[llvm] r347159 - [X86][SSE] Add some generic masked gather codegen tests
Simon Pilgrim via llvm-commits
llvm-commits at lists.llvm.org
Sun Nov 18 06:35:58 PST 2018
Author: rksimon
Date: Sun Nov 18 06:35:57 2018
New Revision: 347159
URL: http://llvm.org/viewvc/llvm-project?rev=347159&view=rev
Log:
[X86][SSE] Add some generic masked gather codegen tests
Added:
llvm/trunk/test/CodeGen/X86/masked_gather.ll
Added: llvm/trunk/test/CodeGen/X86/masked_gather.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/masked_gather.ll?rev=347159&view=auto
==============================================================================
--- llvm/trunk/test/CodeGen/X86/masked_gather.ll (added)
+++ llvm/trunk/test/CodeGen/X86/masked_gather.ll Sun Nov 18 06:35:57 2018
@@ -0,0 +1,1156 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+sse4.2 < %s | FileCheck %s --check-prefixes=ALL,SSE,SSE42
+; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+avx < %s | FileCheck %s --check-prefixes=ALL,AVX,AVX1
+; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+avx2 < %s | FileCheck %s --check-prefixes=ALL,AVX,AVX2
+; RUN: llc -mtriple=x86_64-unknown-linux-gnu -mattr=+avx512f < %s | FileCheck %s --check-prefixes=ALL,AVX512
+
+;
+; vXf32
+;
+
+define <4 x float> @gather_v4f32_ptr_v4i32(<4 x float*> %ptr, <4 x i32> %trigger, <4 x float> %passthru) {
+; SSE-LABEL: gather_v4f32_ptr_v4i32:
+; SSE: # %bb.0:
+; SSE-NEXT: pxor %xmm4, %xmm4
+; SSE-NEXT: pcmpeqd %xmm2, %xmm4
+; SSE-NEXT: pextrb $0, %xmm4, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB0_2
+; SSE-NEXT: # %bb.1: # %cond.load
+; SSE-NEXT: movq %xmm0, %rax
+; SSE-NEXT: movss {{.*#+}} xmm5 = mem[0],zero,zero,zero
+; SSE-NEXT: blendps {{.*#+}} xmm3 = xmm5[0],xmm3[1,2,3]
+; SSE-NEXT: .LBB0_2: # %else
+; SSE-NEXT: pextrb $4, %xmm4, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB0_4
+; SSE-NEXT: # %bb.3: # %cond.load1
+; SSE-NEXT: pextrq $1, %xmm0, %rax
+; SSE-NEXT: insertps {{.*#+}} xmm3 = xmm3[0],mem[0],xmm3[2,3]
+; SSE-NEXT: .LBB0_4: # %else2
+; SSE-NEXT: pxor %xmm0, %xmm0
+; SSE-NEXT: pcmpeqd %xmm0, %xmm2
+; SSE-NEXT: pextrb $8, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB0_6
+; SSE-NEXT: # %bb.5: # %cond.load4
+; SSE-NEXT: movq %xmm1, %rax
+; SSE-NEXT: insertps {{.*#+}} xmm3 = xmm3[0,1],mem[0],xmm3[3]
+; SSE-NEXT: .LBB0_6: # %else5
+; SSE-NEXT: pextrb $12, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB0_8
+; SSE-NEXT: # %bb.7: # %cond.load7
+; SSE-NEXT: pextrq $1, %xmm1, %rax
+; SSE-NEXT: insertps {{.*#+}} xmm3 = xmm3[0,1,2],mem[0]
+; SSE-NEXT: .LBB0_8: # %else8
+; SSE-NEXT: movaps %xmm3, %xmm0
+; SSE-NEXT: retq
+;
+; AVX1-LABEL: gather_v4f32_ptr_v4i32:
+; AVX1: # %bb.0:
+; AVX1-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX1-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm3
+; AVX1-NEXT: vpextrb $0, %xmm3, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB0_2
+; AVX1-NEXT: # %bb.1: # %cond.load
+; AVX1-NEXT: vmovq %xmm0, %rax
+; AVX1-NEXT: vmovss {{.*#+}} xmm4 = mem[0],zero,zero,zero
+; AVX1-NEXT: vblendps {{.*#+}} xmm2 = xmm4[0],xmm2[1,2,3]
+; AVX1-NEXT: .LBB0_2: # %else
+; AVX1-NEXT: vpextrb $4, %xmm3, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB0_4
+; AVX1-NEXT: # %bb.3: # %cond.load1
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],mem[0],xmm2[2,3]
+; AVX1-NEXT: .LBB0_4: # %else2
+; AVX1-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX1-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm1
+; AVX1-NEXT: vpextrb $8, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB0_6
+; AVX1-NEXT: # %bb.5: # %cond.load4
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-NEXT: vmovq %xmm3, %rax
+; AVX1-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],mem[0],xmm2[3]
+; AVX1-NEXT: .LBB0_6: # %else5
+; AVX1-NEXT: vpextrb $12, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB0_8
+; AVX1-NEXT: # %bb.7: # %cond.load7
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1,2],mem[0]
+; AVX1-NEXT: .LBB0_8: # %else8
+; AVX1-NEXT: vmovaps %xmm2, %xmm0
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: gather_v4f32_ptr_v4i32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX2-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm3
+; AVX2-NEXT: vpextrb $0, %xmm3, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB0_2
+; AVX2-NEXT: # %bb.1: # %cond.load
+; AVX2-NEXT: vmovq %xmm0, %rax
+; AVX2-NEXT: vmovss {{.*#+}} xmm4 = mem[0],zero,zero,zero
+; AVX2-NEXT: vblendps {{.*#+}} xmm2 = xmm4[0],xmm2[1,2,3]
+; AVX2-NEXT: .LBB0_2: # %else
+; AVX2-NEXT: vpextrb $4, %xmm3, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB0_4
+; AVX2-NEXT: # %bb.3: # %cond.load1
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],mem[0],xmm2[2,3]
+; AVX2-NEXT: .LBB0_4: # %else2
+; AVX2-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX2-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm1
+; AVX2-NEXT: vpextrb $8, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB0_6
+; AVX2-NEXT: # %bb.5: # %cond.load4
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm3
+; AVX2-NEXT: vmovq %xmm3, %rax
+; AVX2-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],mem[0],xmm2[3]
+; AVX2-NEXT: .LBB0_6: # %else5
+; AVX2-NEXT: vpextrb $12, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB0_8
+; AVX2-NEXT: # %bb.7: # %cond.load7
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1,2],mem[0]
+; AVX2-NEXT: .LBB0_8: # %else8
+; AVX2-NEXT: vmovaps %xmm2, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
