Home
last modified time | relevance | path

Searched refs:pmullw (Results 1 – 25 of 44) sorted by relevance

12

/external/libvpx/libvpx/vp8/common/x86/
Dsubpixel_sse2.asm77 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 4
[all …]
Dsubpixel_mmx.asm59 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 …]
Ddequantize_mmx.asm30 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]
Dmfqe_sse2.asm57 pmullw xmm2, xmm0
58 pmullw xmm3, xmm0
64 pmullw xmm4, xmm1
65 pmullw xmm5, xmm1
132 pmullw xmm2, xmm0
136 pmullw xmm3, xmm1
Dvariance_impl_mmx.asm553 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 …]
Didctllm_sse2.asm39 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]
Dpostproc_mmx.asm100 pmullw mm1, mm1 ;
125 pmullw mm2, mm2
134 pmullw mm1, mm1
151 pmullw mm1, mm1
Dvariance_impl_sse2.asm460 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] ;
Dpostproc_sse2.asm319 pmullw xmm1, xmm1 ;
344 pmullw xmm2, xmm2
353 pmullw xmm1, xmm1
370 pmullw xmm1, xmm1
/external/libvpx/libvpx/vpx_scale/win32/
Dscaleopt.c114 pmullw mm1, mm5 in horizontal_line_5_4_scale_mmx()
116 pmullw mm0, mm6 in horizontal_line_5_4_scale_mmx()
166 pmullw mm1, three_fourths in vertical_band_5_4_scale_mmx()
168 pmullw mm2, one_fourths in vertical_band_5_4_scale_mmx()
171 pmullw mm3, two_fourths in vertical_band_5_4_scale_mmx()
175 pmullw mm4, two_fourths in vertical_band_5_4_scale_mmx()
180 pmullw mm5, one_fourths in vertical_band_5_4_scale_mmx()
189 pmullw mm6, three_fourths in vertical_band_5_4_scale_mmx()
265 pmullw mm0, mm6 in horizontal_line_5_3_scale_mmx()
267 pmullw mm1, mm5 in horizontal_line_5_3_scale_mmx()
[all …]
/external/llvm/test/CodeGen/X86/
Dvec_shift6.ll8 ; Check that we produce a SSE2 packed integer multiply (pmullw) instead.
15 ; CHECK: pmullw
24 ; CHECK: pmullw
55 ; into two pmullw instructions. With AVX2, the test case below would produce
63 ; SSE: pmullw
64 ; SSE-NEXT: pmullw
87 ; parts and then we convert each part into a pmullw.
94 ; SSE: pmullw
95 ; SSE-NEXT: pmullw
96 ; SSE-NEXT: pmullw
[all …]
D2008-02-26-AsmDirectMemOp.ll15 …mm0 \0A\09movq 8($0, $3), %mm1 \0A\09pmullw %mm6, %mm0 \0A\09pmullw …
Dwiden_arith-4.ll3 ; CHECK-NEXT: pmullw
/external/libvpx/libvpx/vp9/encoder/x86/
Dvp9_subpel_variance.asm420 ; 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
863 pmullw m2, filter_x_a
[all …]
Dvp9_quantize_ssse3_x86_64.asm94 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
279 pmullw m8, m3 ; dqc[i] = qc[i] * q
281 pmullw m13, m3 ; dqc[i] = qc[i] * q
331 pmullw m14, m3 ; dqc[i] = qc[i] * q
332 pmullw m13, m3 ; dqc[i] = qc[i] * q
/external/mesa3d/src/mesa/x86/
Dread_rgba_span_x86.S561 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/libvpx/libvpx/vp9/common/x86/
Dvp9_subpixel_bilinear_sse2.asm39 pmullw xmm0, xmm4 ;multiply the filter factors
87 pmullw xmm0, xmm6
88 pmullw xmm1, xmm7
110 pmullw xmm0, xmm6
111 pmullw xmm1, xmm7
112 pmullw xmm2, xmm6
113 pmullw xmm3, xmm7
