diff options
Diffstat (limited to 'test/CodeGen/X86/vector-shuffle-avx512.ll')
-rw-r--r-- | test/CodeGen/X86/vector-shuffle-avx512.ll | 333 |
1 files changed, 333 insertions, 0 deletions
diff --git a/test/CodeGen/X86/vector-shuffle-avx512.ll b/test/CodeGen/X86/vector-shuffle-avx512.ll new file mode 100644 index 000000000000..defc3e918b24 --- /dev/null +++ b/test/CodeGen/X86/vector-shuffle-avx512.ll @@ -0,0 +1,333 @@ +; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py +; RUN: llc < %s -mtriple=x86_64-pc-linux-gnu -mcpu=skx | FileCheck %s --check-prefix=SKX +; RUN: llc < %s -mtriple=x86_64-pc-linux-gnu -mcpu=knl | FileCheck %s --check-prefix=KNL + +;expand 128 -> 256 include <4 x float> <2 x double> +define <8 x float> @expand(<4 x float> %a) { +; SKX-LABEL: expand: +; SKX: # BB#0: +; SKX-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; SKX-NEXT: movb $5, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vexpandps %ymm0, %ymm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand: +; KNL: # BB#0: +; KNL-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[0,1,1,3] +; KNL-NEXT: vxorps %ymm1, %ymm1, %ymm1 +; KNL-NEXT: vblendps {{.*#+}} ymm0 = ymm0[0],ymm1[1],ymm0[2],ymm1[3,4,5,6,7] +; KNL-NEXT: retq + %res = shufflevector <4 x float> %a, <4 x float> zeroinitializer, <8 x i32> <i32 0, i32 5, i32 1, i32 5, i32 5, i32 5, i32 5, i32 5> + ret <8 x float> %res +} + +define <8 x float> @expand1(<4 x float> %a ) { +; SKX-LABEL: expand1: +; SKX: # BB#0: +; SKX-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; SKX-NEXT: movb $-86, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vexpandps %ymm0, %ymm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand1: +; KNL: # BB#0: +; KNL-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; KNL-NEXT: vmovaps {{.*#+}} ymm1 = <u,0,u,1,u,2,u,3> +; KNL-NEXT: vpermps %ymm0, %ymm1, %ymm0 +; KNL-NEXT: vxorps %ymm1, %ymm1, %ymm1 +; KNL-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7] +; KNL-NEXT: retq + %res = shufflevector <4 x float> zeroinitializer, <4 x float> %a, <8 x i32> <i32 0, i32 4, i32 1, i32 5, i32 2, i32 6, i32 3, i32 7> + ret <8 x float> %res +} + +;Expand 128 -> 256 test <2 x double> -> <4 x double> +define <4 x double> @expand2(<2 x double> %a) { +; SKX-LABEL: expand2: +; SKX: # BB#0: +; SKX-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; SKX-NEXT: movb $9, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vexpandpd %ymm0, %ymm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand2: +; KNL: # BB#0: +; KNL-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; KNL-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,1,2,1] +; KNL-NEXT: vxorpd %ymm1, %ymm1, %ymm1 +; KNL-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2],ymm0[3] +; KNL-NEXT: retq + %res = shufflevector <2 x double> %a, <2 x double> zeroinitializer, <4 x i32> <i32 0, i32 2, i32 2, i32 1> + ret <4 x double> %res +} + +;expand 128 -> 256 include case <4 x i32> <8 x i32> +define <8 x i32> @expand3(<4 x i32> %a ) { +; SKX-LABEL: expand3: +; SKX: # BB#0: +; SKX-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; SKX-NEXT: movb $-127, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vpexpandd %ymm0, %ymm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand3: +; KNL: # BB#0: +; KNL-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; KNL-NEXT: vpbroadcastq %xmm0, %ymm0 +; KNL-NEXT: vpxor %ymm1, %ymm1, %ymm1 +; KNL-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3,4,5,6],ymm0[7] +; KNL-NEXT: retq + %res = shufflevector <4 x i32> zeroinitializer, <4 x i32> %a, <8 x i32> <i32 4, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0,i32 5> + ret <8 x i32> %res +} + +;expand 128 -> 256 include case <2 x i64> <4 x i64> +define <4 x i64> @expand4(<2 x i64> %a ) { +; SKX-LABEL: expand4: +; SKX: # BB#0: +; SKX-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; SKX-NEXT: movb $9, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vpexpandq %ymm0, %ymm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand4: +; KNL: # BB#0: +; KNL-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; KNL-NEXT: vpermq {{.*#+}} ymm0 = ymm0[0,1,2,1] +; KNL-NEXT: vpxor %ymm1, %ymm1, %ymm1 +; KNL-NEXT: vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5],ymm0[6,7] +; KNL-NEXT: retq + %res = shufflevector <2 x i64> zeroinitializer, <2 x i64> %a, <4 x i32> <i32 2, i32 0, i32 0, i32 3> + ret <4 x i64> %res +} + +;Negative test for 128-> 256 +define <8 x float> @expand5(<4 x float> %a ) { +; SKX-LABEL: expand5: +; SKX: # BB#0: +; SKX-NEXT: vbroadcastss %xmm0, %ymm0 +; SKX-NEXT: vxorps %ymm1, %ymm1, %ymm1 +; SKX-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7] +; SKX-NEXT: retq +; +; KNL-LABEL: expand5: +; KNL: # BB#0: +; KNL-NEXT: vbroadcastss %xmm0, %ymm0 +; KNL-NEXT: vxorps %ymm1, %ymm1, %ymm1 +; KNL-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0],ymm0[1],ymm1[2],ymm0[3],ymm1[4],ymm0[5],ymm1[6],ymm0[7] +; KNL-NEXT: retq + %res = shufflevector <4 x float> zeroinitializer, <4 x float> %a, <8 x i32> <i32 0, i32 4, i32 1, i32 4, i32 2, i32 4, i32 3, i32 4> + ret <8 x float> %res +} + +;expand 256 -> 512 include <8 x float> <16 x float> +define <8 x float> @expand6(<4 x float> %a ) { +; SKX-LABEL: expand6: +; SKX: # BB#0: +; SKX-NEXT: vxorps %xmm1, %xmm1, %xmm1 +; SKX-NEXT: vinsertf{{.*}}$1, %xmm0, %ymm1, %ymm0 +; SKX-NEXT: retq +; +; KNL-LABEL: expand6: +; KNL: # BB#0: +; KNL-NEXT: vxorps %xmm1, %xmm1, %xmm1 +; KNL-NEXT: vinsertf128 $1, %xmm0, %ymm1, %ymm0 +; KNL-NEXT: retq + %res = shufflevector <4 x float> zeroinitializer, <4 x float> %a, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7> + ret <8 x float> %res +} + +define <16 x float> @expand7(<8 x float> %a) { +; SKX-LABEL: expand7: +; SKX: # BB#0: +; SKX-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; SKX-NEXT: movw $1285, %ax # imm = 0x505 +; SKX-NEXT: kmovw %eax, %k1 +; SKX-NEXT: vexpandps %zmm0, %zmm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand7: +; KNL: # BB#0: +; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; KNL-NEXT: movw $1285, %ax # imm = 0x505 +; KNL-NEXT: kmovw %eax, %k1 +; KNL-NEXT: vexpandps %zmm0, %zmm0 {%k1} {z} +; KNL-NEXT: retq + %res = shufflevector <8 x float> %a, <8 x float> zeroinitializer, <16 x i32> <i32 0, i32 8, i32 1, i32 8, i32 8, i32 8, i32 8, i32 8, i32 2, i32 8, i32 3, i32 8, i32 8, i32 8, i32 8, i32 8> + ret <16 x float> %res +} + +define <16 x float> @expand8(<8 x float> %a ) { +; SKX-LABEL: expand8: +; SKX: # BB#0: +; SKX-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; SKX-NEXT: movw $-21846, %ax # imm = 0xAAAA +; SKX-NEXT: kmovw %eax, %k1 +; SKX-NEXT: vexpandps %zmm0, %zmm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand8: +; KNL: # BB#0: +; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; KNL-NEXT: movw $-21846, %ax # imm = 0xAAAA +; KNL-NEXT: kmovw %eax, %k1 +; KNL-NEXT: vexpandps %zmm0, %zmm0 {%k1} {z} +; KNL-NEXT: retq + %res = shufflevector <8 x float> zeroinitializer, <8 x float> %a, <16 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11, i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15> + ret <16 x float> %res +} + +;expand 256 -> 512 include <4 x double> <8 x double> +define <8 x double> @expand9(<4 x double> %a) { +; SKX-LABEL: expand9: +; SKX: # BB#0: +; SKX-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; SKX-NEXT: movb $-127, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vexpandpd %zmm0, %zmm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand9: +; KNL: # BB#0: +; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; KNL-NEXT: movb $-127, %al +; KNL-NEXT: kmovw %eax, %k1 +; KNL-NEXT: vexpandpd %zmm0, %zmm0 {%k1} {z} +; KNL-NEXT: retq + %res = shufflevector <4 x double> %a, <4 x double> zeroinitializer, <8 x i32> <i32 0, i32 4, i32 4, i32 4, i32 4, i32 4, i32 4, i32 1> + ret <8 x double> %res +} + +define <16 x i32> @expand10(<8 x i32> %a ) { +; SKX-LABEL: expand10: +; SKX: # BB#0: +; SKX-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; SKX-NEXT: movw $-21846, %ax # imm = 0xAAAA +; SKX-NEXT: kmovw %eax, %k1 +; SKX-NEXT: vpexpandd %zmm0, %zmm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand10: +; KNL: # BB#0: +; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; KNL-NEXT: movw $-21846, %ax # imm = 0xAAAA +; KNL-NEXT: kmovw %eax, %k1 +; KNL-NEXT: vpexpandd %zmm0, %zmm0 {%k1} {z} +; KNL-NEXT: retq + %res = shufflevector <8 x i32> zeroinitializer, <8 x i32> %a, <16 x i32> <i32 0, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11, i32 4, i32 12, i32 5, i32 13, i32 6, i32 14, i32 7, i32 15> + ret <16 x i32> %res +} + +define <8 x i64> @expand11(<4 x i64> %a) { +; SKX-LABEL: expand11: +; SKX: # BB#0: +; SKX-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; SKX-NEXT: movb $-127, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vpexpandq %zmm0, %zmm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand11: +; KNL: # BB#0: +; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; KNL-NEXT: movb $-127, %al +; KNL-NEXT: kmovw %eax, %k1 +; KNL-NEXT: vpexpandq %zmm0, %zmm0 {%k1} {z} +; KNL-NEXT: retq + %res = shufflevector <4 x i64> %a, <4 x i64> zeroinitializer, <8 x i32> <i32 0, i32 4, i32 4, i32 4, i32 4, i32 4, i32 4, i32 1> + ret <8 x i64> %res +} + +;Negative test for 256-> 512 +define <16 x float> @expand12(<8 x float> %a) { +; SKX-LABEL: expand12: +; SKX: # BB#0: +; SKX-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; SKX-NEXT: vmovaps {{.*#+}} zmm2 = [0,16,2,16,4,16,6,16,0,16,1,16,2,16,3,16] +; SKX-NEXT: vxorps %zmm1, %zmm1, %zmm1 +; SKX-NEXT: vpermt2ps %zmm0, %zmm2, %zmm1 +; SKX-NEXT: vmovaps %zmm1, %zmm0 +; SKX-NEXT: retq +; +; KNL-LABEL: expand12: +; KNL: # BB#0: +; KNL-NEXT: # kill: %YMM0<def> %YMM0<kill> %ZMM0<def> +; KNL-NEXT: vmovaps {{.