/external/llvm/test/MC/X86/ |
D | x86-32-avx.s | 5 vaddss %xmm4, %xmm6, %xmm2 9 vmulss %xmm4, %xmm6, %xmm2 13 vsubss %xmm4, %xmm6, %xmm2 17 vdivss %xmm4, %xmm6, %xmm2 21 vaddsd %xmm4, %xmm6, %xmm2 25 vmulsd %xmm4, %xmm6, %xmm2 29 vsubsd %xmm4, %xmm6, %xmm2 33 vdivsd %xmm4, %xmm6, %xmm2 37 vaddss 3735928559(%ebx,%ecx,8), %xmm2, %xmm5 41 vsubss 3735928559(%ebx,%ecx,8), %xmm2, %xmm5 [all …]
|
D | x86_64-xop-encoding.s | 29 vphsubbw %xmm2, %xmm1 37 vphaddwq %xmm6, %xmm2 61 vphadduwd %xmm2, %xmm1 69 vphaddudq %xmm6, %xmm2 85 vphaddubq %xmm2, %xmm2 117 vphaddbq %xmm2, %xmm0 180 vpshlw %xmm0, %xmm1, %xmm2 183 vpshlw (%rax), %xmm1, %xmm2 186 vpshlw %xmm0, (%rax,%rcx), %xmm2 191 vpshlq %xmm2, %xmm4, %xmm6 [all …]
|
D | shuffle-comments.s | 18 vpalignr $8, %xmm0, %xmm1, %xmm2 19 # CHECK: xmm2 = xmm0[8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7] 20 vpalignr $8, (%rax), %xmm1, %xmm2 21 # CHECK: xmm2 = mem[8,9,10,11,12,13,14,15],xmm1[0,1,2,3,4,5,6,7] 23 vpalignr $16, %xmm0, %xmm1, %xmm2 24 # CHECK: xmm2 = xmm1[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15] 25 vpalignr $16, (%rax), %xmm1, %xmm2 26 # CHECK: xmm2 = xmm1[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15] 28 vpalignr $0, %xmm0, %xmm1, %xmm2 29 # CHECK: xmm2 = xmm0[0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15] [all …]
|
D | x86-32-fma3.s | 5 vfmadd132pd %xmm2, %xmm5, %xmm1 13 vfmadd132ps %xmm2, %xmm5, %xmm1 21 vfmadd213pd %xmm2, %xmm5, %xmm1 29 vfmadd213ps %xmm2, %xmm5, %xmm1 37 vfmadd231pd %xmm2, %xmm5, %xmm1 45 vfmadd231ps %xmm2, %xmm5, %xmm1 101 vfmadd132pd %xmm2, %xmm5, %xmm1 109 vfmadd132ps %xmm2, %xmm5, %xmm1 117 vfmadd213pd %xmm2, %xmm5, %xmm1 125 vfmadd213ps %xmm2, %xmm5, %xmm1 [all …]
|
/external/llvm/test/CodeGen/X86/ |
D | bswap-vector.ll | 17 ; CHECK-NOSSSE3-NEXT: movdqa %xmm0, %xmm2 18 … {{.*#+}} xmm2 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1… 19 ; CHECK-NOSSSE3-NEXT: pshuflw {{.*#+}} xmm2 = xmm2[1,0,3,2,4,5,6,7] 20 ; CHECK-NOSSSE3-NEXT: pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,5,4,7,6] 24 ; CHECK-NOSSSE3-NEXT: packuswb %xmm2, %xmm0 50 ; CHECK-NOSSSE3-NEXT: movdqa %xmm0, %xmm2 51 … {{.*#+}} xmm2 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1… 52 ; CHECK-NOSSSE3-NEXT: pshuflw {{.*#+}} xmm2 = xmm2[3,2,1,0,4,5,6,7] 53 ; CHECK-NOSSSE3-NEXT: pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,7,6,5,4] 57 ; CHECK-NOSSSE3-NEXT: packuswb %xmm2, %xmm0 [all …]
|
