History

Merge pull request #517 from mstorsjo/simplify-x86-asm-func-macro

Fold ALIGN 16 and the function label into WELS_EXTERN

volvet committed 12 years ago

Commit e75cd222 parents 6892127 eec968221 files · +75 −411Browse files

Showing the changes to codec/processing/src (3 of 21 files). Show all

Unified · Split

3 files
  1. codec/processing/src/asm/denoisefilter.asm +2 −6
  2. codec/processing/src/asm/downsample_bilinear.asm +9 −29
  3. codec/processing/src/asm/vaa.asm +12 −36
codec/processing/src/asm/denoisefilter.asm+2−6
@@ -163,8 +163,6 @@ SECTION .text · 162 unmodified lines
163 paddw %3, %2163 paddw %3, %2
164%endmacro164%endmacro
165 165
166ALIGN 16
167WELS_EXTERN BilateralLumaFilter8_sse2
168;***********************************************************************166;***********************************************************************
169; BilateralLumaFilter8_sse2(uint8_t *pixels, int stride);167; BilateralLumaFilter8_sse2(uint8_t *pixels, int stride);
170;***********************************************************************168;***********************************************************************
@@ -173,7 +171,7 @@ WELS_EXTERN BilateralLumaFilter8_sse2 · 2 unmodified lines
173; 6 7 8171; 6 7 8
174; 0: the center point172; 0: the center point
175 173
176BilateralLumaFilter8_sse2:174WELS_EXTERN BilateralLumaFilter8_sse2
177 175
178 push r3176 push r3
179 %assign push_num 1177 %assign push_num 1
@@ -219,7 +217,6 @@ BilateralLumaFilter8_sse2: · 39 unmodified lines
219 217
220 ret218 ret
221 219
222WELS_EXTERN WaverageChromaFilter8_sse2
223;***********************************************************************220;***********************************************************************
224; void WaverageChromaFilter8_sse2(uint8_t *pixels, int stride);221; void WaverageChromaFilter8_sse2(uint8_t *pixels, int stride);
225;***********************************************************************222;***********************************************************************
@@ -230,8 +227,7 @@ WELS_EXTERN WaverageChromaFilter8_sse2 · 4 unmodified lines
230;1 2 4 2 1227;1 2 4 2 1
231;1 1 2 1 1228;1 1 2 1 1
232 229
233ALIGN 16230WELS_EXTERN WaverageChromaFilter8_sse2
234WaverageChromaFilter8_sse2:
235 231
236 push r3232 push r3
237 233
codec/processing/src/asm/downsample_bilinear.asm+9−29
@@ -66,22 +66,18 @@ shufb_mask_high: · 65 unmodified lines
66 db 01h, 80h, 03h, 80h, 05h, 80h, 07h, 80h, 09h, 80h, 0bh, 80h, 0dh, 80h, 0fh, 80h66 db 01h, 80h, 03h, 80h, 05h, 80h, 07h, 80h, 09h, 80h, 0bh, 80h, 0dh, 80h, 0fh, 80h
67 67
68 68
69ALIGN 16
70
71;***********************************************************************69;***********************************************************************
72; Code70; Code
73;***********************************************************************71;***********************************************************************
74 72
75SECTION .text73SECTION .text
76 74
77WELS_EXTERN DyadicBilinearDownsamplerWidthx32_sse
78;***********************************************************************75;***********************************************************************
79; void DyadicBilinearDownsamplerWidthx32_sse( unsigned char* pDst, const int iDstStride,76; void DyadicBilinearDownsamplerWidthx32_sse( unsigned char* pDst, const int iDstStride,
80; unsigned char* pSrc, const int iSrcStride,77; unsigned char* pSrc, const int iSrcStride,
81; const int iSrcWidth, const int iSrcHeight );78; const int iSrcWidth, const int iSrcHeight );
82;***********************************************************************79;***********************************************************************
83ALIGN 1680WELS_EXTERN DyadicBilinearDownsamplerWidthx32_sse
84DyadicBilinearDownsamplerWidthx32_sse:
85 push ebx81 push ebx
86 push edx82 push edx
87 push esi83 push esi
@@ -227,14 +223,12 @@ DyadicBilinearDownsamplerWidthx32_sse: · 139 unmodified lines
227 pop ebx223 pop ebx
228 ret224 ret
229 225
230WELS_EXTERN DyadicBilinearDownsamplerWidthx16_sse
231;***********************************************************************226;***********************************************************************
232; void DyadicBilinearDownsamplerWidthx16_sse( unsigned char* pDst, const int iDstStride,227; void DyadicBilinearDownsamplerWidthx16_sse( unsigned char* pDst, const int iDstStride,
233; unsigned char* pSrc, const int iSrcStride,228; unsigned char* pSrc, const int iSrcStride,
234; const int iSrcWidth, const int iSrcHeight );229; const int iSrcWidth, const int iSrcHeight );
