/external/libvpx/libvpx/vp8/common/x86/ |
subpixel_sse2.asm | 77 pmullw xmm3, XMMWORD PTR [rdx] ; x[-2] * H[-2]; Tap 1 81 pmullw xmm4, XMMWORD PTR [rdx+16] ; x[-1] * H[-1]; Tap 2 87 pmullw xmm5, [rdx+32] ; x[ 0] * H[ 0]; Tap 3 92 pmullw xmm6, [rdx+48] ; x[ 1] * h[ 1] ; Tap 4 98 pmullw xmm7, [rdx+64] ; x[ 2] * h[ 2] ; Tap 5 101 pmullw xmm1, [rdx+80] ; x[ 3] * h[ 3] ; Tap 6 202 pmullw xmm3, XMMWORD PTR [rdx] ; x[-2] * H[-2]; Tap 1 206 pmullw xmm4, XMMWORD PTR [rdx+16] ; x[-1] * H[-1]; Tap 2 212 pmullw xmm5, [rdx+32] ; x[ 0] * H[ 0]; Tap 3 217 pmullw xmm6, [rdx+48] ; x[ 1] * h[ 1] ; Tap [all...] |
subpixel_mmx.asm | 59 pmullw mm3, mm1 ; mm3 *= kernel 1 modifiers. 63 pmullw mm4, mm7 ; mm5 *= kernel 4 modifiers 69 pmullw mm5, mm2 ; mm5 *= kernel 2 modifiers 75 pmullw mm4, mm6 ; mm5 *= kernel 3 modifiers 81 pmullw mm4, [rdx+80] ; mm5 *= kernel 0 modifiers 85 pmullw mm5, [rdx] ; mm5 *= kernel 5 modifiers 158 pmullw mm3, mm1 ; mm3 *= kernel 1 modifiers. 162 pmullw mm4, mm7 ; mm4 *= kernel 4 modifiers. 166 pmullw mm4, mm2 ; mm4 *= kernel 2 modifiers. 170 pmullw mm4, [rbx] ; mm4 *= kernel 0 modifiers [all...] |
dequantize_mmx.asm | 30 pmullw mm1, [rax+0] ; mm4 *= kernel 0 modifiers. 34 pmullw mm1, [rax+8] ; mm4 *= kernel 0 modifiers. 38 pmullw mm1, [rax+16] ; mm4 *= kernel 0 modifiers. 42 pmullw mm1, [rax+24] ; mm4 *= kernel 0 modifiers. 72 pmullw mm0, [rdx] 75 pmullw mm1, [rdx +8] 78 pmullw mm2, [rdx+16] 81 pmullw mm3, [rdx+24]
|
mfqe_sse2.asm | 57 pmullw xmm2, xmm0 58 pmullw xmm3, xmm0 64 pmullw xmm4, xmm1 65 pmullw xmm5, xmm1 132 pmullw xmm2, xmm0 136 pmullw xmm3, xmm1
|
idctllm_sse2.asm | 39 pmullw xmm4, xmm5 140 pmullw xmm0, [rdx] 141 pmullw xmm2, [rdx+16] 142 pmullw xmm1, [rdx] 143 pmullw xmm3, [rdx+16] 472 pmullw xmm0, [rdx] 473 pmullw xmm2, [rdx+16] 474 pmullw xmm1, [rdx] 475 pmullw xmm3, [rdx+16]
|
variance_impl_mmx.asm | 553 pmullw mm1, [rax] ; 556 pmullw mm3, [rax+8] ; 577 pmullw mm1, [rax] ; 580 pmullw mm3, [rax+8] ; 589 pmullw mm3, [rdx] ; 591 pmullw mm1, [rdx+8] ; 701 pmullw mm1, [rax] ; 703 pmullw mm2, [rax] ; 707 pmullw mm3, [rax+8] ; 709 pmullw mm4, [rax+8] [all...] |
postproc_mmx.asm | 100 pmullw mm1, mm1 ; 125 pmullw mm2, mm2 134 pmullw mm1, mm1 151 pmullw mm1, mm1
|
