_v16f32_const(float* %base, <16 x float> %V) {
; SKX-LABEL: compressstore_v16f32_const:
; SKX:       # %bb.0:
; SKX-NEXT:    movw $-2049, %ax # imm = 0xF7FF
; SKX-NEXT:    kmovd %eax, %k1
; SKX-NEXT:    vcompressps %zmm0, (%rdi) {%k1}
; SKX-NEXT:    vzeroupper
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v16f32_const:
; KNL:       # %bb.0:
; KNL-NEXT:    movw $-2049, %ax # imm = 0xF7FF
; KNL-NEXT:    kmovw %eax, %k1
; KNL-NEXT:    vcompressps %zmm0, (%rdi) {%k1}
; KNL-NEXT:    retq
  call void @llvm.masked.compressstore.v16f32(<16 x float> %V, float* %base, <16 x i1> <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 false, i1 true, i1 true, i1 true, i1 true>)
  ret void
}

define void @compressstore_v8f32_v8i1(float* %base, <8 x float> %V, <8 x i1> %mask) {
; SKX-LABEL: compressstore_v8f32_v8i1:
; SKX:       # %bb.0:
; SKX-NEXT:    vpsllw $15, %xmm1, %xmm1
; SKX-NEXT:    vpmovw2m %xmm1, %k1
; SKX-NEXT:    vcompressps %ymm0, (%rdi) {%k1}
; SKX-NEXT:    vzeroupper
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v8f32_v8i1:
; KNL:       # %bb.0:
; KNL-NEXT:    # kill: def $ymm0 killed $ymm0 def $zmm0
; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
; KNL-NEXT:    vpsllq $63, %zmm1, %zmm1
; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
; KNL-NEXT:    vcompressps %zmm0, (%rdi) {%k1}
; KNL-NEXT:    retq
  call void @llvm.masked.compressstore.v8f32(<8 x float> %V, float* %base, <8 x i1> %mask)
  ret void
}

define void @compressstore_v8f64_v8i1(double* %base, <8 x double> %V, <8 x i1> %mask) {
; SKX-LABEL: compressstore_v8f64_v8i1:
; SKX:       # %bb.0:
; SKX-NEXT:    vpsllw $15, %xmm1, %xmm1
; SKX-NEXT:    vpmovw2m %xmm1, %k1
; SKX-NEXT:    vcompresspd %zmm0, (%rdi) {%k1}
; SKX-NEXT:    vzeroupper
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v8f64_v8i1:
; KNL:       # %bb.0:
; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
; KNL-NEXT:    vpsllq $63, %zmm1, %zmm1
; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
; KNL-NEXT:    vcompresspd %zmm0, (%rdi) {%k1}
; KNL-NEXT:    retq
  call void @llvm.masked.compressstore.v8f64(<8 x double> %V, double* %base, <8 x i1> %mask)
  ret void
}

define void @compressstore_v8i64_v8i1(i64* %base, <8 x i64> %V, <8 x i1> %mask) {
; SKX-LABEL: compressstore_v8i64_v8i1:
; SKX:       # %bb.0:
; SKX-NEXT:    vpsllw $15, %xmm1, %xmm1
; SKX-NEXT:    vpmovw2m %xmm1, %k1
; SKX-NEXT:    vpcompressq %zmm0, (%rdi) {%k1}
; SKX-NEXT:    vzeroupper
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v8i64_v8i1:
; KNL:       # %bb.0:
; KNL-NEXT:    vpmovzxwq {{.*#+}} zmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero,xmm1[4],zero,zero,zero,xmm1[5],zero,zero,zero,xmm1[6],zero,zero,zero,xmm1[7],zero,zero,zero
; KNL-NEXT:    vpsllq $63, %zmm1, %zmm1
; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k1
; KNL-NEXT:    vpcompressq %zmm0, (%rdi) {%k1}
; KNL-NEXT:    retq
    call void @llvm.masked.compressstore.v8i64(<8 x i64> %V, i64* %base, <8 x i1> %mask)
  ret void
}

