summaryrefslogtreecommitdiff
path: root/libavcodec/h264.h
Commit message (Collapse)AuthorAge
* Remove DECLARE_ALIGNED_{8,16} macrosMåns Rullgård2010-03-06
| | | | | | | These macros are redundant. All uses are replaced with the generic DECLARE_ALIGNED macro instead. Originally committed as revision 22233 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Port Optimizations about *_type init from decode to filter code.Michael Niedermayer2010-03-04
| | | | | | 1 cpu cycle faster Originally committed as revision 22193 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Optimize *_type init, 1.5 cpu cycles faster.Michael Niedermayer2010-03-04
| | | | Originally committed as revision 22192 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Reorder indexes in weight tables.Michael Niedermayer2010-03-03
| | | | | | 5 cpu cycles faster. Originally committed as revision 22183 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move all context fields that are not used in the mb and block layersMichael Niedermayer2010-03-03
| | | | | | | to the end of the structure. 4 cpu cycles faster in 3k cpu cycles Originally committed as revision 22181 to svn://svn.ffmpeg.org/ffmpeg/trunk
* remove unused left_border field from context.Michael Niedermayer2010-03-03
| | | | Originally committed as revision 22179 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Note about luma/chroma_weight tables and their datatype.Michael Niedermayer2010-03-03
| | | | Originally committed as revision 22177 to svn://svn.ffmpeg.org/ffmpeg/trunk
* move svq3 specific fields to the end of the contextMichael Niedermayer2010-03-03
| | | | Originally committed as revision 22171 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Merge weight & offset tables, 15 cpu cycles faster.Michael Niedermayer2010-03-03
| | | | Originally committed as revision 22169 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Another 3 useless zeroing instructions.Michael Niedermayer2010-03-03
| | | | Originally committed as revision 22162 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Load the whole left side of mv&ref only when needed.Michael Niedermayer2010-03-03
| | | | | | 30 cpu cycles faster Originally committed as revision 22161 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Merge h->slice_table[left_xy[0/1] ] checks, 4 cpu cycles speedupMichael Niedermayer2010-02-27
| | | | Originally committed as revision 22086 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Split *_type setting up, 4 cpu cycles faster.Michael Niedermayer2010-02-27
| | | | Originally committed as revision 22085 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Only load the topleft mv/ref when the topright is unavailable.Michael Niedermayer2010-02-26
| | | | | | 8 cpu cycles faster. Originally committed as revision 22079 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove some useless operations from the code setting left_cbp.Michael Niedermayer2010-02-26
| | | | | | maybe 0.5 cpu cycles faster Originally committed as revision 22078 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Simplify code to set cbp_*Michael Niedermayer2010-02-26
| | | | | | | | this seems 1 cpu cycle slower even though we practically just remove code. Speed loss seems caused by the merge of if(left_type), iam commiting this anyway as i cant imagine this to be anything but compiler messup. Originally committed as revision 22073 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move init of right side of ref_cache from fill_caches() to ↵Michael Niedermayer2010-02-26
| | | | | | init_the_darn_decoder(). Originally committed as revision 22071 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove 3 mv_cache zeroing instructions that zeroed the right side.Michael Niedermayer2010-02-26
| | | | | | This seems unneeded as nothing seems to ever set it to non zero values. Originally committed as revision 22070 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove useless check of the 2 left MBs of a pair being in the same slice.Michael Niedermayer2010-02-26
| | | | Originally committed as revision 22069 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove unneeded line of code from the neighbor setting code in h264.Michael Niedermayer2010-02-26
| | | | Originally committed as revision 22067 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Get rid of mb2b8_xy and b8_stride, change arrays organized based on b8_stride toMichael Niedermayer2010-02-25
| | | | | | | ones based on mb_stride in h264. about 20 cpu cycles faster overall per MB Originally committed as revision 22065 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Store data in direct_table interleaved.Michael Niedermayer2010-02-25
| | | | | | seems 20cpu cycles faster Originally committed as revision 22055 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Simplify intra4x4_pred_mode_cache init.Michael Niedermayer2010-02-25
| | | | Originally committed as revision 22054 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Reorder intra4x4_pred_mode so that we can read/write 4 values at once.Michael Niedermayer2010-02-25
| | | | | | 3-7 cpu cycles faster Originally committed as revision 22053 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Store intra4x4_pred_mode per row only.Michael Niedermayer2010-02-25
| | | | | | | | about 5 cpu cycles slower in the local code but should be overall faster due to reduced cache use. (my sample though has too few intra4x4 blocks for this to be meassureable easily either way) Originally committed as revision 22052 to svn://svn.ffmpeg.org/ffmpeg/trunk
* unroll tiny and trivial loop. Same speed but clearer.Michael Niedermayer2010-02-25
| | | | Originally committed as revision 22051 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Cut the size of mvd_table by yet another factor of 2.Michael Niedermayer2010-02-25