+;
+; AVX512-LABEL: gather_v4f32_ptr_v4i32:
+; AVX512: # %bb.0:
+; AVX512-NEXT: # kill: def $xmm2 killed $xmm2 def $ymm2
+; AVX512-NEXT: # kill: def $xmm1 killed $xmm1 def $zmm1
+; AVX512-NEXT: # kill: def $ymm0 killed $ymm0 def $zmm0
+; AVX512-NEXT: vptestnmd %zmm1, %zmm1, %k0
+; AVX512-NEXT: kshiftlw $12, %k0, %k0
+; AVX512-NEXT: kshiftrw $12, %k0, %k1
+; AVX512-NEXT: vgatherqps (,%zmm0), %ymm2 {%k1}
+; AVX512-NEXT: vmovaps %xmm2, %xmm0
+; AVX512-NEXT: vzeroupper
+; AVX512-NEXT: retq
+ %mask = icmp eq <4 x i32> %trigger, zeroinitializer
+ %res = call <4 x float> @llvm.masked.gather.v4f32.v4p0f32(<4 x float*> %ptr, i32 4, <4 x i1> %mask, <4 x float> %passthru)
+ ret <4 x float> %res
+}
+
+define <4 x float> @gather_v4f32_v4i32_v4i32(float* %base, <4 x i32> %idx, <4 x i32> %trigger, <4 x float> %passthru) {
+; SSE-LABEL: gather_v4f32_v4i32_v4i32:
+; SSE: # %bb.0:
+; SSE-NEXT: movq %rdi, %xmm3
+; SSE-NEXT: pshufd {{.*#+}} xmm3 = xmm3[0,1,0,1]
+; SSE-NEXT: pmovsxdq %xmm0, %xmm4
+; SSE-NEXT: psllq $2, %xmm4
+; SSE-NEXT: paddq %xmm3, %xmm4
+; SSE-NEXT: pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; SSE-NEXT: pmovsxdq %xmm0, %xmm0
+; SSE-NEXT: pxor %xmm5, %xmm5
+; SSE-NEXT: pcmpeqd %xmm1, %xmm5
+; SSE-NEXT: pextrb $0, %xmm5, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB1_2
+; SSE-NEXT: # %bb.1: # %cond.load
+; SSE-NEXT: movq %xmm4, %rax
+; SSE-NEXT: movss {{.*#+}} xmm6 = mem[0],zero,zero,zero
+; SSE-NEXT: blendps {{.*#+}} xmm2 = xmm6[0],xmm2[1,2,3]
+; SSE-NEXT: .LBB1_2: # %else
+; SSE-NEXT: psllq $2, %xmm0
+; SSE-NEXT: pextrb $4, %xmm5, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB1_4
+; SSE-NEXT: # %bb.3: # %cond.load1
+; SSE-NEXT: pextrq $1, %xmm4, %rax
+; SSE-NEXT: insertps {{.*#+}} xmm2 = xmm2[0],mem[0],xmm2[2,3]
+; SSE-NEXT: .LBB1_4: # %else2
+; SSE-NEXT: paddq %xmm0, %xmm3
+; SSE-NEXT: pxor %xmm0, %xmm0
+; SSE-NEXT: pcmpeqd %xmm0, %xmm1
+; SSE-NEXT: pextrb $8, %xmm1, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB1_6
+; SSE-NEXT: # %bb.5: # %cond.load4
+; SSE-NEXT: movq %xmm3, %rax
+; SSE-NEXT: insertps {{.*#+}} xmm2 = xmm2[0,1],mem[0],xmm2[3]
+; SSE-NEXT: .LBB1_6: # %else5
+; SSE-NEXT: pextrb $12, %xmm1, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB1_8
+; SSE-NEXT: # %bb.7: # %cond.load7
+; SSE-NEXT: pextrq $1, %xmm3, %rax
+; SSE-NEXT: insertps {{.*#+}} xmm2 = xmm2[0,1,2],mem[0]
+; SSE-NEXT: .LBB1_8: # %else8
+; SSE-NEXT: movaps %xmm2, %xmm0
+; SSE-NEXT: retq
+;
+; AVX1-LABEL: gather_v4f32_v4i32_v4i32:
+; AVX1: # %bb.0:
+; AVX1-NEXT: vmovq %rdi, %xmm3
+; AVX1-NEXT: vpshufd {{.*#+}} xmm3 = xmm3[0,1,0,1]
+; AVX1-NEXT: vpshufd {{.*#+}} xmm4 = xmm0[2,3,0,1]
+; AVX1-NEXT: vpmovsxdq %xmm4, %xmm4
+; AVX1-NEXT: vpsllq $2, %xmm4, %xmm4
+; AVX1-NEXT: vpaddq %xmm4, %xmm3, %xmm4
+; AVX1-NEXT: vpmovsxdq %xmm0, %xmm0
+; AVX1-NEXT: vpsllq $2, %xmm0, %xmm0
+; AVX1-NEXT: vpaddq %xmm0, %xmm3, %xmm0
+; AVX1-NEXT: vinsertf128 $1, %xmm4, %ymm0, %ymm0
+; AVX1-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX1-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm3
+; AVX1-NEXT: vpextrb $0, %xmm3, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB1_2
+; AVX1-NEXT: # %bb.1: # %cond.load
+; AVX1-NEXT: vmovq %xmm0, %rax
+; AVX1-NEXT: vmovd {{.*#+}} xmm4 = mem[0],zero,zero,zero
+; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm4[0,1],xmm2[2,3,4,5,6,7]
+; AVX1-NEXT: .LBB1_2: # %else
+; AVX1-NEXT: vpextrb $4, %xmm3, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB1_4
+; AVX1-NEXT: # %bb.3: # %cond.load1
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],mem[0],xmm2[2,3]
+; AVX1-NEXT: .LBB1_4: # %else2
+; AVX1-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX1-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm1
+; AVX1-NEXT: vpextrb $8, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB1_6
+; AVX1-NEXT: # %bb.5: # %cond.load4
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-NEXT: vmovq %xmm3, %rax
+; AVX1-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],mem[0],xmm2[3]
+; AVX1-NEXT: .LBB1_6: # %else5
+; AVX1-NEXT: vpextrb $12, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB1_8
+; AVX1-NEXT: # %bb.7: # %cond.load7
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1,2],mem[0]
+; AVX1-NEXT: .LBB1_8: # %else8
+; AVX1-NEXT: vmovaps %xmm2, %xmm0
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: gather_v4f32_v4i32_v4i32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vmovq %rdi, %xmm3
+; AVX2-NEXT: vpbroadcastq %xmm3, %ymm3
+; AVX2-NEXT: vpmovsxdq %xmm0, %ymm0
+; AVX2-NEXT: vpsllq $2, %ymm0, %ymm0
+; AVX2-NEXT: vpaddq %ymm0, %ymm3, %ymm0
+; AVX2-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX2-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm3
+; AVX2-NEXT: vpextrb $0, %xmm3, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB1_2
+; AVX2-NEXT: # %bb.1: # %cond.load
+; AVX2-NEXT: vmovq %xmm0, %rax
+; AVX2-NEXT: vmovss {{.*#+}} xmm4 = mem[0],zero,zero,zero
+; AVX2-NEXT: vblendps {{.*#+}} xmm2 = xmm4[0],xmm2[1,2,3]
+; AVX2-NEXT: .LBB1_2: # %else
+; AVX2-NEXT: vpextrb $4, %xmm3, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB1_4
+; AVX2-NEXT: # %bb.3: # %cond.load1
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],mem[0],xmm2[2,3]
+; AVX2-NEXT: .LBB1_4: # %else2
+; AVX2-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX2-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm1
+; AVX2-NEXT: vpextrb $8, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB1_6
+; AVX2-NEXT: # %bb.5: # %cond.load4
