History

Fold ALIGN 16 and the function label into WELS_EXTERN

This simplifies the structure for all x86 assembly functions, reducing the amount of duplicated code structure.

Martin Storsjö committed 12 years ago

Commit eec96821 parent 918b21121 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
@@ -255,12 +255,10 @@ SECTION .text · 254 unmodified lines
255 255
256%ifdef X86_32256%ifdef X86_32
257 257
258WELS_EXTERN SampleVariance16x16_sse2
259;***********************************************************************258;***********************************************************************
260; void SampleVariance16x16_sse2( uint8_t * y_ref, int32_t y_ref_stride, uint8_t * y_src, int32_t y_src_stride,SMotionTextureUnit* pMotionTexture );259; void SampleVariance16x16_sse2( uint8_t * y_ref, int32_t y_ref_stride, uint8_t * y_src, int32_t y_src_stride,SMotionTextureUnit* pMotionTexture );
261;***********************************************************************260;***********************************************************************
262ALIGN 16261WELS_EXTERN SampleVariance16x16_sse2
263SampleVariance16x16_sse2:
264 push esi262 push esi
265 push edi263 push edi
266 push ebx264 push ebx
@@ -357,15 +355,13 @@ SampleVariance16x16_sse2: · 90 unmodified lines
357 355
358 356
359 357
360WELS_EXTERN VAACalcSad_sse2
361;*************************************************************************************************************358;*************************************************************************************************************
362;void VAACalcSad_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight359;void VAACalcSad_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight
363; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8)360; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8)
364;*************************************************************************************************************361;*************************************************************************************************************
365 362
366 363
367ALIGN 16364WELS_EXTERN VAACalcSad_sse2
368VAACalcSad_sse2:
369%define cur_data esp + pushsize + 4365%define cur_data esp + pushsize + 4
370%define ref_data esp + pushsize + 8366%define ref_data esp + pushsize + 8
371%define iPicWidth esp + pushsize + 12367%define iPicWidth esp + pushsize + 12
@@ -451,12 +447,10 @@ width_loop: · 79 unmodified lines
451 447
452%else ;64-bit448%else ;64-bit
453 449
454WELS_EXTERN SampleVariance16x16_sse2
455;***********************************************************************450;***********************************************************************
456; void SampleVariance16x16_sse2( uint8_t * y_ref, int32_t y_ref_stride, uint8_t * y_src, int32_t y_src_stride,SMotionTextureUnit* pMotionTexture );451; void SampleVariance16x16_sse2( uint8_t * y_ref, int32_t y_ref_stride, uint8_t * y_src, int32_t y_src_stride,SMotionTextureUnit* pMotionTexture );
457;***********************************************************************452;***********************************************************************
458ALIGN 16453WELS_EXTERN SampleVariance16x16_sse2
459SampleVariance16x16_sse2:
460 %define SUM r10;[esp]454 %define SUM r10;[esp]
461 %define SUM_CUR r11;[esp+4]455 %define SUM_CUR r11;[esp+4]
462 %define SQR r13;[esp+8]456 %define SQR r13;[esp+8]
@@ -549,15 +543,13 @@ SampleVariance16x16_sse2: · 86 unmodified lines
549 ret543 ret
550 544
551 545
552WELS_EXTERN VAACalcSad_sse2
553;*************************************************************************************************************546;*************************************************************************************************************
554;void VAACalcSad_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight547;void VAACalcSad_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight
555; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8)548; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8)
556;*************************************************************************************************************549;*************************************************************************************************************
557 550
558 551
559ALIGN 16552WELS_EXTERN VAACalcSad_sse2
560VAACalcSad_sse2:
561%define cur_data r0553%define cur_data r0
562%define ref_data r1554%define ref_data r1
563%define iPicWidth r2555%define iPicWidth r2
@@ -647,15 +639,13 @@ width_loop: · 83 unmodified lines
647 639
648 640
649%ifdef X86_32641%ifdef X86_32
650WELS_EXTERN VAACalcSadVar_sse2
651;*************************************************************************************************************642;*************************************************************************************************************