235;***********************************************************************230;***********************************************************************
236ALIGN 16231WELS_EXTERN DyadicBilinearDownsamplerWidthx16_sse
237DyadicBilinearDownsamplerWidthx16_sse:
238 push ebx232 push ebx
239 push edx233 push edx
240 push esi234 push esi
@@ -331,14 +325,12 @@ DyadicBilinearDownsamplerWidthx16_sse: · 90 unmodified lines
331 pop ebx325 pop ebx
332 ret326 ret
333 327
334WELS_EXTERN DyadicBilinearDownsamplerWidthx8_sse
335;***********************************************************************328;***********************************************************************
336; void DyadicBilinearDownsamplerWidthx8_sse( unsigned char* pDst, const int iDstStride,329; void DyadicBilinearDownsamplerWidthx8_sse( unsigned char* pDst, const int iDstStride,
337; unsigned char* pSrc, const int iSrcStride,330; unsigned char* pSrc, const int iSrcStride,
338; const int iSrcWidth, const int iSrcHeight );331; const int iSrcWidth, const int iSrcHeight );
339;***********************************************************************332;***********************************************************************
340ALIGN 16333WELS_EXTERN DyadicBilinearDownsamplerWidthx8_sse
341DyadicBilinearDownsamplerWidthx8_sse:
342 push ebx334 push ebx
343 push edx335 push edx
344 push esi336 push esi
@@ -422,14 +414,12 @@ DyadicBilinearDownsamplerWidthx8_sse: · 77 unmodified lines
422 414
423 415
424; got about 50% improvement over DyadicBilinearDownsamplerWidthx32_sse416; got about 50% improvement over DyadicBilinearDownsamplerWidthx32_sse
425WELS_EXTERN DyadicBilinearDownsamplerWidthx32_ssse3
426;***********************************************************************417;***********************************************************************
427; void DyadicBilinearDownsamplerWidthx32_ssse3( unsigned char* pDst, const int iDstStride,418; void DyadicBilinearDownsamplerWidthx32_ssse3( unsigned char* pDst, const int iDstStride,
428; unsigned char* pSrc, const int iSrcStride,419; unsigned char* pSrc, const int iSrcStride,
429; const int iSrcWidth, const int iSrcHeight );420; const int iSrcWidth, const int iSrcHeight );
430;***********************************************************************421;***********************************************************************
431ALIGN 16422WELS_EXTERN DyadicBilinearDownsamplerWidthx32_ssse3
432DyadicBilinearDownsamplerWidthx32_ssse3:
433 push ebx423 push ebx
434 push edx424 push edx
435 push esi425 push esi
@@ -533,14 +523,12 @@ DyadicBilinearDownsamplerWidthx32_ssse3: · 97 unmodified lines
533 pop ebx523 pop ebx
534 ret524 ret
535 525
536WELS_EXTERN DyadicBilinearDownsamplerWidthx16_ssse3
537;***********************************************************************526;***********************************************************************
538; void DyadicBilinearDownsamplerWidthx16_ssse3( unsigned char* pDst, const int iDstStride,527; void DyadicBilinearDownsamplerWidthx16_ssse3( unsigned char* pDst, const int iDstStride,
539; unsigned char* pSrc, const int iSrcStride,528; unsigned char* pSrc, const int iSrcStride,
540; const int iSrcWidth, const int iSrcHeight );529; const int iSrcWidth, const int iSrcHeight );
541;***********************************************************************530;***********************************************************************
542ALIGN 16531WELS_EXTERN DyadicBilinearDownsamplerWidthx16_ssse3
543DyadicBilinearDownsamplerWidthx16_ssse3:
544 push ebx532 push ebx
545 push edx533 push edx
546 push esi534 push esi
@@ -623,14 +611,12 @@ DyadicBilinearDownsamplerWidthx16_ssse3: · 76 unmodified lines
623 ret611 ret
624 612
625; got about 65% improvement over DyadicBilinearDownsamplerWidthx32_sse613; got about 65% improvement over DyadicBilinearDownsamplerWidthx32_sse
626WELS_EXTERN DyadicBilinearDownsamplerWidthx32_sse4
627;***********************************************************************614;***********************************************************************
628; void DyadicBilinearDownsamplerWidthx32_sse4( unsigned char* pDst, const int iDstStride,615; void DyadicBilinearDownsamplerWidthx32_sse4( unsigned char* pDst, const int iDstStride,
629; unsigned char* pSrc, const int iSrcStride,616; unsigned char* pSrc, const int iSrcStride,
630; const int iSrcWidth, const int iSrcHeight );617; const int iSrcWidth, const int iSrcHeight );
631;***********************************************************************618;***********************************************************************