define void @compressstore_v4i64_v4i1(i64* %base, <4 x i64> %V, <4 x i1> %mask) {
; SKX-LABEL: compressstore_v4i64_v4i1:
; SKX:       # %bb.0:
; SKX-NEXT:    vpslld $31, %xmm1, %xmm1
; SKX-NEXT:    vpmovd2m %xmm1, %k1
; SKX-NEXT:    vpcompressq %ymm0, (%rdi) {%k1}
; SKX-NEXT:    vzeroupper
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v4i64_v4i1:
; KNL:       # %bb.0:
; KNL-NEXT:    # kill: def $ymm0 killed $ymm0 def $zmm0
; KNL-NEXT:    vpslld $31, %xmm1, %xmm1
; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k0
; KNL-NEXT:    kshiftlw $12, %k0, %k0
; KNL-NEXT:    kshiftrw $12, %k0, %k1
; KNL-NEXT:    vpcompressq %zmm0, (%rdi) {%k1}
; KNL-NEXT:    retq
    call void @llvm.masked.compressstore.v4i64(<4 x i64> %V, i64* %base, <4 x i1> %mask)
  ret void
}

define void @compressstore_v2i64_v2i1(i64* %base, <2 x i64> %V, <2 x i1> %mask) {
; SKX-LABEL: compressstore_v2i64_v2i1:
; SKX:       # %bb.0:
; SKX-NEXT:    vpsllq $63, %xmm1, %xmm1
; SKX-NEXT:    vpmovq2m %xmm1, %k1
; SKX-NEXT:    vpcompressq %xmm0, (%rdi) {%k1}
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v2i64_v2i1:
; KNL:       # %bb.0:
; KNL-NEXT:    # kill: def $xmm0 killed $xmm0 def $zmm0
; KNL-NEXT:    vpsllq $63, %xmm1, %xmm1
; KNL-NEXT:    vptestmq %zmm1, %zmm1, %k0
; KNL-NEXT:    kshiftlw $14, %k0, %k0
; KNL-NEXT:    kshiftrw $14, %k0, %k1
; KNL-NEXT:    vpcompressq %zmm0, (%rdi) {%k1}
; KNL-NEXT:    retq
    call void @llvm.masked.compressstore.v2i64(<2 x i64> %V, i64* %base, <2 x i1> %mask)
  ret void
}

define void @compressstore_v4f32_v4i1(float* %base, <4 x float> %V, <4 x i1> %mask) {
; SKX-LABEL: compressstore_v4f32_v4i1:
; SKX:       # %bb.0:
; SKX-NEXT:    vpslld $31, %xmm1, %xmm1
; SKX-NEXT:    vpmovd2m %xmm1, %k1
; SKX-NEXT:    vcompressps %xmm0, (%rdi) {%k1}
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v4f32_v4i1:
; KNL:       # %bb.0:
; KNL-NEXT:    # kill: def $xmm0 killed $xmm0 def $zmm0
; KNL-NEXT:    vpslld $31, %xmm1, %xmm1
; KNL-NEXT:    vptestmd %zmm1, %zmm1, %k0
; KNL-NEXT:    kshiftlw $12, %k0, %k0
; KNL-NEXT:    kshiftrw $12, %k0, %k1
; KNL-NEXT:    vcompressps %zmm0, (%rdi) {%k1}
; KNL-NEXT:    retq
    call void @llvm.masked.compressstore.v4f32(<4 x float> %V, float* %base, <4 x i1> %mask)
  ret void
}

define <2 x float> @expandload_v2f32_v2i1(float* %base, <2 x float> %src0, <2 x i32> %trigger) {
; SKX-LABEL: expandload_v2f32_v2i1:
; SKX:       # %bb.0:
; SKX-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; SKX-NEXT:    vpblendd {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3]
; SKX-NEXT:    vptestnmq %xmm1, %xmm1, %k1
; SKX-NEXT:    vexpandps (%rdi), %xmm0 {%k1}
; SKX-NEXT:    retq
;
; KNL-LABEL: expandload_v2f32_v2i1:
; KNL:       # %bb.0:
; KNL-NEXT:    # kill: def $xmm0 killed $xmm0 def $zmm0
; KNL-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; KNL-NEXT:    vpblendd {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3]
; KNL-NEXT:    vptestnmq %zmm1, %zmm1, %k0
; KNL-NEXT:    kshiftlw $14, %k0, %k0
; KNL-NEXT:    kshiftrw $14, %k0, %k1
; KNL-NEXT:    vexpandps (%rdi), %zmm0 {%k1}
; KNL-NEXT:    # kill: def $xmm0 killed $xmm0 killed $zmm0
; KNL-NEXT:    retq
  %mask = icmp eq <2 x i32> %trigger, zeroinitializer
  %res = call <2 x float> @llvm.masked.expandload.v2f32(float* %base, <2 x i1> %mask, <2 x float> %src0)
  ret <2 x float> %res
}