652;void VAACalcSadVar_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight643;void VAACalcSadVar_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight
653; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16)644; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16)
654;*************************************************************************************************************645;*************************************************************************************************************
655 646
656 647
657ALIGN 16648WELS_EXTERN VAACalcSadVar_sse2
658VAACalcSadVar_sse2:
659%define localsize 8649%define localsize 8
660%define cur_data esp + pushsize + localsize + 4650%define cur_data esp + pushsize + localsize + 4
661%define ref_data esp + pushsize + localsize + 8651%define ref_data esp + pushsize + localsize + 8
@@ -783,15 +773,13 @@ var_width_loop: · 121 unmodified lines
783 773
784%else ;64-bit774%else ;64-bit
785 775
786WELS_EXTERN VAACalcSadVar_sse2
787;*************************************************************************************************************776;*************************************************************************************************************
788;void VAACalcSadVar_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight777;void VAACalcSadVar_sse2( const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight
789; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16)778; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16)
790;*************************************************************************************************************779;*************************************************************************************************************
791 780
792 781
793ALIGN 16782WELS_EXTERN VAACalcSadVar_sse2
794VAACalcSadVar_sse2:
795%define cur_data arg1 ;r0783%define cur_data arg1 ;r0
796%define ref_data arg2 ;r1784%define ref_data arg2 ;r1
797%define iPicWidth arg3 ;r2785%define iPicWidth arg3 ;r2
@@ -926,15 +914,13 @@ var_width_loop: · 128 unmodified lines
926 914
927%ifdef X86_32915%ifdef X86_32
928 916
929WELS_EXTERN VAACalcSadSsd_sse2
930;*************************************************************************************************************917;*************************************************************************************************************
931;void VAACalcSadSsd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,918;void VAACalcSadSsd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
932; int32_t iPicStride,int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16, int32_t *psqdiff16x16)919; int32_t iPicStride,int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16, int32_t *psqdiff16x16)
933;*************************************************************************************************************920;*************************************************************************************************************
934 921
935 922
936ALIGN 16923WELS_EXTERN VAACalcSadSsd_sse2
937VAACalcSadSsd_sse2:
938%define localsize 12924%define localsize 12
939%define cur_data esp + pushsize + localsize + 4925%define cur_data esp + pushsize + localsize + 4
940%define ref_data esp + pushsize + localsize + 8926%define ref_data esp + pushsize + localsize + 8
@@ -1082,15 +1068,13 @@ sqdiff_width_loop: · 141 unmodified lines
1082%else1068%else
1083 1069
1084 1070
1085WELS_EXTERN VAACalcSadSsd_sse2
1086;*************************************************************************************************************1071;*************************************************************************************************************
1087;void VAACalcSadSsd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1072;void VAACalcSadSsd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1088; int32_t iPicStride,int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16, int32_t *psqdiff16x16)1073; int32_t iPicStride,int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16, int32_t *psqdiff16x16)
1089;*************************************************************************************************************1074;*************************************************************************************************************
1090 1075
1091 1076
1092ALIGN 161077WELS_EXTERN VAACalcSadSsd_sse2
1093VAACalcSadSsd_sse2:
1094%define localsize 121078%define localsize 12
1095%define cur_data arg1;r01079%define cur_data arg1;r0
1096%define ref_data arg2;r11080%define ref_data arg2;r1
@@ -1246,15 +1230,13 @@ sqdiff_width_loop: · 149 unmodified lines
1246%endif1230%endif
1247 1231
1248%ifdef X86_321232%ifdef X86_32
1249WELS_EXTERN VAACalcSadBgd_sse2
1250;*************************************************************************************************************1233;*************************************************************************************************************