variance_impl_sse2.asm | 460 pmullw xmm1, [rax] ; 462 pmullw xmm3, [rax+16] ; 480 pmullw xmm1, [rax] ; 482 pmullw xmm3, [rax+16] ; 491 pmullw xmm3, [rdx] ; 492 pmullw xmm1, [rdx+16] ; 544 pmullw xmm1, [rdx] ; 545 pmullw xmm3, [rdx+16] ; 611 pmullw xmm1, [rax] ; 613 pmullw xmm3, [rax+16] [all...] |
postproc_sse2.asm | 319 pmullw xmm1, xmm1 ; 344 pmullw xmm2, xmm2 353 pmullw xmm1, xmm1 370 pmullw xmm1, xmm1
|
/external/libvpx/libvpx/vpx_scale/win32/ |
scaleopt.c | 114 pmullw mm1, mm5 116 pmullw mm0, mm6 166 pmullw mm1, three_fourths 168 pmullw mm2, one_fourths 171 pmullw mm3, two_fourths 175 pmullw mm4, two_fourths 180 pmullw mm5, one_fourths 189 pmullw mm6, three_fourths 265 pmullw mm0, mm6 267 pmullw mm1, mm [all...] |
/external/libvpx/libvpx/vp9/encoder/x86/ |
vp9_subpel_variance.asm | 420 ; slightly faster because of pmullw latency. It would also cut our rodata 422 pmullw m2, filter_y_a 423 pmullw m3, filter_y_b 425 pmullw m0, filter_y_a 426 pmullw m4, filter_y_b 464 pmullw m0, filter_y_a 465 pmullw m1, m2, filter_y_b 468 pmullw m2, filter_y_a 469 pmullw m4, filter_y_b 719 pmullw m2, filter_y_ [all...] |
vp9_temporal_filter_apply_sse2.asm | 104 pmullw xmm0, xmm0 ; modifer[ 0- 7]^2 105 pmullw xmm1, xmm1 ; modifer[ 8-15]^2 108 pmullw xmm0, [GLOBAL(_const_3w)] 109 pmullw xmm1, [GLOBAL(_const_3w)] 127 pmullw xmm2, [rsp + filter_weight] 128 pmullw xmm3, [rsp + filter_weight] 150 pmullw xmm0, xmm2 151 pmullw xmm1, xmm3
|
vp9_quantize_ssse3.asm | 94 pmullw m8, m3 ; dqc[i] = qc[i] * q 96 pmullw m13, m3 ; dqc[i] = qc[i] * q 148 pmullw m14, m3 ; dqc[i] = qc[i] * q 149 pmullw m13, m3 ; dqc[i] = qc[i] * q
|
/external/libvpx/libvpx/vp8/encoder/x86/ |
temporal_filter_apply_sse2.asm | 104 pmullw xmm0, xmm0 ; modifer[ 0- 7]^2 105 pmullw xmm1, xmm1 ; modifer[ 8-15]^2 108 pmullw xmm0, [GLOBAL(_const_3w)] 109 pmullw xmm1, [GLOBAL(_const_3w)] 127 pmullw xmm2, [rsp + filter_weight] 128 pmullw xmm3, [rsp + filter_weight] 150 pmullw xmm0, xmm2 151 pmullw xmm1, xmm3
|
quantize_mmx.asm | 66 pmullw mm3, mm2 106 pmullw mm7, mm6 147 pmullw mm7, mm6 188 pmullw mm7, mm6
|
quantize_ssse3.asm | 97 pmullw xmm0, xmm1 98 pmullw xmm4, xmm5
|
quantize_sse4.asm | 200 pmullw xmm0, xmm4 201 pmullw xmm1, xmm5
|
/external/qemu/distrib/sdl-1.2.15/src/audio/ |
SDL_mixer_MMX_VC.c | 77 pmullw mm1, mm0 //%%mm0,%%mm1\n" // mm1 = l(a*v)|l(b*v)|l(c*v)|l(d*v) 80 pmullw mm4, mm0 //%%mm0,%%mm4\n" // mm4 = l(e*v)|l(f*v)|l(g*v)|l(h*v) 161 pmullw mm1, mm0 //%%mm0,%%mm1 // mm1 = v*a|v*b|v*c|v*d 163 pmullw mm3, mm0 //%%mm0,%%mm3 // mm3 = v*e|v*f|v*g|v*h