Dvp9_subpixel_8t_sse2.asm61 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, k5
[all …]
/external/libvpx/libvpx/vp8/encoder/x86/
Dquantize_mmx.asm66 pmullw mm3, mm2
106 pmullw mm7, mm6
147 pmullw mm7, mm6
188 pmullw mm7, mm6
/external/valgrind/VEX/test/
Dmmxtest.c286 #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
555 do_test("pmullw", pmullw(ma,mb)); in main()
/external/libyuv/files/source/
Drow_win.cc1574 __asm pmullw xmm3, kYToRgb \
1602 __asm pmullw xmm3, kYToRgb \
3037 pmullw xmm2, xmm3 // _r_b * alpha in ARGBBlendRow_SSE2()
3042 pmullw xmm1, xmm3 // _a_g * alpha in ARGBBlendRow_SSE2()
3068 pmullw xmm2, xmm3 // _r_b * alpha in ARGBBlendRow_SSE2()
3073 pmullw xmm1, xmm3 // _a_g * alpha in ARGBBlendRow_SSE2()
3099 pmullw xmm2, xmm3 // _r_b * alpha in ARGBBlendRow_SSE2()
3104 pmullw xmm1, xmm3 // _a_g * alpha in ARGBBlendRow_SSE2()
3169 pmullw xmm2, xmm3 // _r_b * alpha in ARGBBlendRow_SSSE3()
3174 pmullw xmm1, xmm3 // _a_g * alpha in ARGBBlendRow_SSSE3()
[all …]
/external/llvm/test/MC/X86/
Dx86-32-coverage.s4085 pmullw 0xdeadbeef(%ebx,%ecx,8),%mm3
4089 pmullw 0x45,%mm3
4093 pmullw 0x7eed,%mm3
4097 pmullw 0xbabecafe,%mm3
4101 pmullw 0x12345678,%mm3
4105 pmullw %mm3,%mm3
4109 pmullw 0xdeadbeef(%ebx,%ecx,8),%xmm5
4113 pmullw 0x45,%xmm5
4117 pmullw 0x7eed,%xmm5
4121 pmullw 0xbabecafe,%xmm5
[all …]
/external/valgrind/none/tests/x86/
Dinsn_mmx.def48 pmullw mm.sw[1111,2222,-1111,-2222] mm.sw[3333,-4444,3333,-4444] => 1.uw[0x80b3,0x5378,0x7f4d,0xac8…
49 pmullw m64.sw[1111,2222,-1111,-2222] mm.sw[3333,-4444,3333,-4444] => 1.uw[0x80b3,0x5378,0x7f4d,0xac…
/external/valgrind/memcheck/tests/x86/
Dsse2_memory.stdout.exp731 r pmullw 2f78bc0bc33ea15eabc329b69557b556 461496d69c647f5dfc7a5bc01095877e da882808d46adf6e541a…
732 r pmullw 750d5eee9b305e7bc74ed6ccc0273578 69ab1e114c0462e07d895184df8a7370 fdab042d04dafc792bf6…
733 r pmullw fcd338b1bd8feb996c82123248ba8748 b6dda44447b4b374209dcb20bdd6008e 2841e0676bc1511b8089…
734 r pmullw 9faaba3b4ce641878612dd20636b35e3 292db7f6d2a18a87f209e96bff73d598 7746f66d58b60a30ac38…
735 r pmullw 9b57b554cea0b8f4619098aca4742749 1efd498d90d360695fff2d1a593cfc2c 2a739dd8e03d003dff32…
736 m pmullw a9d5ab2046b329a9712d5c70ae7c39af 8c7a59a12606096fab7003cc02f0a659 6c6273e6644071b97bca…
737 m pmullw 9e3739b6b92754a1c7bbfc648439512a 6f9838a89a454d656f033e56cfd95ad0 82ed78444ac244aa49c0…
738 m pmullw 54eb1cbd8fe4c5a2d70578a78186e2cb a11140d2ce65fb261b7ccb02c20b4b4c d493003f125627d5adc1…
739 m pmullw c876e6ba0c15951bae60f112fa5c934e de2ac463c653ecdf7aed8965ed3ee24e 70d1180a482b5c38ec28…
740 m pmullw 56398515d6c43b5ab8fce759a3956f54 42037382e7d759a0ba4abaf7c60fe832 2ccabf341a57833eb0cd…
/external/valgrind/none/tests/amd64/
Dinsn_mmx.def68 pmullw mm.sw[1111,2222,-1111,-2222] mm.sw[3333,-4444,3333,-4444] => 1.uw[0x80b3,0x5378,0x7f4d,0xac8…
69 pmullw m64.sw[1111,2222,-1111,-2222] mm.sw[3333,-4444,3333,-4444] => 1.uw[0x80b3,0x5378,0x7f4d,0xac…

12