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm3
+; AVX2-NEXT: vmovq %xmm3, %rax
+; AVX2-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],mem[0],xmm2[3]
+; AVX2-NEXT: .LBB1_6: # %else5
+; AVX2-NEXT: vpextrb $12, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB1_8
+; AVX2-NEXT: # %bb.7: # %cond.load7
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1,2],mem[0]
+; AVX2-NEXT: .LBB1_8: # %else8
+; AVX2-NEXT: vmovaps %xmm2, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
+;
+; AVX512-LABEL: gather_v4f32_v4i32_v4i32:
+; AVX512: # %bb.0:
+; AVX512-NEXT: # kill: def $xmm2 killed $xmm2 def $zmm2
+; AVX512-NEXT: # kill: def $xmm1 killed $xmm1 def $zmm1
+; AVX512-NEXT: # kill: def $xmm0 killed $xmm0 def $zmm0
+; AVX512-NEXT: vptestnmd %zmm1, %zmm1, %k0
+; AVX512-NEXT: kshiftlw $12, %k0, %k0
+; AVX512-NEXT: kshiftrw $12, %k0, %k1
+; AVX512-NEXT: vgatherdps (%rdi,%zmm0,4), %zmm2 {%k1}
+; AVX512-NEXT: vmovaps %xmm2, %xmm0
+; AVX512-NEXT: vzeroupper
+; AVX512-NEXT: retq
+ %vptr0 = insertelement <4 x float*> undef, float* %base, i32 0
+ %vptr1 = shufflevector <4 x float*> %vptr0, <4 x float*> undef, <4 x i32> zeroinitializer
+ %vptr2 = getelementptr float, <4 x float*> %vptr1, <4 x i32> %idx
+
+ %mask = icmp eq <4 x i32> %trigger, zeroinitializer
+ %res = call <4 x float> @llvm.masked.gather.v4f32.v4p0f32(<4 x float*> %vptr2, i32 4, <4 x i1> %mask, <4 x float> %passthru)
+ ret <4 x float> %res
+}
+
+define <4 x float> @gather_v4f32_v4i64_v4i32(float* %base, <4 x i64> %idx, <4 x i32> %trigger, <4 x float> %passthru) {
+; SSE-LABEL: gather_v4f32_v4i64_v4i32:
+; SSE: # %bb.0:
+; SSE-NEXT: movq %rdi, %xmm4
+; SSE-NEXT: pshufd {{.*#+}} xmm4 = xmm4[0,1,0,1]
+; SSE-NEXT: psllq $2, %xmm0
+; SSE-NEXT: paddq %xmm4, %xmm0
+; SSE-NEXT: pxor %xmm5, %xmm5
+; SSE-NEXT: pcmpeqd %xmm2, %xmm5
+; SSE-NEXT: pextrb $0, %xmm5, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB2_2
+; SSE-NEXT: # %bb.1: # %cond.load
+; SSE-NEXT: movq %xmm0, %rax
+; SSE-NEXT: movss {{.*#+}} xmm6 = mem[0],zero,zero,zero
+; SSE-NEXT: blendps {{.*#+}} xmm3 = xmm6[0],xmm3[1,2,3]
+; SSE-NEXT: .LBB2_2: # %else
+; SSE-NEXT: psllq $2, %xmm1
+; SSE-NEXT: pextrb $4, %xmm5, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB2_4
+; SSE-NEXT: # %bb.3: # %cond.load1
+; SSE-NEXT: pextrq $1, %xmm0, %rax
+; SSE-NEXT: insertps {{.*#+}} xmm3 = xmm3[0],mem[0],xmm3[2,3]
+; SSE-NEXT: .LBB2_4: # %else2
+; SSE-NEXT: paddq %xmm1, %xmm4
+; SSE-NEXT: pxor %xmm0, %xmm0
+; SSE-NEXT: pcmpeqd %xmm0, %xmm2
+; SSE-NEXT: pextrb $8, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB2_6
+; SSE-NEXT: # %bb.5: # %cond.load4
+; SSE-NEXT: movq %xmm4, %rax
+; SSE-NEXT: insertps {{.*#+}} xmm3 = xmm3[0,1],mem[0],xmm3[3]
+; SSE-NEXT: .LBB2_6: # %else5
+; SSE-NEXT: pextrb $12, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB2_8
+; SSE-NEXT: # %bb.7: # %cond.load7
+; SSE-NEXT: pextrq $1, %xmm4, %rax
+; SSE-NEXT: insertps {{.*#+}} xmm3 = xmm3[0,1,2],mem[0]
+; SSE-NEXT: .LBB2_8: # %else8
+; SSE-NEXT: movaps %xmm3, %xmm0
+; SSE-NEXT: retq
+;
+; AVX1-LABEL: gather_v4f32_v4i64_v4i32:
+; AVX1: # %bb.0:
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-NEXT: vpsllq $2, %xmm3, %xmm3
+; AVX1-NEXT: vmovq %rdi, %xmm4
+; AVX1-NEXT: vpshufd {{.*#+}} xmm4 = xmm4[0,1,0,1]
+; AVX1-NEXT: vpaddq %xmm3, %xmm4, %xmm3
+; AVX1-NEXT: vpsllq $2, %xmm0, %xmm0
+; AVX1-NEXT: vpaddq %xmm0, %xmm4, %xmm0
+; AVX1-NEXT: vinsertf128 $1, %xmm3, %ymm0, %ymm0
+; AVX1-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX1-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm3
+; AVX1-NEXT: vpextrb $0, %xmm3, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB2_2
+; AVX1-NEXT: # %bb.1: # %cond.load
+; AVX1-NEXT: vmovq %xmm0, %rax
+; AVX1-NEXT: vmovd {{.*#+}} xmm4 = mem[0],zero,zero,zero
+; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm4[0,1],xmm2[2,3,4,5,6,7]
+; AVX1-NEXT: .LBB2_2: # %else
+; AVX1-NEXT: vpextrb $4, %xmm3, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB2_4
+; AVX1-NEXT: # %bb.3: # %cond.load1
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],mem[0],xmm2[2,3]
+; AVX1-NEXT: .LBB2_4: # %else2
+; AVX1-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX1-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm1
+; AVX1-NEXT: vpextrb $8, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB2_6
+; AVX1-NEXT: # %bb.5: # %cond.load4
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm3
+; AVX1-NEXT: vmovq %xmm3, %rax
+; AVX1-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],mem[0],xmm2[3]
+; AVX1-NEXT: .LBB2_6: # %else5
+; AVX1-NEXT: vpextrb $12, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB2_8
+; AVX1-NEXT: # %bb.7: # %cond.load7
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1,2],mem[0]
+; AVX1-NEXT: .LBB2_8: # %else8
+; AVX1-NEXT: vmovaps %xmm2, %xmm0
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: gather_v4f32_v4i64_v4i32:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vmovq %rdi, %xmm3
+; AVX2-NEXT: vpbroadcastq %xmm3, %ymm3
+; AVX2-NEXT: vpsllq $2, %ymm0, %ymm0
+; AVX2-NEXT: vpaddq %ymm0, %ymm3, %ymm0
+; AVX2-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX2-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm3
+; AVX2-NEXT: vpextrb $0, %xmm3, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB2_2
+; AVX2-NEXT: # %bb.1: # %cond.load
+; AVX2-NEXT: vmovq %xmm0, %rax
+; AVX2-NEXT: vmovss {{.*#+}} xmm4 = mem[0],zero,zero,zero
+; AVX2-NEXT: vblendps {{.*#+}} xmm2 = xmm4[0],xmm2[1,2,3]