|
/external/libvpx/libvpx/vp9/common/x86/ |
vp9_postproc_mmx.asm | 71 pmullw mm3, mm6 ; mm3 *= kernel 2 modifiers 76 pmullw mm6, mm5 ; mm6 *= p0..p3 * kernel 3 modifiers 89 pmullw mm6, mm5 ; mm5 *= kernel 4 modifiers 105 pmullw mm6, mm5 ; mm5 *= kernel 0 modifiers 119 pmullw mm6, mm4 ; mm4 *= kernel 1 modifiers. 166 pmullw mm3, mm6 ; mm3 *= kernel 2 modifiers 172 pmullw mm6, mm5 ; mm6 *= p1..p4 * kernel 3 modifiers 186 pmullw mm6, mm5 ; mm5 *= kernel 4 modifiers 202 pmullw mm6, mm5 ; mm5 *= kernel 0 modifiers 216 pmullw mm6, mm4 ; mm4 *= kernel 1 modifiers [all...] |
vp9_subpixel_8t_sse2.asm | 61 pmullw xmm0, k0k1 ;multiply the filter factors 62 pmullw xmm6, k6k7 63 pmullw xmm2, k2k3 64 pmullw xmm5, k5k4 152 pmullw xmm0, k0 153 pmullw xmm1, k1 154 pmullw xmm6, k6 155 pmullw xmm7, k7 156 pmullw xmm2, k2 157 pmullw xmm5, k [all...] |
/external/chromium_org/third_party/mesa/src/src/mesa/x86/ |
read_rgba_span_x86.S | 561 pmullw %mm6, %mm0 562 pmullw %mm6, %mm2 596 pmullw %mm6, %mm0 597 pmullw %mm6, %mm2 634 pmullw %mm6, %mm0 635 pmullw %mm6, %mm2 665 pmullw %mm6, %mm0
|
/external/mesa3d/src/mesa/x86/ |
read_rgba_span_x86.S | 561 pmullw %mm6, %mm0 562 pmullw %mm6, %mm2 596 pmullw %mm6, %mm0 597 pmullw %mm6, %mm2 634 pmullw %mm6, %mm0 635 pmullw %mm6, %mm2 665 pmullw %mm6, %mm0
|
/external/valgrind/main/VEX/test/ |
mmxtest.c | 286 #define pmullw_m2r(var, reg) mmx_m2r(pmullw, var, reg) 287 #define pmullw_r2r(regs, regd) mmx_r2r(pmullw, regs, regd) 288 #define pmullw(vars, vard) mmx_m2m(pmullw, vars, vard) macro 299 (muls like pmullw, then adds adjacent 16-bit fields 555 do_test("pmullw", pmullw(ma,mb));
|
/external/qemu/distrib/sdl-1.2.15/src/video/ |
mmx.h | 479 #define pmullw_m2r(var, reg) mmx_m2r(pmullw, var, reg) 480 #define pmullw_r2r(regs, regd) mmx_r2r(pmullw, regs, regd) 481 #define pmullw(vars, vard) mmx_m2m(pmullw, vars, vard) macro 492 (muls like pmullw, then adds adjacent 16-bit fields
|
/external/valgrind/main/none/tests/amd64/ |
insn_mmx.def | 68 pmullw mm.sw[1111,2222,-1111,-2222] mm.sw[3333,-4444,3333,-4444] => 1.uw[0x80b3,0x5378,0x7f4d,0xac88] 69 pmullw m64.sw[1111,2222,-1111,-2222] mm.sw[3333,-4444,3333,-4444] => 1.uw[0x80b3,0x5378,0x7f4d,0xac88]
|