632ALIGN 16619WELS_EXTERN DyadicBilinearDownsamplerWidthx32_sse4
633DyadicBilinearDownsamplerWidthx32_sse4:
634 push ebx620 push ebx
635 push edx621 push edx
636 push esi622 push esi
@@ -733,14 +719,12 @@ DyadicBilinearDownsamplerWidthx32_sse4: · 96 unmodified lines
733 pop ebx719 pop ebx
734 ret720 ret
735 721
736WELS_EXTERN DyadicBilinearDownsamplerWidthx16_sse4
737;***********************************************************************722;***********************************************************************
738; void DyadicBilinearDownsamplerWidthx16_sse4( unsigned char* pDst, const int iDstStride,723; void DyadicBilinearDownsamplerWidthx16_sse4( unsigned char* pDst, const int iDstStride,
739; unsigned char* pSrc, const int iSrcStride,724; unsigned char* pSrc, const int iSrcStride,
740; const int iSrcWidth, const int iSrcHeight );725; const int iSrcWidth, const int iSrcHeight );
741;***********************************************************************726;***********************************************************************
742ALIGN 16727WELS_EXTERN DyadicBilinearDownsamplerWidthx16_sse4
743DyadicBilinearDownsamplerWidthx16_sse4:
744 push ebx728 push ebx
745 push edx729 push edx
746 push esi730 push esi
@@ -825,7 +809,6 @@ DyadicBilinearDownsamplerWidthx16_sse4: · 78 unmodified lines
825 809
826 810
827 811
828WELS_EXTERN GeneralBilinearAccurateDownsampler_sse2
829;**************************************************************************************************************812;**************************************************************************************************************
830;int GeneralBilinearAccurateDownsampler_sse2( unsigned char* pDst, const int iDstStride, const int iDstWidth, const int iDstHeight,813;int GeneralBilinearAccurateDownsampler_sse2( unsigned char* pDst, const int iDstStride, const int iDstWidth, const int iDstHeight,
831; unsigned char* pSrc, const int iSrcStride, const int iSrcWidth, const int iSrcHeight,814; unsigned char* pSrc, const int iSrcStride, const int iSrcWidth, const int iSrcHeight,
@@ -833,8 +816,7 @@ WELS_EXTERN GeneralBilinearAccurateDownsampler_sse2 · 1 unmodified line
833;{816;{
834;**************************************************************************************************************817;**************************************************************************************************************
835 818
836ALIGN 16819WELS_EXTERN GeneralBilinearAccurateDownsampler_sse2
837GeneralBilinearAccurateDownsampler_sse2:
838 push ebp820 push ebp
839 push esi821 push esi
840 push edi822 push edi
@@ -1029,7 +1011,6 @@ LAST_ROW_END: · 188 unmodified lines
1029 1011
1030 1012
1031 1013
1032WELS_EXTERN GeneralBilinearFastDownsampler_sse2
1033;**************************************************************************************************************1014;**************************************************************************************************************
1034;int GeneralBilinearFastDownsampler_sse2( unsigned char* pDst, const int iDstStride, const int iDstWidth, const int iDstHeight,1015;int GeneralBilinearFastDownsampler_sse2( unsigned char* pDst, const int iDstStride, const int iDstWidth, const int iDstHeight,
1035; unsigned char* pSrc, const int iSrcStride, const int iSrcWidth, const int iSrcHeight,1016; unsigned char* pSrc, const int iSrcStride, const int iSrcWidth, const int iSrcHeight,
@@ -1037,8 +1018,7 @@ WELS_EXTERN GeneralBilinearFastDownsampler_sse2 · 1 unmodified line
1037;{1018;{
1038;**************************************************************************************************************1019;**************************************************************************************************************
1039 1020
1040ALIGN 161021WELS_EXTERN GeneralBilinearFastDownsampler_sse2
1041GeneralBilinearFastDownsampler_sse2:
1042 push ebp1022 push ebp
1043 push esi1023 push esi
1044 push edi1024 push edi
codec/processing/src/asm/vaa.asm+12−36
@@ -245,12 +245,10 @@ SECTION .text · 244 unmodified lines
245 245
246%ifdef X86_32246%ifdef X86_32
247 247
248WELS_EXTERN SampleVariance16x16_sse2
249;***********************************************************************248;***********************************************************************
250; void SampleVariance16x16_sse2( uint8_t * y_ref, int32_t y_ref_stride, uint8_t * y_src, int32_t y_src_stride,SMotionTextureUnit* pMotionTexture );249; void SampleVariance16x16_sse2( uint8_t * y_ref, int32_t y_ref_stride, uint8_t * y_src, int32_t y_src_stride,SMotionTextureUnit* pMotionTexture );