1251;void VAACalcSadBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1234;void VAACalcSadBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1252; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *p_sd8x8, uint8_t *p_mad8x8)1235; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *p_sd8x8, uint8_t *p_mad8x8)
1253;*************************************************************************************************************1236;*************************************************************************************************************
1254 1237
1255 1238
1256ALIGN 161239WELS_EXTERN VAACalcSadBgd_sse2
1257VAACalcSadBgd_sse2:
1258%define localsize 121240%define localsize 12
1259%define cur_data esp + pushsize + localsize + 41241%define cur_data esp + pushsize + localsize + 4
1260%define ref_data esp + pushsize + localsize + 81242%define ref_data esp + pushsize + localsize + 8
@@ -1425,7 +1407,6 @@ bgd_width_loop: · 164 unmodified lines
1425 1407
1426 1408
1427 1409
1428WELS_EXTERN VAACalcSadSsdBgd_sse2
1429;*************************************************************************************************************1410;*************************************************************************************************************
1430;void VAACalcSadSsdBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1411;void VAACalcSadSsdBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1431; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16,1412; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16,
@@ -1433,8 +1414,7 @@ WELS_EXTERN VAACalcSadSsdBgd_sse2 · 1 unmodified line
1433;*************************************************************************************************************1414;*************************************************************************************************************
1434 1415
1435 1416
1436ALIGN 161417WELS_EXTERN VAACalcSadSsdBgd_sse2
1437VAACalcSadSsdBgd_sse2:
1438%define localsize 161418%define localsize 16
1439%define cur_data esp + pushsize + localsize + 41419%define cur_data esp + pushsize + localsize + 4
1440%define ref_data esp + pushsize + localsize + 81420%define ref_data esp + pushsize + localsize + 8
@@ -1656,15 +1636,13 @@ sqdiff_bgd_width_loop: · 215 unmodified lines
1656 ret1636 ret
1657%else1637%else
1658 1638
1659WELS_EXTERN VAACalcSadBgd_sse2
1660;*************************************************************************************************************1639;*************************************************************************************************************
1661;void VAACalcSadBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1640;void VAACalcSadBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1662; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *p_sd8x8, uint8_t *p_mad8x8)1641; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *p_sd8x8, uint8_t *p_mad8x8)
1663;*************************************************************************************************************1642;*************************************************************************************************************
1664 1643
1665 1644
1666ALIGN 161645WELS_EXTERN VAACalcSadBgd_sse2
1667VAACalcSadBgd_sse2:
1668%define cur_data arg1;1646%define cur_data arg1;
1669%define ref_data arg2;1647%define ref_data arg2;
1670%define iPicWidth arg3;1648%define iPicWidth arg3;
@@ -1827,7 +1805,6 @@ bgd_width_loop: · 156 unmodified lines
1827 1805
1828 1806
1829 1807
1830WELS_EXTERN VAACalcSadSsdBgd_sse2
1831;*************************************************************************************************************1808;*************************************************************************************************************
1832;void VAACalcSadSsdBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,1809;void VAACalcSadSsdBgd_sse2(const uint8_t *cur_data, const uint8_t *ref_data, int32_t iPicWidth, int32_t iPicHeight,
1833; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16,1810; int32_t iPicStride, int32_t *psadframe, int32_t *psad8x8, int32_t *psum16x16, int32_t *psqsum16x16,
@@ -1835,8 +1812,7 @@ WELS_EXTERN VAACalcSadSsdBgd_sse2 · 1 unmodified line
1835;*************************************************************************************************************1812;*************************************************************************************************************
1836 1813
1837 1814
1838ALIGN 161815WELS_EXTERN VAACalcSadSsdBgd_sse2
1839VAACalcSadSsdBgd_sse2:
1840%define cur_data arg1;1816%define cur_data arg1;
1841%define ref_data arg2;1817%define ref_data arg2;
1842%define iPicWidth arg3;1818%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