| | | | | | | The code read/write code itself was 1 cycle faster, overall its likely more due to cache effects Originally committed as revision 22048 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Keep mvd_table values of only 2 mb rows.Michael Niedermayer2010-02-25
| | | | Originally committed as revision 22047 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Change mvd_cache & mvd_table to 8bit, this is overall a bit fasterMichael Niedermayer2010-02-24
| | | | | | | for high resolution videos. about 20cycles faster per MB for cathederal. Originally committed as revision 22038 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Replace /2 by faster >>1 as the mvd values are now all positive.Michael Niedermayer2010-02-24
| | | | Originally committed as revision 22013 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove unused variable. Seems i forgot to commit this.Michael Niedermayer2010-02-24
| | | | Originally committed as revision 22012 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove unused variable, fixes warnings of the type:Diego Biurrun2010-02-21
| | | | | | libavcodec/h264.h:816: warning: unused variable `mb_xy' Originally committed as revision 21941 to svn://svn.ffmpeg.org/ffmpeg/trunk
* H264: use alias-safe macrosMåns Rullgård2010-02-18
| | | | | | | This eliminates all aliasing violation warnings in h264 code. No measurable speed difference with gcc-4.4.3 on i7. Originally committed as revision 21881 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move predict_field_decoding_flag() from h264.h to .c as its only used there ↵Michael Niedermayer2010-02-17
| | | | | | | | and belongs there as well. Originally committed as revision 21861 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move check for and call of predict_field_decoding_flag() from the mb code toMichael Niedermayer2010-02-17
| | | | | | the row code. This function would only be needed on a MB basis for MBAFF+FMO Originally committed as revision 21860 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Split setting neighboring MBs from fill_decode_caches()Michael Niedermayer2010-02-15
| | | | | | no speed change. Originally committed as revision 21842 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Store sub_mb_type in direct_cache/direct_table.Michael Niedermayer2010-02-14
| | | | | | This is equal complexity but could be more usefull. Originally committed as revision 21821 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Remove slice_table checks from decode_cabac_mb_cbp_luma() and set ↵Michael Niedermayer2010-02-14
| | | | | | | | left/top_cbp so these checks arent needed. Originally committed as revision 21819 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Revert r21814Michael Niedermayer2010-02-14
| | | | | | | | | | | | | | Log: h264: Fix pointer warnings by removing redundant [0] Fixes: h264.h:1222:38: warning: initialization from incompatible pointer type h264.h:1299:38: warning: initialization from incompatible pointer type h264.h:1314:42: warning: initialization from incompatible pointer type Reason: breaks h264 decoding & fate Originally committed as revision 21818 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Direct temporal skiped MBs dont need fill_decode_caches() at all so dont call itMichael Niedermayer2010-02-13
| | | | | | for them. Originally committed as revision 21816 to svn://svn.ffmpeg.org/ffmpeg/trunk
* h264: Remove unused variables.Alexander Strange2010-02-13
| | | | Originally committed as revision 21815 to svn://svn.ffmpeg.org/ffmpeg/trunk
* h264: Fix pointer warnings by removing redundant [0]Alexander Strange2010-02-13
| | | | | | | | | Fixes: h264.h:1222:38: warning: initialization from incompatible pointer type h264.h:1299:38: warning: initialization from incompatible pointer type h264.h:1314:42: warning: initialization from incompatible pointer type Originally committed as revision 21814 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Fix integer overflow warnings in h264.hAlexander Strange2010-02-13
| | | | | | | | | Fixes: h264.h: In function 'fill_filter_caches': h264.h:1216:73: warning: integer overflow in expression h264.h:1307:81: warning: integer overflow in expression Originally committed as revision 21813 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Merge (IS_SKIP(mb_type) || IS_DIRECT(mb_type)Michael Niedermayer2010-02-13
| | | | Originally committed as revision 21812 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Skiped MBs dont need the cbp stuff so skip initing that.Michael Niedermayer2010-02-13
| | | | Originally committed as revision 21811 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Also skip direct/mvd_cache init for skiped blocks.Michael Niedermayer2010-02-13
| | | | | | Odd thing is i thought ive tryed this already and it failed previously. Originally committed as revision 21809 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move more code under if(!IS_DIRECT(mb_type)).Michael Niedermayer2010-02-13
| | | | Originally committed as revision 21806 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Skip some more code that isnt needed for direct MBs.Michael Niedermayer2010-02-13
| | | | Originally committed as revision 21798 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Move setting MB_TYPE_L0L1 for direct MBs up, this is simpler.Michael Niedermayer2010-02-13
| | | | Originally committed as revision 21794 to svn://svn.ffmpeg.org/ffmpeg/trunk
* Dont calculate any surrounding MVs for temporal MBsMichael Niedermayer2010-02-13
| | | | Originally committed as revision 21793 to svn://svn.ffmpeg.org/ffmpeg/trunk
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-15 22:36:13 +0000