251;***********************************************************************250;***********************************************************************
252ALIGN 16251WELS_EXTERN SampleVariance16x16_sse2
253SampleVariance16x16_sse2:
254 push esi252 push esi
255 push edi253 push edi
256 push ebx254 push ebx
@@ -347,15 +345,13 @@ SampleVariance16x16_sse2: · 90 unmodified lines
347 345
348 346
349 347
350WELS_EXTERN VAACalcSad_sse2
351;*************************************************************************************************************348;*************************************************************************************************************
352;void VAACalcSad_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight349;void VAACalcSad_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight
353; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8)350; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8)
354;*************************************************************************************************************351;*************************************************************************************************************
355 352
356 353
357ALIGN 16354WELS_EXTERN VAACalcSad_sse2
358VAACalcSad_sse2:
359%define cur_data esp + pushsize + 4355%define cur_data esp + pushsize + 4
360%define ref_data esp + pushsize + 8356%define ref_data esp + pushsize + 8
361%define iPicWidth esp + pushsize + 12357%define iPicWidth esp + pushsize + 12
@@ -441,12 +437,10 @@ width_loop: · 79 unmodified lines
441 437
442%else ;64-bit438%else ;64-bit
443 439
444WELS_EXTERN SampleVariance16x16_sse2
445;***********************************************************************440;***********************************************************************
446; void SampleVariance16x16_sse2( uint8_t * y_ref, int32_t y_ref_stride, uint8_t * y_src, int32_t y_src_stride,SMotionTextureUnit* pMotionTexture );441; void SampleVariance16x16_sse2( uint8_t * y_ref, int32_t y_ref_stride, uint8_t * y_src, int32_t y_src_stride,SMotionTextureUnit* pMotionTexture );
447;***********************************************************************442;***********************************************************************
448ALIGN 16443WELS_EXTERN SampleVariance16x16_sse2
449SampleVariance16x16_sse2:
450 %define SUM r10;[esp]444 %define SUM r10;[esp]
451 %define SUM_CUR r11;[esp+4]445 %define SUM_CUR r11;[esp+4]
452 %define SQR r13;[esp+8]446 %define SQR r13;[esp+8]
@@ -539,15 +533,13 @@ SampleVariance16x16_sse2: · 86 unmodified lines
539 ret533 ret
540 534
541 535
542WELS_EXTERN VAACalcSad_sse2
543;*************************************************************************************************************536;*************************************************************************************************************
544;void VAACalcSad_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight537;void VAACalcSad_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight
545; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8)538; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8)
546;*************************************************************************************************************539;*************************************************************************************************************
547 540
548 541
549ALIGN 16542WELS_EXTERN VAACalcSad_sse2
550VAACalcSad_sse2:
551%define cur_data r0543%define cur_data r0
552%define ref_data r1544%define ref_data r1
553%define iPicWidth r2545%define iPicWidth r2
@@ -637,15 +629,13 @@ width_loop: · 83 unmodified lines
637 629
638 630
639%ifdef X86_32631%ifdef X86_32
640WELS_EXTERN VAACalcSadVar_sse2
641;*************************************************************************************************************632;*************************************************************************************************************
642;void VAACalcSadVar_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight633;void VAACalcSadVar_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight
643; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16)634; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16)
644;*************************************************************************************************************635;*************************************************************************************************************
645 636
646 637
647ALIGN 16638WELS_EXTERN VAACalcSadVar_sse2
648VAACalcSadVar_sse2:
649%define localsize 8639%define localsize 8