D | vector-shift-lshr-128.ll | 20 ; SSE2-NEXT: movdqa %xmm0, %xmm2 21 ; SSE2-NEXT: psrlq %xmm3, %xmm2 23 ; SSE2-NEXT: movsd {{.*#+}} xmm2 = xmm0[0],xmm2[1] 24 ; SSE2-NEXT: movapd %xmm2, %xmm0 29 ; SSE41-NEXT: movdqa %xmm0, %xmm2 30 ; SSE41-NEXT: psrlq %xmm1, %xmm2 33 ; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm2[0,1,2,3],xmm0[4,5,6,7] 38 ; AVX1-NEXT: vpsrlq %xmm1, %xmm0, %xmm2 41 ; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm2[0,1,2,3],xmm0[4,5,6,7] 51 ; XOPAVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2 [all …]
|
D | vector-shift-ashr-128.ll | 20 ; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [9223372036854775808,9223372036854775808] 21 ; SSE2-NEXT: movdqa %xmm2, %xmm4 23 ; SSE2-NEXT: psrlq %xmm1, %xmm2 24 ; SSE2-NEXT: movsd {{.*#+}} xmm4 = xmm2[0],xmm4[1] 25 ; SSE2-NEXT: movdqa %xmm0, %xmm2 26 ; SSE2-NEXT: psrlq %xmm3, %xmm2 28 ; SSE2-NEXT: movsd {{.*#+}} xmm2 = xmm0[0],xmm2[1] 29 ; SSE2-NEXT: xorpd %xmm4, %xmm2 30 ; SSE2-NEXT: psubq %xmm4, %xmm2 31 ; SSE2-NEXT: movdqa %xmm2, %xmm0 [all …]
|
D | vector-shift-lshr-256.ll | 14 ; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2 16 ; AVX1-NEXT: vpsrlq %xmm2, %xmm3, %xmm4 17 ; AVX1-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[2,3,0,1] 18 ; AVX1-NEXT: vpsrlq %xmm2, %xmm3, %xmm2 19 ; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm4[0,1,2,3],xmm2[4,5,6,7] 24 ; AVX1-NEXT: vinsertf128 $1, %xmm2, %ymm0, %ymm0 34 ; XOPAVX1-NEXT: vextractf128 $1, %ymm1, %xmm2 36 ; XOPAVX1-NEXT: vpsubq %xmm2, %xmm3, %xmm2 38 ; XOPAVX1-NEXT: vpshlq %xmm2, %xmm4, %xmm2 41 ; XOPAVX1-NEXT: vinsertf128 $1, %xmm2, %ymm0, %ymm0 [all …]
|
D | vector-lzcnt-256.ll | 16 ; AVX1-NEXT: vmovq %rax, %xmm2 22 ; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0] 27 ; AVX1-NEXT: vmovq %rax, %xmm2 33 ; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0] 45 ; AVX2-NEXT: vmovq %rax, %xmm2 51 ; AVX2-NEXT: vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0] 56 ; AVX2-NEXT: vmovq %rax, %xmm2 62 ; AVX2-NEXT: vpunpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0] 87 ; AVX1-NEXT: vmovq %rax, %xmm2 92 ; AVX1-NEXT: vpunpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm2[0] [all …]
|
D | vector-tzcnt-128.ll | 158 ; SSE2-NEXT: pxor %xmm2, %xmm2 159 ; SSE2-NEXT: psubd %xmm0, %xmm2 160 ; SSE2-NEXT: pand %xmm0, %xmm2 161 ; SSE2-NEXT: psubd {{.*}}(%rip), %xmm2 162 ; SSE2-NEXT: movdqa %xmm2, %xmm0 165 ; SSE2-NEXT: psubd %xmm0, %xmm2 167 ; SSE2-NEXT: movdqa %xmm2, %xmm3 169 ; SSE2-NEXT: psrld $2, %xmm2 170 ; SSE2-NEXT: pand %xmm0, %xmm2 171 ; SSE2-NEXT: paddd %xmm3, %xmm2 [all …]