+; AVX2-NEXT: .LBB2_2: # %else
+; AVX2-NEXT: vpextrb $4, %xmm3, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB2_4
+; AVX2-NEXT: # %bb.3: # %cond.load1
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0],mem[0],xmm2[2,3]
+; AVX2-NEXT: .LBB2_4: # %else2
+; AVX2-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX2-NEXT: vpcmpeqd %xmm3, %xmm1, %xmm1
+; AVX2-NEXT: vpextrb $8, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB2_6
+; AVX2-NEXT: # %bb.5: # %cond.load4
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm3
+; AVX2-NEXT: vmovq %xmm3, %rax
+; AVX2-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1],mem[0],xmm2[3]
+; AVX2-NEXT: .LBB2_6: # %else5
+; AVX2-NEXT: vpextrb $12, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB2_8
+; AVX2-NEXT: # %bb.7: # %cond.load7
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vinsertps {{.*#+}} xmm2 = xmm2[0,1,2],mem[0]
+; AVX2-NEXT: .LBB2_8: # %else8
+; AVX2-NEXT: vmovaps %xmm2, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
+;
+; AVX512-LABEL: gather_v4f32_v4i64_v4i32:
+; AVX512: # %bb.0:
+; AVX512-NEXT: # kill: def $xmm2 killed $xmm2 def $ymm2
+; AVX512-NEXT: # kill: def $xmm1 killed $xmm1 def $zmm1
+; AVX512-NEXT: # kill: def $ymm0 killed $ymm0 def $zmm0
+; AVX512-NEXT: vptestnmd %zmm1, %zmm1, %k0
+; AVX512-NEXT: kshiftlw $12, %k0, %k0
+; AVX512-NEXT: kshiftrw $12, %k0, %k1
+; AVX512-NEXT: vgatherqps (%rdi,%zmm0,4), %ymm2 {%k1}
+; AVX512-NEXT: vmovaps %xmm2, %xmm0
+; AVX512-NEXT: vzeroupper
+; AVX512-NEXT: retq
+ %vptr0 = insertelement <4 x float*> undef, float* %base, i32 0
+ %vptr1 = shufflevector <4 x float*> %vptr0, <4 x float*> undef, <4 x i32> zeroinitializer
+ %vptr2 = getelementptr float, <4 x float*> %vptr1, <4 x i64> %idx
+
+ %mask = icmp eq <4 x i32> %trigger, zeroinitializer
+ %res = call <4 x float> @llvm.masked.gather.v4f32.v4p0f32(<4 x float*> %vptr2, i32 4, <4 x i1> %mask, <4 x float> %passthru)
+ ret <4 x float> %res
+}
+
+;
+; vXi8
+;
+
+define <16 x i8> @gather_v16i8_v16i32_v16i8(i8* %base, <16 x i32> %idx, <16 x i8> %trigger, <16 x i8> %passthru) {
+; SSE-LABEL: gather_v16i8_v16i32_v16i8:
+; SSE: # %bb.0:
+; SSE-NEXT: movq %rdi, %xmm6
+; SSE-NEXT: pshufd {{.*#+}} xmm8 = xmm6[0,1,0,1]
+; SSE-NEXT: pshufd {{.*#+}} xmm7 = xmm0[2,3,0,1]
+; SSE-NEXT: pmovsxdq %xmm0, %xmm0
+; SSE-NEXT: paddq %xmm8, %xmm0
+; SSE-NEXT: pxor %xmm6, %xmm6
+; SSE-NEXT: pcmpeqb %xmm4, %xmm6
+; SSE-NEXT: pextrb $0, %xmm6, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_2
+; SSE-NEXT: # %bb.1: # %cond.load
+; SSE-NEXT: movq %xmm0, %rax
+; SSE-NEXT: pinsrb $0, (%rax), %xmm5
+; SSE-NEXT: .LBB3_2: # %else
+; SSE-NEXT: pmovsxdq %xmm7, %xmm7
+; SSE-NEXT: pextrb $1, %xmm6, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_4
+; SSE-NEXT: # %bb.3: # %cond.load1
+; SSE-NEXT: pextrq $1, %xmm0, %rax
+; SSE-NEXT: pinsrb $1, (%rax), %xmm5
+; SSE-NEXT: .LBB3_4: # %else2
+; SSE-NEXT: paddq %xmm8, %xmm7
+; SSE-NEXT: pxor %xmm6, %xmm6
+; SSE-NEXT: pcmpeqb %xmm4, %xmm6
+; SSE-NEXT: pextrb $2, %xmm6, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_6
+; SSE-NEXT: # %bb.5: # %cond.load4
+; SSE-NEXT: movq %xmm7, %rax
+; SSE-NEXT: pinsrb $2, (%rax), %xmm5
+; SSE-NEXT: .LBB3_6: # %else5
+; SSE-NEXT: pmovsxdq %xmm1, %xmm0
+; SSE-NEXT: pextrb $3, %xmm6, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_8
+; SSE-NEXT: # %bb.7: # %cond.load7
+; SSE-NEXT: pextrq $1, %xmm7, %rax
+; SSE-NEXT: pinsrb $3, (%rax), %xmm5
+; SSE-NEXT: .LBB3_8: # %else8
+; SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm1[2,3,0,1]
+; SSE-NEXT: paddq %xmm8, %xmm0
+; SSE-NEXT: pxor %xmm6, %xmm6
+; SSE-NEXT: pcmpeqb %xmm4, %xmm6
+; SSE-NEXT: pextrb $4, %xmm6, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_10
+; SSE-NEXT: # %bb.9: # %cond.load10
+; SSE-NEXT: movq %xmm0, %rax
+; SSE-NEXT: pinsrb $4, (%rax), %xmm5
+; SSE-NEXT: .LBB3_10: # %else11
+; SSE-NEXT: pmovsxdq %xmm1, %xmm1
+; SSE-NEXT: pextrb $5, %xmm6, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_12
+; SSE-NEXT: # %bb.11: # %cond.load13
+; SSE-NEXT: pextrq $1, %xmm0, %rax
+; SSE-NEXT: pinsrb $5, (%rax), %xmm5
+; SSE-NEXT: .LBB3_12: # %else14
+; SSE-NEXT: paddq %xmm8, %xmm1
+; SSE-NEXT: pxor %xmm6, %xmm6
+; SSE-NEXT: pcmpeqb %xmm4, %xmm6
+; SSE-NEXT: pextrb $6, %xmm6, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_14
+; SSE-NEXT: # %bb.13: # %cond.load16
+; SSE-NEXT: movq %xmm1, %rax
+; SSE-NEXT: pinsrb $6, (%rax), %xmm5
+; SSE-NEXT: .LBB3_14: # %else17
+; SSE-NEXT: pmovsxdq %xmm2, %xmm0
+; SSE-NEXT: pextrb $7, %xmm6, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_16
+; SSE-NEXT: # %bb.15: # %cond.load19
+; SSE-NEXT: pextrq $1, %xmm1, %rax
+; SSE-NEXT: pinsrb $7, (%rax), %xmm5
+; SSE-NEXT: .LBB3_16: # %else20
+; SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm2[2,3,0,1]
+; SSE-NEXT: paddq %xmm8, %xmm0
+; SSE-NEXT: pxor %xmm2, %xmm2
+; SSE-NEXT: pcmpeqb %xmm4, %xmm2
+; SSE-NEXT: pextrb $8, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_18
+; SSE-NEXT: # %bb.17: # %cond.load22
+; SSE-NEXT: movq %xmm0, %rax
+; SSE-NEXT: pinsrb $8, (%rax), %xmm5
+; SSE-NEXT: .LBB3_18: # %else23
+; SSE-NEXT: pmovsxdq %xmm1, %xmm1
+; SSE-NEXT: pextrb $9, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_20
+; SSE-NEXT: # %bb.19: # %cond.load25
+; SSE-NEXT: pextrq $1, %xmm0, %rax
+; SSE-NEXT: pinsrb $9, (%rax), %xmm5
+; SSE-NEXT: .LBB3_20: # %else26
+; SSE-NEXT: paddq %xmm8, %xmm1
+; SSE-NEXT: pxor %xmm2, %xmm2
+; SSE-NEXT: pcmpeqb %xmm4, %xmm2