650%define cur_data esp + pushsize + localsize + 4640%define cur_data esp + pushsize + localsize + 4
651%define ref_data esp + pushsize + localsize + 8641%define ref_data esp + pushsize + localsize + 8
@@ -773,15 +763,13 @@ var_width_loop: · 121 unmodified lines
773 763
774%else ;64-bit764%else ;64-bit
775 765
776WELS_EXTERN VAACalcSadVar_sse2
777;*************************************************************************************************************766;*************************************************************************************************************
778;void VAACalcSadVar_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight767;void VAACalcSadVar_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight
779; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16)768; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16)
780;*************************************************************************************************************769;*************************************************************************************************************
781 770
782 771
783ALIGN 16772WELS_EXTERN VAACalcSadVar_sse2
784VAACalcSadVar_sse2:
785%define cur_data arg1 ;r0773%define cur_data arg1 ;r0
786%define ref_data arg2 ;r1774%define ref_data arg2 ;r1
787%define iPicWidth arg3 ;r2775%define iPicWidth arg3 ;r2
@@ -916,15 +904,13 @@ var_width_loop: · 128 unmodified lines
916 904
917%ifdef X86_32905%ifdef X86_32
918 906
919WELS_EXTERN VAACalcSadSsd_sse2
920;*************************************************************************************************************907;*************************************************************************************************************
921;void VAACalcSadSsd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,908;void VAACalcSadSsd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
922; int32_t iPicStride,int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16, int32_t *psqdiff16x16)909; int32_t iPicStride,int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16, int32_t *psqdiff16x16)
923;*************************************************************************************************************910;*************************************************************************************************************
924 911
925 912
926ALIGN 16913WELS_EXTERN VAACalcSadSsd_sse2
927VAACalcSadSsd_sse2:
928%define localsize 12914%define localsize 12
929%define cur_data esp + pushsize + localsize + 4915%define cur_data esp + pushsize + localsize + 4
930%define ref_data esp + pushsize + localsize + 8916%define ref_data esp + pushsize + localsize + 8
@@ -1072,15 +1058,13 @@ sqdiff_width_loop: · 141 unmodified lines
1072%else1058%else
1073 1059
1074 1060
1075WELS_EXTERN VAACalcSadSsd_sse2
1076;*************************************************************************************************************1061;*************************************************************************************************************
1077;void VAACalcSadSsd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1062;void VAACalcSadSsd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1078; int32_t iPicStride,int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16, int32_t *psqdiff16x16)1063; int32_t iPicStride,int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16, int32_t *psqdiff16x16)
1079;*************************************************************************************************************1064;*************************************************************************************************************
1080 1065
1081 1066
1082ALIGN 161067WELS_EXTERN VAACalcSadSsd_sse2
1083VAACalcSadSsd_sse2:
1084%define localsize 121068%define localsize 12
1085%define cur_data arg1;r01069%define cur_data arg1;r0
1086%define ref_data arg2;r11070%define ref_data arg2;r1
@@ -1236,15 +1220,13 @@ sqdiff_width_loop: · 149 unmodified lines
1236%endif1220%endif
1237 1221
1238%ifdef X86_321222%ifdef X86_32
1239WELS_EXTERN VAACalcSadBgd_sse2
1240;*************************************************************************************************************1223;*************************************************************************************************************
1241;void VAACalcSadBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1224;void VAACalcSadBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1242; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *p_sd8x8, uint8_t *p_mad8x8)1225; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *p_sd8x8, uint8_t *p_mad8x8)