*#+}} zmm2 = [0,16,2,16,4,16,6,16,0,16,1,16,2,16,3,16] +; KNL-NEXT: vpxord %zmm1, %zmm1, %zmm1 +; KNL-NEXT: vpermt2ps %zmm0, %zmm2, %zmm1 +; KNL-NEXT: vmovaps %zmm1, %zmm0 +; KNL-NEXT: retq + %res = shufflevector <8 x float> zeroinitializer, <8 x float> %a, <16 x i32> <i32 0, i32 8, i32 1, i32 8, i32 2, i32 8, i32 3, i32 8,i32 0, i32 8, i32 1, i32 8, i32 2, i32 8, i32 3, i32 8> + ret <16 x float> %res +} + +define <16 x float> @expand13(<8 x float> %a ) { +; SKX-LABEL: expand13: +; SKX: # BB#0: +; SKX-NEXT: vxorps %ymm1, %ymm1, %ymm1 +; SKX-NEXT: vinsertf32x8 $1, %ymm0, %zmm1, %zmm0 +; SKX-NEXT: retq +; +; KNL-LABEL: expand13: +; KNL: # BB#0: +; KNL-NEXT: vxorpd %ymm1, %ymm1, %ymm1 +; KNL-NEXT: vinsertf64x4 $1, %ymm0, %zmm1, %zmm0 +; KNL-NEXT: retq + %res = shufflevector <8 x float> zeroinitializer, <8 x float> %a, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7,i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15> + ret <16 x float> %res +} + +; The function checks for a case where the vector is mixed values vector ,and the mask points on zero elements from this vector. + +define <8 x float> @expand14(<4 x float> %a) { +; SKX-LABEL: expand14: +; SKX: # BB#0: +; SKX-NEXT: # kill: %XMM0<def> %XMM0<kill> %YMM0<def> +; SKX-NEXT: movb $20, %al +; SKX-NEXT: kmovb %eax, %k1 +; SKX-NEXT: vexpandps %ymm0, %ymm0 {%k1} {z} +; SKX-NEXT: retq +; +; KNL-LABEL: expand14: +; KNL: # BB#0: +; KNL-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[0,1,1,3] +; KNL-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,0,1,3] +; KNL-NEXT: vmovaps {{.*#+}} ymm1 = <0,2,4,0,u,u,u,u> +; KNL-NEXT: vpermilps {{.*#+}} xmm1 = xmm1[3,3,0,0] +; KNL-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,1,1] +; KNL-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3],ymm0[4],ymm1[5,6,7] +; KNL-NEXT: retq + %addV = fadd <4 x float> <float 0.0,float 1.0,float 2.0,float 0.0> , <float 0.0,float 1.0,float 2.0,float 0.0> + %res = shufflevector <4 x float> %addV, <4 x float> %a, <8 x i32> <i32 3, i32 3, i32 4, i32 0, i32 5, i32 0, i32 0, i32 0> + ret <8 x float> %res +} + +;Negative test. +define <8 x float> @expand15(<4 x float> %a) { +; SKX-LABEL: expand15: +; SKX: # BB#0: +; SKX-NEXT: vpermilps {{.*#+}} xmm1 = xmm0[0,1,1,3] +; SKX-NEXT: vmovaps {{.*#+}} ymm0 = <0,2,4,0,u,u,u,u> +; SKX-NEXT: vpermilps {{.*#+}} xmm2 = xmm0[0,1,0,0] +; SKX-NEXT: vmovaps {{.*#+}} ymm0 = [0,1,8,3,10,3,2,3] +; SKX-NEXT: vpermi2ps %ymm1, %ymm2, %ymm0 +; SKX-NEXT: retq +; +; KNL-LABEL: expand15: +; KNL: # BB#0: +; KNL-NEXT: vpermilps {{.*#+}} xmm0 = xmm0[0,1,1,3] +; KNL-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[0,0,1,3] +; KNL-NEXT: vmovaps {{.*#+}} ymm1 = <0,2,4,0,u,u,u,u> +; KNL-NEXT: vpermilps {{.*#+}} xmm1 = xmm1[0,1,0,0] +; KNL-NEXT: vpermpd {{.*#+}} ymm1 = ymm1[0,1,1,1] +; KNL-NEXT: vblendps {{.*#+}} ymm0 = ymm1[0,1],ymm0[2],ymm1[3],ymm0[4],ymm1[5,6,7] +; KNL-NEXT: retq + %addV = fadd <4 x float> <float 0.0,float 1.0,float 2.0,float 0.0> , <float 0.0,float 1.0,float 2.0,float 0.0> + %res = shufflevector <4 x float> %addV, <4 x float> %a, <8 x i32> <i32 0, i32 1, i32 4, i32 0, i32 5, i32 0, i32 0, i32 0> + ret <8 x float> %res +} |