Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | bfin: add README | Michael Niedermayer | 2014-06-02 | 1 | -0/+6 |
| | | | | | | Based on sh4/README Suggested-by: jamrial Signed-off-by: Michael Niedermayer <michaelni@gmx.at> | ||||
* | Remove all Blackfin architecture optimizations | Diego Biurrun | 2014-06-02 | 12 | -2066/+0 |
| | | | | | | Blackfin is a painful platform to work with, no test machines are available and the range of multimedia applications is dubious. Thus it only represents a maintenance burden. | ||||
* | dsputil: Move ff_zigzag_direct and ff_crop_tab declarations to mathops.h | Diego Biurrun | 2014-03-22 | 1 | -1/+1 |
| | |||||
* | dsputil: Use correct type in me_cmp_func function pointer | Diego Biurrun | 2014-03-20 | 1 | -11/+12 |
| | |||||
* | dsputil: Conditionally compile dsputil code on all architectures | Diego Biurrun | 2014-03-20 | 1 | -8/+8 |
| | |||||
* | dsputil: Propagate bit depth information to all (sub)init functions | Diego Biurrun | 2014-03-20 | 1 | -7/+5 |
| | | | | This avoids recalculating the value over and over again. | ||||
* | bfin: Use more sensible file names | Diego Biurrun | 2014-03-13 | 8 | -10/+10 |
| | | | | | Drop non-informative _bfin suffix, rename some files for consistency with other architectures, rename others to reflect their content. | ||||
* | bfin: hpeldsp: Split no_rnd pixel operations off into a separate file | Diego Biurrun | 2014-03-13 | 3 | -64/+82 |
| | | | | This saves some space in configurations that do not enable hpeldsp. | ||||
* | bfin: dsputil: Unconditionally compile hpel pixels code | Diego Biurrun | 2014-03-13 | 1 | -2/+2 |
| | | | | The functions it contains are used from unconditionally compiled code. | ||||
* | bfin: hpeldsp: "nornd" ---> "no_rnd" function name suffix for consistency | Diego Biurrun | 2014-03-13 | 2 | -34/+34 |
| | |||||
* | bfin: Move all assembly declarations only used within one .c file there | Diego Biurrun | 2014-03-13 | 6 | -95/+43 |
| | | | | This is done similarly for all other architectures. | ||||
* | bfin: hpeldsp: K&R formatting cosmetics | Diego Biurrun | 2014-03-13 | 2 | -35/+62 |
| | |||||
* | bfin: Refactor duplicated assembly-related macros | Diego Biurrun | 2014-03-13 | 6 | -99/+15 |
| | |||||
* | bfin: Refactor duplicated l1_text/l1_data_B macros | Diego Biurrun | 2014-03-13 | 2 | -17/+2 |
| | |||||
* | bfin: dsputil: K&R formatting cosmetics | Diego Biurrun | 2014-03-13 | 3 | -94/+112 |
| | | | | Also add a few #endif comments. | ||||
* | bfin: dsputil: Drop broken and trivial profiling code | Diego Biurrun | 2014-03-13 | 1 | -40/+0 |
| | |||||
* | bfin: hpeldsp: Drop broken put_pixels_*_xy2_nornd bits | Diego Biurrun | 2014-03-13 | 3 | -114/+0 |
| | | | | They were marked as broken and disabled in 2009 and will never get fixed. | ||||
* | bfin: dsputil: Drop broken vsad bits | Diego Biurrun | 2014-03-13 | 1 | -12/+0 |
| | | | | They were marked as broken and disabled in 2009 and will never get fixed. | ||||
* | bfin: dsputil: Remove unused global variable | Diego Biurrun | 2014-03-13 | 1 | -2/+0 |
| | |||||
* | dsputil: cosmetics: Lose camelCase on ff_cropTbl and ff_squareTbl names | Diego Biurrun | 2014-03-13 | 1 | -2/+2 |
| | | | | Also switch from "tbl" to "tab" name suffixes. | ||||
* | Update dsputil- and SIMD-related comments to match reality more closely | Diego Biurrun | 2014-03-13 | 1 | -1/+1 |
| | |||||
* | bfin: vp3: Mark all binary object sections appropriately | Diego Biurrun | 2014-01-22 | 1 | -1/+1 |
| | | | | This fixes a leftover from e44d7c659ba56d7efd6de10d5d99a1c44fd997fa. | ||||
* | mpegvideo: remove disabled bfin asm | Anton Khirnov | 2014-01-03 | 2 | -148/+0 |
| | | | | It has been disabled as 'broken' over 3 years ago in b716a792 | ||||
* | cosmetics: bfin: Fix indentation in the dsputil init function | Martin Storsjö | 2013-04-19 | 1 | -2/+2 |
| | | | | Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | bfin: hpeldsp: Move half-pel assembly from dsputil to hpeldsp | Ronald S. Bultje | 2013-04-19 | 7 | -465/+552 |
| | | | | Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | vp3: Use full transpose for all IDCTs | Ronald S. Bultje | 2013-04-15 | 1 | -1/+0 |
| | | | | | | | | | | | This way, the special IDCT permutations are no longer needed. This is similar to how H264 does it, and removes the dsputil dependency imposed by the scantable code. Also remove the unused type == 0 cases from the plain C version of the idct. Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | vp3: bfin: Transpose the IDCTs | Martin Storsjö | 2013-04-15 | 1 | -2/+8 |