1243;*************************************************************************************************************1226;*************************************************************************************************************
1244 1227
1245 1228
1246ALIGN 161229WELS_EXTERN VAACalcSadBgd_sse2
1247VAACalcSadBgd_sse2:
1248%define localsize 121230%define localsize 12
1249%define cur_data esp + pushsize + localsize + 41231%define cur_data esp + pushsize + localsize + 4
1250%define ref_data esp + pushsize + localsize + 81232%define ref_data esp + pushsize + localsize + 8
@@ -1415,7 +1397,6 @@ bgd_width_loop: · 164 unmodified lines
1415 1397
1416 1398
1417 1399
1418WELS_EXTERN VAACalcSadSsdBgd_sse2
1419;*************************************************************************************************************1400;*************************************************************************************************************
1420;void VAACalcSadSsdBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1401;void VAACalcSadSsdBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1421; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16,1402; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16,
@@ -1423,8 +1404,7 @@ WELS_EXTERN VAACalcSadSsdBgd_sse2 · 1 unmodified line
1423;*************************************************************************************************************1404;*************************************************************************************************************
1424 1405
1425 1406
1426ALIGN 161407WELS_EXTERN VAACalcSadSsdBgd_sse2
1427VAACalcSadSsdBgd_sse2:
1428%define localsize 161408%define localsize 16
1429%define cur_data esp + pushsize + localsize + 41409%define cur_data esp + pushsize + localsize + 4
1430%define ref_data esp + pushsize + localsize + 81410%define ref_data esp + pushsize + localsize + 8
@@ -1646,15 +1626,13 @@ sqdiff_bgd_width_loop: · 215 unmodified lines
1646 ret1626 ret
1647%else1627%else
1648 1628
1649WELS_EXTERN VAACalcSadBgd_sse2
1650;*************************************************************************************************************1629;*************************************************************************************************************
1651;void VAACalcSadBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1630;void VAACalcSadBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1652; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *p_sd8x8, uint8_t *p_mad8x8)1631; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *p_sd8x8, uint8_t *p_mad8x8)
1653;*************************************************************************************************************1632;*************************************************************************************************************
1654 1633
1655 1634
1656ALIGN 161635WELS_EXTERN VAACalcSadBgd_sse2
1657VAACalcSadBgd_sse2:
1658%define cur_data arg1;1636%define cur_data arg1;
1659%define ref_data arg2;1637%define ref_data arg2;
1660%define iPicWidth arg3;1638%define iPicWidth arg3;
@@ -1817,7 +1795,6 @@ bgd_width_loop: · 156 unmodified lines
1817 1795
1818 1796
1819 1797
1820WELS_EXTERN VAACalcSadSsdBgd_sse2
1821;*************************************************************************************************************1798;*************************************************************************************************************
1822;void VAACalcSadSsdBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1799;void VAACalcSadSsdBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1823; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16,1800; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16,
@@ -1825,8 +1802,7 @@ WELS_EXTERN VAACalcSadSsdBgd_sse2 · 1 unmodified line
1825;*************************************************************************************************************1802;*************************************************************************************************************
1826 1803
1827 1804
1828ALIGN 161805WELS_EXTERN VAACalcSadSsdBgd_sse2
1829VAACalcSadSsdBgd_sse2:
1830%define cur_data arg1;1806%define cur_data arg1;
1831%define ref_data arg2;1807%define ref_data arg2;
1832%define iPicWidth arg3;1808%define iPicWidth arg3;

Keyboard shortcuts

General

Show keyboard shortcuts
?
Focus the current search
/

Issues and pull requests

Open filters
f

Issues, pull requests and Inbox

Select next item
j↓
Select previous item
k↑
Open selected item
Enter
Dismiss filters or clear selection
Esc

Repository code

Go to file
t