define void @compressstore_v2f32_v2i32(float* %base, <2 x float> %V, <2 x i32> %trigger) {
; SKX-LABEL: compressstore_v2f32_v2i32:
; SKX:       # %bb.0:
; SKX-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; SKX-NEXT:    vpblendd {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3]
; SKX-NEXT:    vptestnmq %xmm1, %xmm1, %k1
; SKX-NEXT:    vcompressps %xmm0, (%rdi) {%k1}
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v2f32_v2i32:
; KNL:       # %bb.0:
; KNL-NEXT:    # kill: def $xmm0 killed $xmm0 def $zmm0
; KNL-NEXT:    vpxor %xmm2, %xmm2, %xmm2
; KNL-NEXT:    vpblendd {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3]
; KNL-NEXT:    vptestnmq %zmm1, %zmm1, %k0
; KNL-NEXT:    kshiftlw $14, %k0, %k0
; KNL-NEXT:    kshiftrw $14, %k0, %k1
; KNL-NEXT:    vcompressps %zmm0, (%rdi) {%k1}
; KNL-NEXT:    retq
  %mask = icmp eq <2 x i32> %trigger, zeroinitializer
  call void @llvm.masked.compressstore.v2f32(<2 x float> %V, float* %base, <2 x i1> %mask)
  ret void
}

define <32 x float> @expandload_v32f32_v32i32(float* %base, <32 x float> %src0, <32 x i32> %trigger) {
; ALL-LABEL: expandload_v32f32_v32i32:
; ALL:       # %bb.0:
; ALL-NEXT:    vptestnmd %zmm3, %zmm3, %k1
; ALL-NEXT:    vptestnmd %zmm2, %zmm2, %k2
; ALL-NEXT:    kmovw %k2, %eax
; ALL-NEXT:    popcntl %eax, %eax
; ALL-NEXT:    vexpandps (%rdi,%rax,4), %zmm1 {%k1}
; ALL-NEXT:    vexpandps (%rdi), %zmm0 {%k2}
; ALL-NEXT:    retq
  %mask = icmp eq <32 x i32> %trigger, zeroinitializer
  %res = call <32 x float> @llvm.masked.expandload.v32f32(float* %base, <32 x i1> %mask, <32 x float> %src0)
  ret <32 x float> %res
}

define <16 x double> @compressstore_v16f64_v16i32(double* %base, <16 x double> %src0, <16 x i32> %trigger) {
; SKX-LABEL: compressstore_v16f64_v16i32:
; SKX:       # %bb.0:
; SKX-NEXT:    vextracti64x4 $1, %zmm2, %ymm3
; SKX-NEXT:    vptestnmd %ymm3, %ymm3, %k1
; SKX-NEXT:    vptestnmd %ymm2, %ymm2, %k2
; SKX-NEXT:    kmovb %k2, %eax
; SKX-NEXT:    popcntl %eax, %eax
; SKX-NEXT:    vexpandpd (%rdi,%rax,8), %zmm1 {%k1}
; SKX-NEXT:    vexpandpd (%rdi), %zmm0 {%k2}
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v16f64_v16i32:
; KNL:       # %bb.0:
; KNL-NEXT:    vextracti64x4 $1, %zmm2, %ymm3
; KNL-NEXT:    vptestnmd %zmm3, %zmm3, %k1
; KNL-NEXT:    vptestnmd %zmm2, %zmm2, %k2
; KNL-NEXT:    vexpandpd (%rdi), %zmm0 {%k2}
; KNL-NEXT:    kmovw %k2, %eax
; KNL-NEXT:    movzbl %al, %eax
; KNL-NEXT:    popcntl %eax, %eax
; KNL-NEXT:    vexpandpd (%rdi,%rax,8), %zmm1 {%k1}
; KNL-NEXT:    retq
  %mask = icmp eq <16 x i32> %trigger, zeroinitializer
  %res = call <16 x double> @llvm.masked.expandload.v16f64(double* %base, <16 x i1> %mask, <16 x double> %src0)
  ret <16 x double> %res
}

define void @compressstore_v32f32_v32i32(float* %base, <32 x float> %V, <32 x i32> %trigger) {
; SKX-LABEL: compressstore_v32f32_v32i32:
; SKX:       # %bb.0:
; SKX-NEXT:    vptestnmd %zmm3, %zmm3, %k1
; SKX-NEXT:    vptestnmd %zmm2, %zmm2, %k2
; SKX-NEXT:    kmovw %k2, %eax
; SKX-NEXT:    popcntl %eax, %eax
; SKX-NEXT:    vcompressps %zmm1, (%rdi,%rax,4) {%k1}
; SKX-NEXT:    vcompressps %zmm0, (%rdi) {%k2}
; SKX-NEXT:    vzeroupper
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v32f32_v32i32:
; KNL:       # %bb.0:
; KNL-NEXT:    vptestnmd %zmm3, %zmm3, %k1
; KNL-NEXT:    vptestnmd %zmm2, %zmm2, %k2
; KNL-NEXT:    kmovw %k2, %eax
; KNL-NEXT:    popcntl %eax, %eax
; KNL-NEXT:    vcompressps %zmm1, (%rdi,%rax,4) {%k1}
; KNL-NEXT:    vcompressps %zmm0, (%rdi) {%k2}
; KNL-NEXT:    retq
  %mask = icmp eq <32 x i32> %trigger, zeroinitializer
  call void @llvm.masked.compressstore.v32f32(<32 x float> %V, float* %base, <32 x i1> %mask)
  ret void
}