+; SSE-NEXT: pextrb $10, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_22
+; SSE-NEXT: # %bb.21: # %cond.load28
+; SSE-NEXT: movq %xmm1, %rax
+; SSE-NEXT: pinsrb $10, (%rax), %xmm5
+; SSE-NEXT: .LBB3_22: # %else29
+; SSE-NEXT: pmovsxdq %xmm3, %xmm0
+; SSE-NEXT: pextrb $11, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_24
+; SSE-NEXT: # %bb.23: # %cond.load31
+; SSE-NEXT: pextrq $1, %xmm1, %rax
+; SSE-NEXT: pinsrb $11, (%rax), %xmm5
+; SSE-NEXT: .LBB3_24: # %else32
+; SSE-NEXT: pshufd {{.*#+}} xmm1 = xmm3[2,3,0,1]
+; SSE-NEXT: paddq %xmm8, %xmm0
+; SSE-NEXT: pxor %xmm2, %xmm2
+; SSE-NEXT: pcmpeqb %xmm4, %xmm2
+; SSE-NEXT: pextrb $12, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_26
+; SSE-NEXT: # %bb.25: # %cond.load34
+; SSE-NEXT: movq %xmm0, %rax
+; SSE-NEXT: pinsrb $12, (%rax), %xmm5
+; SSE-NEXT: .LBB3_26: # %else35
+; SSE-NEXT: pmovsxdq %xmm1, %xmm1
+; SSE-NEXT: pextrb $13, %xmm2, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_28
+; SSE-NEXT: # %bb.27: # %cond.load37
+; SSE-NEXT: pextrq $1, %xmm0, %rax
+; SSE-NEXT: pinsrb $13, (%rax), %xmm5
+; SSE-NEXT: .LBB3_28: # %else38
+; SSE-NEXT: paddq %xmm1, %xmm8
+; SSE-NEXT: pxor %xmm0, %xmm0
+; SSE-NEXT: pcmpeqb %xmm0, %xmm4
+; SSE-NEXT: pextrb $14, %xmm4, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_30
+; SSE-NEXT: # %bb.29: # %cond.load40
+; SSE-NEXT: movq %xmm8, %rax
+; SSE-NEXT: pinsrb $14, (%rax), %xmm5
+; SSE-NEXT: .LBB3_30: # %else41
+; SSE-NEXT: pextrb $15, %xmm4, %eax
+; SSE-NEXT: testb $1, %al
+; SSE-NEXT: je .LBB3_32
+; SSE-NEXT: # %bb.31: # %cond.load43
+; SSE-NEXT: pextrq $1, %xmm8, %rax
+; SSE-NEXT: pinsrb $15, (%rax), %xmm5
+; SSE-NEXT: .LBB3_32: # %else44
+; SSE-NEXT: movdqa %xmm5, %xmm0
+; SSE-NEXT: retq
+;
+; AVX1-LABEL: gather_v16i8_v16i32_v16i8:
+; AVX1: # %bb.0:
+; AVX1-NEXT: vmovq %rdi, %xmm4
+; AVX1-NEXT: vpshufd {{.*#+}} xmm4 = xmm4[0,1,0,1]
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm5
+; AVX1-NEXT: vpshufd {{.*#+}} xmm6 = xmm0[2,3,0,1]
+; AVX1-NEXT: vpmovsxdq %xmm6, %xmm6
+; AVX1-NEXT: vpaddq %xmm6, %xmm4, %xmm6
+; AVX1-NEXT: vpmovsxdq %xmm0, %xmm0
+; AVX1-NEXT: vpaddq %xmm0, %xmm4, %xmm0
+; AVX1-NEXT: vinsertf128 $1, %xmm6, %ymm0, %ymm0
+; AVX1-NEXT: vpxor %xmm6, %xmm6, %xmm6
+; AVX1-NEXT: vpcmpeqb %xmm6, %xmm2, %xmm6
+; AVX1-NEXT: vpextrb $0, %xmm6, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_2
+; AVX1-NEXT: # %bb.1: # %cond.load
+; AVX1-NEXT: vmovq %xmm0, %rax
+; AVX1-NEXT: vpinsrb $0, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_2: # %else
+; AVX1-NEXT: vpshufd {{.*#+}} xmm7 = xmm5[2,3,0,1]
+; AVX1-NEXT: vpmovsxdq %xmm5, %xmm5
+; AVX1-NEXT: vpextrb $1, %xmm6, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_4
+; AVX1-NEXT: # %bb.3: # %cond.load1
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vpinsrb $1, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_4: # %else2
+; AVX1-NEXT: vpmovsxdq %xmm7, %xmm6
+; AVX1-NEXT: vpaddq %xmm5, %xmm4, %xmm8
+; AVX1-NEXT: vpxor %xmm7, %xmm7, %xmm7
+; AVX1-NEXT: vpcmpeqb %xmm7, %xmm2, %xmm7
+; AVX1-NEXT: vpextrb $2, %xmm7, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_6
+; AVX1-NEXT: # %bb.5: # %cond.load4
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm5
+; AVX1-NEXT: vmovq %xmm5, %rax
+; AVX1-NEXT: vpinsrb $2, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_6: # %else5
+; AVX1-NEXT: vpaddq %xmm6, %xmm4, %xmm6
+; AVX1-NEXT: vpextrb $3, %xmm7, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_8
+; AVX1-NEXT: # %bb.7: # %cond.load7
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vpinsrb $3, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_8: # %else8
+; AVX1-NEXT: vinsertf128 $1, %xmm6, %ymm8, %ymm0
+; AVX1-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX1-NEXT: vpcmpeqb %xmm5, %xmm2, %xmm5
+; AVX1-NEXT: vpextrb $4, %xmm5, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_10
+; AVX1-NEXT: # %bb.9: # %cond.load10
+; AVX1-NEXT: vmovq %xmm0, %rax
+; AVX1-NEXT: vpinsrb $4, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_10: # %else11
+; AVX1-NEXT: vpshufd {{.*#+}} xmm6 = xmm1[2,3,0,1]
+; AVX1-NEXT: vpmovsxdq %xmm1, %xmm7
+; AVX1-NEXT: vpextrb $5, %xmm5, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_12
+; AVX1-NEXT: # %bb.11: # %cond.load13
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vpinsrb $5, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_12: # %else14
+; AVX1-NEXT: vpmovsxdq %xmm6, %xmm6
+; AVX1-NEXT: vpaddq %xmm7, %xmm4, %xmm8
+; AVX1-NEXT: vpxor %xmm7, %xmm7, %xmm7
+; AVX1-NEXT: vpcmpeqb %xmm7, %xmm2, %xmm7
+; AVX1-NEXT: vpextrb $6, %xmm7, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_14
+; AVX1-NEXT: # %bb.13: # %cond.load16
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm5
+; AVX1-NEXT: vmovq %xmm5, %rax
+; AVX1-NEXT: vpinsrb $6, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_14: # %else17
+; AVX1-NEXT: vpaddq %xmm6, %xmm4, %xmm6
+; AVX1-NEXT: vpextrb $7, %xmm7, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_16
+; AVX1-NEXT: # %bb.15: # %cond.load19
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vpinsrb $7, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_16: # %else20
+; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm1
+; AVX1-NEXT: vinsertf128 $1, %xmm6, %ymm8, %ymm0
+; AVX1-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX1-NEXT: vpcmpeqb %xmm5, %xmm2, %xmm5
+; AVX1-NEXT: vpextrb $8, %xmm5, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_18
+; AVX1-NEXT: # %bb.17: # %cond.load22