|
D | vector-shift-ashr-256.ll | 14 ; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2 16 ; AVX1-NEXT: vpsrlq %xmm2, %xmm3, %xmm4 17 ; AVX1-NEXT: vpshufd {{.*#+}} xmm5 = xmm2[2,3,0,1] 21 ; AVX1-NEXT: vpsrlq %xmm2, %xmm6, %xmm2 23 ; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm2[0,1,2,3],xmm5[4,5,6,7] 24 ; AVX1-NEXT: vpxor %xmm4, %xmm2, %xmm2 25 ; AVX1-NEXT: vpsubq %xmm4, %xmm2, %xmm2 35 ; AVX1-NEXT: vinsertf128 $1, %xmm2, %ymm0, %ymm0 49 ; XOPAVX1-NEXT: vextractf128 $1, %ymm1, %xmm2 51 ; XOPAVX1-NEXT: vpsubq %xmm2, %xmm3, %xmm2 [all …]
|
D | fmaxnum.ll | 19 ; SSE: movaps %xmm0, %xmm2 20 ; SSE-NEXT: cmpunordss %xmm2, %xmm2 21 ; SSE-NEXT: movaps %xmm2, %xmm3 24 ; SSE-NEXT: andnps %xmm1, %xmm2 25 ; SSE-NEXT: orps %xmm3, %xmm2 26 ; SSE-NEXT: movaps %xmm2, %xmm0 29 ; AVX: vmaxss %xmm0, %xmm1, %xmm2 31 ; AVX-NEXT: vblendvps %xmm0, %xmm1, %xmm2, %xmm0 62 ; SSE: movaps %xmm0, %xmm2 63 ; SSE-NEXT: cmpunordss %xmm2, %xmm2 [all …]
|
D | vector-zext.ll | 41 ; SSE2-NEXT: pxor %xmm2, %xmm2 42 …}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1],xmm0[2],xmm2[2],xmm0[3],xmm2[3],xmm0[4],xmm2[4],xmm0[5],… 43 …m1 = xmm1[8],xmm2[8],xmm1[9],xmm2[9],xmm1[10],xmm2[10],xmm1[11],xmm2[11],xmm1[12],xmm2[12],xmm1[13… 49 ; SSSE3-NEXT: pxor %xmm2, %xmm2 50 …}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1],xmm0[2],xmm2[2],xmm0[3],xmm2[3],xmm0[4],xmm2[4],xmm0[5],… 51 …m1 = xmm1[8],xmm2[8],xmm1[9],xmm2[9],xmm1[10],xmm2[10],xmm1[11],xmm2[11],xmm1[12],xmm2[12],xmm1[13… 57 ; SSE41-NEXT: pxor %xmm2, %xmm2 59 …m1 = xmm1[8],xmm2[8],xmm1[9],xmm2[9],xmm1[10],xmm2[10],xmm1[11],xmm2[11],xmm1[12],xmm2[12],xmm1[13… 113 ; SSE2-NEXT: pxor %xmm2, %xmm2 114 …}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1],xmm1[2],xmm2[2],xmm1[3],xmm2[3],xmm1[4],xmm2[4],xmm1[5],… [all …]
|
D | vector-shift-shl-128.ll | 20 ; SSE2-NEXT: movdqa %xmm0, %xmm2 21 ; SSE2-NEXT: psllq %xmm3, %xmm2 23 ; SSE2-NEXT: movsd {{.*#+}} xmm2 = xmm0[0],xmm2[1] 24 ; SSE2-NEXT: movapd %xmm2, %xmm0 29 ; SSE41-NEXT: movdqa %xmm0, %xmm2 30 ; SSE41-NEXT: psllq %xmm1, %xmm2 33 ; SSE41-NEXT: pblendw {{.*#+}} xmm0 = xmm2[0,1,2,3],xmm0[4,5,6,7] 38 ; AVX1-NEXT: vpsllq %xmm1, %xmm0, %xmm2 41 ; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm2[0,1,2,3],xmm0[4,5,6,7] 62 ; X32-SSE-NEXT: movdqa %xmm0, %xmm2 [all …]
|