define void @compressstore_v16f64_v16i1(double* %base, <16 x double> %V, <16 x i1> %mask) {
; SKX-LABEL: compressstore_v16f64_v16i1:
; SKX:       # %bb.0:
; SKX-NEXT:    vpsllw $7, %xmm2, %xmm2
; SKX-NEXT:    vpmovb2m %xmm2, %k1
; SKX-NEXT:    kshiftrw $8, %k1, %k2
; SKX-NEXT:    kmovb %k1, %eax
; SKX-NEXT:    popcntl %eax, %eax
; SKX-NEXT:    vcompresspd %zmm1, (%rdi,%rax,8) {%k2}
; SKX-NEXT:    vcompresspd %zmm0, (%rdi) {%k1}
; SKX-NEXT:    vzeroupper
; SKX-NEXT:    retq
;
; KNL-LABEL: compressstore_v16f64_v16i1:
; KNL:       # %bb.0:
; KNL-NEXT:    vpmovzxbd {{.*#+}} zmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero,xmm2[8],zero,zero,zero,xmm2[9],zero,zero,zero,xmm2[10],zero,zero,zero,xmm2[11],zero,zero,zero,xmm2[12],zero,zero,zero,xmm2[13],zero,zero,zero,xmm2[14],zero,zero,zero,xmm2[15],zero,zero,zero
; KNL-NEXT:    vpslld $31, %zmm2, %zmm2
; KNL-NEXT:    vptestmd %zmm2, %zmm2, %k1
; KNL-NEXT:    kshiftrw $8, %k1, %k2
; KNL-NEXT:    vcompresspd %zmm0, (%rdi) {%k1}
; KNL-NEXT:    kmovw %k1, %eax
; KNL-NEXT:    movzbl %al, %eax
; KNL-NEXT:    popcntl %eax, %eax
; KNL-NEXT:    vcompresspd %zmm1, (%rdi,%rax,8) {%k2}
; KNL-NEXT:    retq
  call void @llvm.masked.compressstore.v16f64(<16 x double> %V, double* %base, <16 x i1> %mask)
  ret void
}

declare void @llvm.masked.compressstore.v16f32(<16 x float>, float* , <16 x i1>)
declare void @llvm.masked.compressstore.v8f32(<8 x float>, float* , <8 x i1>)
declare void @llvm.masked.compressstore.v8f64(<8 x double>, double* , <8 x i1>)
declare void @llvm.masked.compressstore.v16i32(<16 x i32>, i32* , <16 x i1>)
declare void @llvm.masked.compressstore.v8i32(<8 x i32>, i32* , <8 x i1>)
declare void @llvm.masked.compressstore.v8i64(<8 x i64>, i64* , <8 x i1>)
declare void @llvm.masked.compressstore.v4i32(<4 x i32>, i32* , <4 x i1>)
declare void @llvm.masked.compressstore.v4f32(<4 x float>, float* , <4 x i1>)
declare void @llvm.masked.compressstore.v4i64(<4 x i64>, i64* , <4 x i1>)
declare void @llvm.masked.compressstore.v2i64(<2 x i64>, i64* , <2 x i1>)
declare void @llvm.masked.compressstore.v2f32(<2 x float>, float* , <2 x i1>)
declare void @llvm.masked.compressstore.v32f32(<32 x float>, float* , <32 x i1>)
declare void @llvm.masked.compressstore.v16f64(<16 x double>, double* , <16 x i1>)
declare void @llvm.masked.compressstore.v32f64(<32 x double>, double* , <32 x i1>)

declare <2 x float> @llvm.masked.expandload.v2f32(float* , <2 x i1> , <2 x float> )
declare <32 x float> @llvm.masked.expandload.v32f32(float* , <32 x i1> , <32 x float> )
declare <16 x double> @llvm.masked.expandload.v16f64(double* , <16 x i1> , <16 x double> )