+; AVX1-NEXT: vmovq %xmm0, %rax
+; AVX1-NEXT: vpinsrb $8, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_18: # %else23
+; AVX1-NEXT: vpshufd {{.*#+}} xmm6 = xmm1[2,3,0,1]
+; AVX1-NEXT: vpmovsxdq %xmm1, %xmm1
+; AVX1-NEXT: vpextrb $9, %xmm5, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_20
+; AVX1-NEXT: # %bb.19: # %cond.load25
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vpinsrb $9, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_20: # %else26
+; AVX1-NEXT: vpmovsxdq %xmm6, %xmm5
+; AVX1-NEXT: vpaddq %xmm1, %xmm4, %xmm1
+; AVX1-NEXT: vpxor %xmm6, %xmm6, %xmm6
+; AVX1-NEXT: vpcmpeqb %xmm6, %xmm2, %xmm6
+; AVX1-NEXT: vpextrb $10, %xmm6, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_22
+; AVX1-NEXT: # %bb.21: # %cond.load28
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm7
+; AVX1-NEXT: vmovq %xmm7, %rax
+; AVX1-NEXT: vpinsrb $10, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_22: # %else29
+; AVX1-NEXT: vpaddq %xmm5, %xmm4, %xmm4
+; AVX1-NEXT: vpextrb $11, %xmm6, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_24
+; AVX1-NEXT: # %bb.23: # %cond.load31
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vpinsrb $11, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_24: # %else32
+; AVX1-NEXT: vinsertf128 $1, %xmm4, %ymm1, %ymm0
+; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT: vpcmpeqb %xmm1, %xmm2, %xmm1
+; AVX1-NEXT: vpextrb $12, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_26
+; AVX1-NEXT: # %bb.25: # %cond.load34
+; AVX1-NEXT: vmovq %xmm0, %rax
+; AVX1-NEXT: vpinsrb $12, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_26: # %else35
+; AVX1-NEXT: vpextrb $13, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_28
+; AVX1-NEXT: # %bb.27: # %cond.load37
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vpinsrb $13, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_28: # %else38
+; AVX1-NEXT: vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT: vpcmpeqb %xmm1, %xmm2, %xmm1
+; AVX1-NEXT: vpextrb $14, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_30
+; AVX1-NEXT: # %bb.29: # %cond.load40
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm2
+; AVX1-NEXT: vmovq %xmm2, %rax
+; AVX1-NEXT: vpinsrb $14, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_30: # %else41
+; AVX1-NEXT: vpextrb $15, %xmm1, %eax
+; AVX1-NEXT: testb $1, %al
+; AVX1-NEXT: je .LBB3_32
+; AVX1-NEXT: # %bb.31: # %cond.load43
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpextrq $1, %xmm0, %rax
+; AVX1-NEXT: vpinsrb $15, (%rax), %xmm3, %xmm3
+; AVX1-NEXT: .LBB3_32: # %else44
+; AVX1-NEXT: vmovdqa %xmm3, %xmm0
+; AVX1-NEXT: vzeroupper
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: gather_v16i8_v16i32_v16i8:
+; AVX2: # %bb.0:
+; AVX2-NEXT: vmovq %rdi, %xmm4
+; AVX2-NEXT: vpbroadcastq %xmm4, %ymm4
+; AVX2-NEXT: vpmovsxdq %xmm0, %ymm5
+; AVX2-NEXT: vpaddq %ymm5, %ymm4, %ymm5
+; AVX2-NEXT: vpxor %xmm6, %xmm6, %xmm6
+; AVX2-NEXT: vpcmpeqb %xmm6, %xmm2, %xmm6
+; AVX2-NEXT: vpextrb $0, %xmm6, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_2
+; AVX2-NEXT: # %bb.1: # %cond.load
+; AVX2-NEXT: vmovq %xmm5, %rax
+; AVX2-NEXT: vpinsrb $0, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_2: # %else
+; AVX2-NEXT: vpextrb $1, %xmm6, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_4
+; AVX2-NEXT: # %bb.3: # %cond.load1
+; AVX2-NEXT: vpextrq $1, %xmm5, %rax
+; AVX2-NEXT: vpinsrb $1, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_4: # %else2
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT: vpxor %xmm6, %xmm6, %xmm6
+; AVX2-NEXT: vpcmpeqb %xmm6, %xmm2, %xmm6
+; AVX2-NEXT: vpextrb $2, %xmm6, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_6
+; AVX2-NEXT: # %bb.5: # %cond.load4
+; AVX2-NEXT: vextracti128 $1, %ymm5, %xmm7
+; AVX2-NEXT: vmovq %xmm7, %rax
+; AVX2-NEXT: vpinsrb $2, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_6: # %else5
+; AVX2-NEXT: vpmovsxdq %xmm0, %ymm0
+; AVX2-NEXT: vpextrb $3, %xmm6, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_8
+; AVX2-NEXT: # %bb.7: # %cond.load7
+; AVX2-NEXT: vextracti128 $1, %ymm5, %xmm5
+; AVX2-NEXT: vpextrq $1, %xmm5, %rax
+; AVX2-NEXT: vpinsrb $3, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_8: # %else8
+; AVX2-NEXT: vpaddq %ymm0, %ymm4, %ymm0
+; AVX2-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX2-NEXT: vpcmpeqb %xmm5, %xmm2, %xmm5
+; AVX2-NEXT: vpextrb $4, %xmm5, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_10
+; AVX2-NEXT: # %bb.9: # %cond.load10
+; AVX2-NEXT: vmovq %xmm0, %rax
+; AVX2-NEXT: vpinsrb $4, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_10: # %else11
+; AVX2-NEXT: vpextrb $5, %xmm5, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_12
+; AVX2-NEXT: # %bb.11: # %cond.load13
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vpinsrb $5, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_12: # %else14
+; AVX2-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX2-NEXT: vpcmpeqb %xmm5, %xmm2, %xmm5
+; AVX2-NEXT: vpextrb $6, %xmm5, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_14
+; AVX2-NEXT: # %bb.13: # %cond.load16
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm6
+; AVX2-NEXT: vmovq %xmm6, %rax
+; AVX2-NEXT: vpinsrb $6, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_14: # %else17
+; AVX2-NEXT: vpmovsxdq %xmm1, %ymm6
+; AVX2-NEXT: vpextrb $7, %xmm5, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_16
+; AVX2-NEXT: # %bb.15: # %cond.load19