D | vector-trunc.ll | 16 ; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm2[0,2,2,3] 26 ; SSSE3-NEXT: pshufd {{.*#+}} xmm1 = xmm2[0,2,2,3] 36 ; SSE41-NEXT: pshufd {{.*#+}} xmm1 = xmm2[0,2,2,3] 42 ; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2 43 ; AVX1-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[0,1,0,2] 45 ; AVX1-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm2[4,5,6,7] 46 ; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm2 47 ; AVX1-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[0,1,0,2] 49 ; AVX1-NEXT: vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm2[4,5,6,7] 76 ; SSE2-NEXT: punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1],xmm0[2],xmm2[2],xmm0[3],x… [all …]
|
D | vector-shift-shl-256.ll | 14 ; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2 16 ; AVX1-NEXT: vpsllq %xmm2, %xmm3, %xmm4 17 ; AVX1-NEXT: vpshufd {{.*#+}} xmm2 = xmm2[2,3,0,1] 18 ; AVX1-NEXT: vpsllq %xmm2, %xmm3, %xmm2 19 ; AVX1-NEXT: vpblendw {{.*#+}} xmm2 = xmm4[0,1,2,3],xmm2[4,5,6,7] 24 ; AVX1-NEXT: vinsertf128 $1, %xmm2, %ymm0, %ymm0 34 ; XOPAVX1-NEXT: vextractf128 $1, %ymm1, %xmm2 36 ; XOPAVX1-NEXT: vpshlq %xmm2, %xmm3, %xmm2 38 ; XOPAVX1-NEXT: vinsertf128 $1, %xmm2, %ymm0, %ymm0 52 ; AVX1-NEXT: vextractf128 $1, %ymm1, %xmm2 [all …]
|
D | vector-lzcnt-128.ll | 207 ; SSE2-NEXT: pshufd {{.*#+}} xmm2 = xmm0[1,1,2,3] 208 ; SSE2-NEXT: movd %xmm2, %eax 212 ; SSE2-NEXT: movd %eax, %xmm2 213 ; SSE2-NEXT: punpckldq {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1] 226 ; SSE2-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1] 239 ; SSE3-NEXT: pshufd {{.*#+}} xmm2 = xmm0[1,1,2,3] 240 ; SSE3-NEXT: movd %xmm2, %eax 244 ; SSE3-NEXT: movd %eax, %xmm2 245 ; SSE3-NEXT: punpckldq {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1] 258 ; SSE3-NEXT: punpckldq {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1] [all …]
|
D | vector-tzcnt-256.ll | 9 ; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2 10 ; AVX1-NEXT: vpsubq %xmm1, %xmm2, %xmm1 11 ; AVX1-NEXT: vpsubq %xmm0, %xmm2, %xmm3 25 ; AVX1-NEXT: vpsadbw %xmm2, %xmm1, %xmm1 33 ; AVX1-NEXT: vpsadbw %xmm2, %xmm0, %xmm0 62 ; AVX1-NEXT: vpxor %xmm2, %xmm2, %xmm2 63 ; AVX1-NEXT: vpsubq %xmm1, %xmm2, %xmm1 64 ; AVX1-NEXT: vpsubq %xmm0, %xmm2, %xmm3 78 ; AVX1-NEXT: vpsadbw %xmm2, %xmm1, %xmm1 86 ; AVX1-NEXT: vpsadbw %xmm2, %xmm0, %xmm0 [all …]
|