| | | | | | | | | | While this change isn't bitexact, the IDCTs weren't bitexact to start with either. This simplifies decoupling vp3 from dsputil. Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | bfin: Match DEFUN_END macros to the right functions | Martin Storsjö | 2013-04-15 | 1 | -1/+2 |
| | | | | Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | lavc: Make pointers to ff_cropTbl const | Martin Storsjö | 2013-04-08 | 1 | -1/+1 |
| | | | | | | There's no point in these pointers not being const. Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | bfin: Make vp3 functions static | Martin Storsjö | 2013-04-08 | 2 | -6/+4 |
| | | | | Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | bfin: Don't use the vp3 idct functions if bitexact behaviour is expected | Martin Storsjö | 2013-04-08 | 1 | -2/+4 |
| | | | | | | | | In the non-bitexact mode, vp3 currently decodes to the same frame crcs as before 28f9ab702 (and the output visually looks correct). Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | Remove unnecessary dsputil.h #includes | Diego Biurrun | 2013-02-26 | 1 | -1/+0 |
| | |||||
* | Add av_cold attributes to arch-specific init functions | Diego Biurrun | 2013-02-05 | 3 | -3/+6 |
| | |||||
* | Use ptrdiff_t instead of int for {avg, put}_pixels line_size parameter. | Diego Biurrun | 2013-02-05 | 1 | -14/+14 |
| | | | | | This avoids SIMD-optimized functions having to sign-extend their line size argument manually to be able to do pointer arithmetic. | ||||
* | bfin: unbreak compilation | Luca Barbato | 2013-01-29 | 2 | -0/+30 |
| | | | | Added a missing header file. | ||||
* | bfin: Separate VP3 initialization code | Diego Biurrun | 2013-01-28 | 4 | -13/+11 |
| | |||||
* | bfin: update VP3 idct | Luca Barbato | 2013-01-28 | 1 | -0/+6 |
| | | | | The block must be set to 0. | ||||
* | Drop DCTELEM typedef | Diego Biurrun | 2013-01-22 | 9 | -29/+30 |
| | | | | | | It does not help as an abstraction and adds dsputil dependencies. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com> | ||||
* | lavc: put FF_IDCT_{VP3,H264,CAVS,BINK,EA,WMV2} under FF_API_IDCT. | Ronald S. Bultje | 2013-01-20 | 1 | -1/+2 |
| | | | | | | This allows us to get rid of them on the next major bump. All of the above are functionally irrelevant, and most of them are unused, except the vp3 one, which is used wrongly in the bfin arch optimizations. | ||||
* | build: non-x86: Only compile mpegvideo optimizations when necessary | Diego Biurrun | 2012-10-09 | 1 | -1/+2 |
| | |||||
* | build: prettyprinting cosmetics | Diego Biurrun | 2012-03-26 | 1 | -7/+7 |
| | |||||
* | mpegvideo: Add ff_ prefix to nonstatic functions | Martin Storsjö | 2012-02-15 | 1 | -1/+1 |
| | | | | Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | dsputil: Add ff_ prefix to the dsputil*_init* functions | Martin Storsjö | 2012-02-15 | 1 | -1/+1 |
| | | | | Signed-off-by: Martin Storsjö <martin@martin.st> | ||||
* | cosmetics: Delete empty lines at end of file. | Diego Biurrun | 2012-02-09 | 7 | -13/+0 |
| | |||||
* | Replace some forgotten FFmpeg references by Libav. | Diego Biurrun | 2011-11-02 | 2 | -8/+6 |
| | |||||
* | dsputil: update per-arch init funcs for non-h264 high bit depth | Mans Rullgard | 2011-07-21 | 1 | -1/+1 |
| | | | | Signed-off-by: Mans Rullgard <mans@mansr.com> | ||||
* | dsputil: template get_pixels() for different bit depths | Mans Rullgard | 2011-07-21 | 1 | -1/+1 |
| | | | | Signed-off-by: Mans Rullgard <mans@mansr.com> | ||||
* | jfdctint: add 10-bit version | Mans Rullgard | 2011-07-21 | 1 | -3/+3 |
| | | | | Signed-off-by: Mans Rullgard <mans@mansr.com> | ||||
* | simple_idct: add 10-bit version | Mans Rullgard | 2011-07-20 | 1 | -10/+12 |
| | | | | Signed-off-by: Mans Rullgard <mans@mansr.com> | ||||
* | Adds 8-, 9- and 10-bit versions of some of the functions used by the h264 ↵ | Oskar Arvidsson | 2011-05-10 | 1 | -0/+5 |
| | | | | | | | | | | | | | | | | | decoder. This patch lets e.g. dsputil_init chose dsp functions with respect to the bit depth to decode. The naming scheme of bit depth dependent functions is <base name>_<bit depth>[_<prefix>] (i.e. the old clear_blocks_c is now named clear_blocks_8_c). Note: Some of the functions for high bit depth is not dependent on the bit depth, but only on the pixel size. This leaves some room for optimizing binary size. Preparatory patch for high bit depth h264 decoding support. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com> |