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vpinsrb $7, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_16: # %else20
+; AVX2-NEXT: vpaddq %ymm6, %ymm4, %ymm0
+; AVX2-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX2-NEXT: vpcmpeqb %xmm5, %xmm2, %xmm5
+; AVX2-NEXT: vpextrb $8, %xmm5, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_18
+; AVX2-NEXT: # %bb.17: # %cond.load22
+; AVX2-NEXT: vmovq %xmm0, %rax
+; AVX2-NEXT: vpinsrb $8, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_18: # %else23
+; AVX2-NEXT: vpextrb $9, %xmm5, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_20
+; AVX2-NEXT: # %bb.19: # %cond.load25
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vpinsrb $9, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_20: # %else26
+; AVX2-NEXT: vextracti128 $1, %ymm1, %xmm1
+; AVX2-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX2-NEXT: vpcmpeqb %xmm5, %xmm2, %xmm5
+; AVX2-NEXT: vpextrb $10, %xmm5, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_22
+; AVX2-NEXT: # %bb.21: # %cond.load28
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm6
+; AVX2-NEXT: vmovq %xmm6, %rax
+; AVX2-NEXT: vpinsrb $10, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_22: # %else29
+; AVX2-NEXT: vpmovsxdq %xmm1, %ymm1
+; AVX2-NEXT: vpextrb $11, %xmm5, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_24
+; AVX2-NEXT: # %bb.23: # %cond.load31
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vpinsrb $11, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_24: # %else32
+; AVX2-NEXT: vpaddq %ymm1, %ymm4, %ymm0
+; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
+; AVX2-NEXT: vpcmpeqb %xmm1, %xmm2, %xmm1
+; AVX2-NEXT: vpextrb $12, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_26
+; AVX2-NEXT: # %bb.25: # %cond.load34
+; AVX2-NEXT: vmovq %xmm0, %rax
+; AVX2-NEXT: vpinsrb $12, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_26: # %else35
+; AVX2-NEXT: vpextrb $13, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_28
+; AVX2-NEXT: # %bb.27: # %cond.load37
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vpinsrb $13, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_28: # %else38
+; AVX2-NEXT: vpxor %xmm1, %xmm1, %xmm1
+; AVX2-NEXT: vpcmpeqb %xmm1, %xmm2, %xmm1
+; AVX2-NEXT: vpextrb $14, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_30
+; AVX2-NEXT: # %bb.29: # %cond.load40
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm2
+; AVX2-NEXT: vmovq %xmm2, %rax
+; AVX2-NEXT: vpinsrb $14, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_30: # %else41
+; AVX2-NEXT: vpextrb $15, %xmm1, %eax
+; AVX2-NEXT: testb $1, %al
+; AVX2-NEXT: je .LBB3_32
+; AVX2-NEXT: # %bb.31: # %cond.load43
+; AVX2-NEXT: vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT: vpextrq $1, %xmm0, %rax
+; AVX2-NEXT: vpinsrb $15, (%rax), %xmm3, %xmm3
+; AVX2-NEXT: .LBB3_32: # %else44
+; AVX2-NEXT: vmovdqa %xmm3, %xmm0
+; AVX2-NEXT: vzeroupper
+; AVX2-NEXT: retq
+;
+; AVX512-LABEL: gather_v16i8_v16i32_v16i8:
+; AVX512: # %bb.0:
+; AVX512-NEXT: vpbroadcastq %rdi, %zmm3
+; AVX512-NEXT: vpmovsxdq %ymm0, %zmm4
+; AVX512-NEXT: vpaddq %zmm4, %zmm3, %zmm4
+; AVX512-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX512-NEXT: vpcmpeqb %xmm5, %xmm1, %xmm5
+; AVX512-NEXT: vpmovsxbd %xmm5, %zmm5
+; AVX512-NEXT: vptestmd %zmm5, %zmm5, %k0
+; AVX512-NEXT: kmovw %k0, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_2
+; AVX512-NEXT: # %bb.1: # %cond.load
+; AVX512-NEXT: vmovq %xmm4, %rax
+; AVX512-NEXT: vpinsrb $0, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_2: # %else
+; AVX512-NEXT: kshiftrw $1, %k0, %k0
+; AVX512-NEXT: kmovw %k0, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_4
+; AVX512-NEXT: # %bb.3: # %cond.load1
+; AVX512-NEXT: vpextrq $1, %xmm4, %rax
+; AVX512-NEXT: vpinsrb $1, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_4: # %else2
+; AVX512-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX512-NEXT: vpcmpeqb %xmm5, %xmm1, %xmm5
+; AVX512-NEXT: vpmovsxbd %xmm5, %zmm5
+; AVX512-NEXT: vptestmd %zmm5, %zmm5, %k0
+; AVX512-NEXT: kshiftrw $2, %k0, %k1
+; AVX512-NEXT: kmovw %k1, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_6
+; AVX512-NEXT: # %bb.5: # %cond.load4
+; AVX512-NEXT: vextracti128 $1, %ymm4, %xmm5
+; AVX512-NEXT: vmovq %xmm5, %rax
+; AVX512-NEXT: vpinsrb $2, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_6: # %else5
+; AVX512-NEXT: kshiftrw $3, %k0, %k0
+; AVX512-NEXT: kmovw %k0, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_8
+; AVX512-NEXT: # %bb.7: # %cond.load7
+; AVX512-NEXT: vextracti128 $1, %ymm4, %xmm5
+; AVX512-NEXT: vpextrq $1, %xmm5, %rax
+; AVX512-NEXT: vpinsrb $3, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_8: # %else8
+; AVX512-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX512-NEXT: vpcmpeqb %xmm5, %xmm1, %xmm5
+; AVX512-NEXT: vpmovsxbd %xmm5, %zmm5
+; AVX512-NEXT: vptestmd %zmm5, %zmm5, %k0
+; AVX512-NEXT: kshiftrw $4, %k0, %k1
+; AVX512-NEXT: kmovw %k1, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_10
+; AVX512-NEXT: # %bb.9: # %cond.load10
+; AVX512-NEXT: vextracti32x4 $2, %zmm4, %xmm5
+; AVX512-NEXT: vmovq %xmm5, %rax
+; AVX512-NEXT: vpinsrb $4, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_10: # %else11
+; AVX512-NEXT: kshiftrw $5, %k0, %k0
+; AVX512-NEXT: kmovw %k0, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_12
+; AVX512-NEXT: # %bb.11: # %cond.load13
+; AVX512-NEXT: vextracti32x4 $2, %zmm4, %xmm5
+; AVX512-NEXT: vpextrq $1, %xmm5, %rax