D | vec_cmp_uint-128.ll | 231 ; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [2147483648,2147483648,2147483648,2147483648] 232 ; SSE2-NEXT: pxor %xmm2, %xmm0 233 ; SSE2-NEXT: pxor %xmm2, %xmm1 234 ; SSE2-NEXT: movdqa %xmm1, %xmm2 235 ; SSE2-NEXT: pcmpgtd %xmm0, %xmm2 236 ; SSE2-NEXT: pshufd {{.*#+}} xmm3 = xmm2[0,0,2,2] 240 ; SSE2-NEXT: pshufd {{.*#+}} xmm1 = xmm2[1,1,3,3] 248 ; SSE41-NEXT: movdqa {{.*#+}} xmm2 = [2147483648,2147483648,2147483648,2147483648] 249 ; SSE41-NEXT: pxor %xmm2, %xmm0 250 ; SSE41-NEXT: pxor %xmm2, %xmm1 [all …]
|
D | lower-vec-shift-2.ll | 16 ; AVX-NEXT: vpxor %xmm2, %xmm2, %xmm2 17 ; AVX-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3,4,5,6,7] 29 ; SSE2-NEXT: xorps %xmm2, %xmm2 30 ; SSE2-NEXT: movss {{.*#+}} xmm2 = xmm1[0],xmm2[1,2,3] 31 ; SSE2-NEXT: pslld %xmm2, %xmm0 36 ; AVX-NEXT: vpxor %xmm2, %xmm2, %xmm2 37 ; AVX-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0,1],xmm2[2,3,4,5,6,7] 73 ; AVX-NEXT: vpxor %xmm2, %xmm2, %xmm2 74 ; AVX-NEXT: vpblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3,4,5,6,7] 86 ; SSE2-NEXT: xorps %xmm2, %xmm2 [all …]
|
/external/libvpx/libvpx/vp8/common/x86/ |
D | idctllm_sse2.asm | 61 movq xmm2, [rax+2*rdx] 66 punpcklbw xmm2, xmm5 73 paddw xmm2, xmm4 79 packuswb xmm2, xmm5 88 movq [rax], xmm2 126 ; note the transpose of xmm1 and xmm2, necessary for shuffle 129 movdqa xmm2, [rax+16] 141 pmullw xmm2, [rdx+16] 154 movdqa xmm4, xmm2 155 punpckldq xmm2, xmm3 [all …]
|
/external/boringssl/linux-x86/crypto/aes/ |
D | aesni-x86.S | 12 movups (%eax),%xmm2 18 xorps %xmm0,%xmm2 28 movups %xmm2,(%eax) 29 pxor %xmm2,%xmm2 40 movups (%eax),%xmm2 46 xorps %xmm0,%xmm2 56 movups %xmm2,(%eax) 57 pxor %xmm2,%xmm2 67 xorps %xmm0,%xmm2 95 xorps %xmm0,%xmm2 [all …]
|
/external/boringssl/mac-x86/crypto/aes/ |
D | aesni-x86.S | 11 movups (%eax),%xmm2 17 xorps %xmm0,%xmm2 27 movups %xmm2,(%eax) 28 pxor %xmm2,%xmm2 37 movups (%eax),%xmm2 43 xorps %xmm0,%xmm2 53 movups %xmm2,(%eax) 54 pxor %xmm2,%xmm2 62 xorps %xmm0,%xmm2 88 xorps %xmm0,%xmm2 [all …]
|
/external/boringssl/src/crypto/aes/asm/ |
D | vpaes-x86_64.pl | 90 movdqa .Lk_ipt(%rip), %xmm2 # iptlo 95 pshufb %xmm0, %xmm2 98 pxor %xmm5, %xmm2 100 pxor %xmm2, %xmm0 109 pshufb %xmm2, %xmm4 # 4 = sb1u 115 pshufb %xmm2, %xmm5 # 4 = sb2u 117 movdqa %xmm14, %xmm2 # 2 : sb2t 118 pshufb %xmm3, %xmm2 # 2 = sb2t 120 pxor %xmm5, %xmm2 # 2 = 2A 123 pxor %xmm2, %xmm0 # 0 = 2A+B [all …]
|
/external/boringssl/win-x86/crypto/aes/ |
D | aesni-x86.asm | 24 movups xmm2,[eax] 30 xorps xmm2,xmm0 40 movups [eax],xmm2 41 pxor xmm2,xmm2 49 movups xmm2,[eax] 55 xorps xmm2,xmm0 65 movups [eax],xmm2 66 pxor xmm2,xmm2 73 xorps xmm2,xmm0 98 xorps xmm2,xmm0 [all …]
|