+; AVX512-NEXT: vpinsrb $5, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_12: # %else14
+; AVX512-NEXT: vextracti64x4 $1, %zmm0, %ymm0
+; AVX512-NEXT: vpxor %xmm5, %xmm5, %xmm5
+; AVX512-NEXT: vpcmpeqb %xmm5, %xmm1, %xmm5
+; AVX512-NEXT: vpmovsxbd %xmm5, %zmm5
+; AVX512-NEXT: vptestmd %zmm5, %zmm5, %k0
+; AVX512-NEXT: kshiftrw $6, %k0, %k1
+; AVX512-NEXT: kmovw %k1, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_14
+; AVX512-NEXT: # %bb.13: # %cond.load16
+; AVX512-NEXT: vextracti32x4 $3, %zmm4, %xmm5
+; AVX512-NEXT: vmovq %xmm5, %rax
+; AVX512-NEXT: vpinsrb $6, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_14: # %else17
+; AVX512-NEXT: vpmovsxdq %ymm0, %zmm0
+; AVX512-NEXT: kshiftrw $7, %k0, %k0
+; AVX512-NEXT: kmovw %k0, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_16
+; AVX512-NEXT: # %bb.15: # %cond.load19
+; AVX512-NEXT: vextracti32x4 $3, %zmm4, %xmm4
+; AVX512-NEXT: vpextrq $1, %xmm4, %rax
+; AVX512-NEXT: vpinsrb $7, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_16: # %else20
+; AVX512-NEXT: vpaddq %zmm0, %zmm3, %zmm0
+; AVX512-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX512-NEXT: vpcmpeqb %xmm3, %xmm1, %xmm3
+; AVX512-NEXT: vpmovsxbd %xmm3, %zmm3
+; AVX512-NEXT: vptestmd %zmm3, %zmm3, %k0
+; AVX512-NEXT: kshiftrw $8, %k0, %k1
+; AVX512-NEXT: kmovw %k1, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_18
+; AVX512-NEXT: # %bb.17: # %cond.load22
+; AVX512-NEXT: vmovq %xmm0, %rax
+; AVX512-NEXT: vpinsrb $8, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_18: # %else23
+; AVX512-NEXT: kshiftrw $9, %k0, %k0
+; AVX512-NEXT: kmovw %k0, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_20
+; AVX512-NEXT: # %bb.19: # %cond.load25
+; AVX512-NEXT: vpextrq $1, %xmm0, %rax
+; AVX512-NEXT: vpinsrb $9, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_20: # %else26
+; AVX512-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX512-NEXT: vpcmpeqb %xmm3, %xmm1, %xmm3
+; AVX512-NEXT: vpmovsxbd %xmm3, %zmm3
+; AVX512-NEXT: vptestmd %zmm3, %zmm3, %k0
+; AVX512-NEXT: kshiftrw $10, %k0, %k1
+; AVX512-NEXT: kmovw %k1, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_22
+; AVX512-NEXT: # %bb.21: # %cond.load28
+; AVX512-NEXT: vextracti128 $1, %ymm0, %xmm3
+; AVX512-NEXT: vmovq %xmm3, %rax
+; AVX512-NEXT: vpinsrb $10, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_22: # %else29
+; AVX512-NEXT: kshiftrw $11, %k0, %k0
+; AVX512-NEXT: kmovw %k0, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_24
+; AVX512-NEXT: # %bb.23: # %cond.load31
+; AVX512-NEXT: vextracti128 $1, %ymm0, %xmm3
+; AVX512-NEXT: vpextrq $1, %xmm3, %rax
+; AVX512-NEXT: vpinsrb $11, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_24: # %else32
+; AVX512-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX512-NEXT: vpcmpeqb %xmm3, %xmm1, %xmm3
+; AVX512-NEXT: vpmovsxbd %xmm3, %zmm3
+; AVX512-NEXT: vptestmd %zmm3, %zmm3, %k0
+; AVX512-NEXT: kshiftrw $12, %k0, %k1
+; AVX512-NEXT: kmovw %k1, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_26
+; AVX512-NEXT: # %bb.25: # %cond.load34
+; AVX512-NEXT: vextracti32x4 $2, %zmm0, %xmm3
+; AVX512-NEXT: vmovq %xmm3, %rax
+; AVX512-NEXT: vpinsrb $12, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_26: # %else35
+; AVX512-NEXT: kshiftrw $13, %k0, %k0
+; AVX512-NEXT: kmovw %k0, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_28
+; AVX512-NEXT: # %bb.27: # %cond.load37
+; AVX512-NEXT: vextracti32x4 $2, %zmm0, %xmm3
+; AVX512-NEXT: vpextrq $1, %xmm3, %rax
+; AVX512-NEXT: vpinsrb $13, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_28: # %else38
+; AVX512-NEXT: vpxor %xmm3, %xmm3, %xmm3
+; AVX512-NEXT: vpcmpeqb %xmm3, %xmm1, %xmm1
+; AVX512-NEXT: vpmovsxbd %xmm1, %zmm1
+; AVX512-NEXT: vptestmd %zmm1, %zmm1, %k0
+; AVX512-NEXT: kshiftrw $14, %k0, %k1
+; AVX512-NEXT: kmovw %k1, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_30
+; AVX512-NEXT: # %bb.29: # %cond.load40
+; AVX512-NEXT: vextracti32x4 $3, %zmm0, %xmm1
+; AVX512-NEXT: vmovq %xmm1, %rax
+; AVX512-NEXT: vpinsrb $14, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_30: # %else41
+; AVX512-NEXT: kshiftrw $15, %k0, %k0
+; AVX512-NEXT: kmovw %k0, %eax
+; AVX512-NEXT: testb $1, %al
+; AVX512-NEXT: je .LBB3_32
+; AVX512-NEXT: # %bb.31: # %cond.load43
+; AVX512-NEXT: vextracti32x4 $3, %zmm0, %xmm0
+; AVX512-NEXT: vpextrq $1, %xmm0, %rax
+; AVX512-NEXT: vpinsrb $15, (%rax), %xmm2, %xmm2
+; AVX512-NEXT: .LBB3_32: # %else44
+; AVX512-NEXT: vmovdqa %xmm2, %xmm0
+; AVX512-NEXT: vzeroupper
+; AVX512-NEXT: retq
+ %vptr0 = insertelement <16 x i8*> undef, i8* %base, i32 0
+ %vptr1 = shufflevector <16 x i8*> %vptr0, <16 x i8*> undef, <16 x i32> zeroinitializer
+ %vptr2 = getelementptr i8, <16 x i8*> %vptr1, <16 x i32> %idx
+
+ %mask = icmp eq <16 x i8> %trigger, zeroinitializer
+ %res = call <16 x i8> @llvm.masked.gather.v16i8.v16p0i8(<16 x i8*> %vptr2, i32 4, <16 x i1> %mask, <16 x i8> %passthru)
+ ret <16 x i8> %res
+}
+
+declare <2 x double> @llvm.masked.gather.v2f64.v2p0f64(<2 x double*>, i32, <2 x i1>, <2 x double>)
+declare <4 x double> @llvm.masked.gather.v4f64.v4p0f64(<4 x double*>, i32, <4 x i1>, <4 x double>)
+
+declare <4 x float> @llvm.masked.gather.v4f32.v4p0f32(<4 x float*>, i32, <4 x i1>, <4 x float>)
+declare <8 x float> @llvm.masked.gather.v8f32.v8p0f32(<8 x float*>, i32, <8 x i1>, <8 x float>)
+
+declare <16 x i8> @llvm.masked.gather.v16i8.v16p0i8(<16 x i8*>, i32, <16 x i1>, <16 x i8>)
More information about the llvm-commits
mailing list