Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | ppc: Move vec_unaligned_load macro to util_altivec | Diego Biurrun | 2014-05-28 |
| | | | | This allows reusing it from multiple files. | ||
* | dsputil: Split off HuffYUV decoding bits into their own context | Diego Biurrun | 2014-05-27 |
| | | | | Also shorten HuffYUV context member names to avoid clutter. | ||
* | ppc: Drop a bunch of unnecessary dsputil_altivec.h #includes | Diego Biurrun | 2014-04-04 |
| | |||
* | On2 VP7 decoder | Peter Ross | 2014-04-04 |
| | | | | | | | | | Further performance improvements and security fixes by Vittorio Giovara, Luca Barbato and Diego Biurrun. Signed-off-by: Vittorio Giovara <vittorio.giovara@gmail.com> Signed-off-by: Luca Barbato <lu_zero@gentoo.org> Signed-off-by: Diego Biurrun <diego@biurrun.de> | ||
* | ppc: dsputil: Move hpeldsp-related declarations to a separate header | Diego Biurrun | 2014-03-22 |
| | |||
* | dsputil: Use correct type in me_cmp_func function pointer | Diego Biurrun | 2014-03-20 |
| | |||
* | build: Group general components separate from de/encoders in arch Makefiles | Diego Biurrun | 2014-03-20 |
| | | | | This is in line with how the top-level libavcodec Makefile is structured. | ||
* | dsputil: Conditionally compile dsputil code on all architectures | Diego Biurrun | 2014-03-20 |
| | |||
* | dsputil: Propagate bit depth information to all (sub)init functions | Diego Biurrun | 2014-03-20 |
| | | | | This avoids recalculating the value over and over again. | ||
* | ppc: dsputil: Drop trailing semicolon from macros | Diego Biurrun | 2014-03-20 |
| | | | | This allows for a more natural macro usage. | ||
* | ppc: dsputil: Merge some declarations and initializations | Diego Biurrun | 2014-03-20 |
| | |||
* | ppc: dsputil: Simplify some ifdeffed function definitions | Diego Biurrun | 2014-03-20 |
| | |||
* | ppc: dsputil: Drop some unnecessary parentheses | Diego Biurrun | 2014-03-20 |
| | |||
* | ppc: dsputil: more K&R formatting cosmetics | Diego Biurrun | 2014-03-20 |
| | |||
* | ppc: dsputil: K&R formatting cosmetics | Diego Biurrun | 2014-03-20 |
| | |||
* | ppc: dsputil: comment formatting and wording/grammar improvements | Diego Biurrun | 2014-03-20 |
| | |||
* | ppc: Add some missing headers | Diego Biurrun | 2014-03-13 |
| | |||
* | dsputil: Remove prototypes for nonexisting optimization functions | Diego Biurrun | 2014-03-13 |
| | |||
* | ppc: reduce overreads when loading 8 pixels in altivec dsp functions | Janne Grunau | 2014-02-14 |
| | | | | | | | | | | | Altivec can only load naturally aligned vectors. To handle possibly unaligned data a second vector is loaded from an offset of the original location and the data is recovered through a vector permutation. Overreads are minimal if the offset for second load points to the last element of data. This is 7 for loading eight 8-bit pixels and overreads are reduced from 16 bytes to 8 bytes if the pixels are 64-bit aligned. For unaligned pixels the overread is reduced from 23 bytes to 15 bytes in the worst case. | ||
* | vp8: fix PPC assembly to work if src_stride != dst_stride | Ronald S. Bultje | 2014-02-09 |
| | | | | | Signed-off-by: Anton Khirnov <anton@khirnov.net> Signed-off-by: Janne Grunau <janne-libav@jannau.net> | ||
* | h264: do not use 422 functions for monochrome | Anton Khirnov | 2014-01-06 |
| | | | | | | | Fixes invalid memory access. Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind CC:libav-stable@libav.org | ||
* | ppc: cosmetics: Consistently format CPU flag detection invocations | Diego Biurrun | 2013-08-29 |
| | |||
* | ppc: Add missing AltiVec cpuflag detection invocations | Diego Biurrun | 2013-08-29 |
| | |||
* | ppc: fdct: Remove vim editor settings comment | Diego Biurrun | 2013-08-28 |
| | |||
* | ppc: Add and use convenience macro to check for AltiVec availability | Diego Biurrun | 2013-08-28 |
| | |||
* | altivec: perform an explicit unaligned load | Kostya Shishkov | 2013-08-16 |
| | | | | | | | | Implicit vector loads on POWER7 hardware can use the VSX instruction set instead of classic Altivec/VMX. Let's force a VMX load in this case. Signed-off-by: Martin Storsjö <martin@martin.st> | ||
* | Consistently use "cpu_flags" as variable/parameter name for CPU flags | Diego Biurrun | 2013-07-18 |
| | |||
* | fmtconvert: Explicitly use int32_t instead of int | Christophe Gisquet | 2013-07-17 |
| | | | | Signed-off-by: Martin Storsjö <martin@martin.st> | ||
* | fix scalarproduct_and_madd_int16_altivec() for orders > 16 | Kostya Shishkov | 2013-05-26 |
| | | | | the second and third sources were incremented only by half of the needed size | ||
* | ppc: Only compile AltiVec FFT assembly when AltiVec is enabled | Diego Biurrun | 2013-05-02 |
| | |||
* | ppc: Drop unnecessary ff_ name prefixes from static functions | Diego Biurrun | 2013-04-30 |
| | |||
* | ppc: More consistent arch initialization | Diego Biurrun | 2013-04-30 |
| | |||
* | ppc: Move AltiVec utility headers out of AltiVec ifdefs | Diego Biurrun | 2013-04-30 |
| | | | | | Now that the headers themselves have ifdef protection this is no longer necessary and more consistent with normal include handling. | ||
* | ppc: More consistent names for H.264 optimizations files | Diego Biurrun | 2013-04-30 |
| | |||
* | mpegaudiosp: More consistent names for ppc/x86 optimization files | Diego Biurrun | 2013-04-30 |
| | |||
* | ppc: hpeldsp: Include attributes.h | Martin Storsjö | 2013-04-20 |
| | | | | | | This fixes building in configurations where altivec is disabled. Signed-off-by: Martin Storsjö <martin@martin.st> | ||
* | ppc: hpeldsp: Move half-pel assembly from dsputil to hpeldsp | Ronald S. Bultje | 2013-04-19 |
| | | | | Signed-off-by: Martin Storsjö <martin@martin.st> | ||
* | vp3: Use full transpose for all IDCTs | Ronald S. Bultje | 2013-04-15 |
| | | | | | | | | | | | 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> | ||
* | h264: Integrate clear_blocks calls with IDCT | Ronald S. Bultje | 2013-04-10 |
| | | | | | | | | | The non-intra-pcm branch in hl_decode_mb (simple, 8bpp) goes from 700 to 672 cycles, and the complete loop of decode_mb_cabac and hl_decode_mb (in the decode_slice loop) goes from 1759 to 1733 cycles on the clip tested (cathedral), i.e. almost 30 cycles per mb faster. Signed-off-by: Martin Storsjö <martin@martin.st> | ||
* | dsputil: convert remaining functions to use ptrdiff_t strides | Luca Barbato | 2013-03-12 |
| | | | | Signed-off-by: Luca Barbato <lu_zero@gentoo.org> | ||
* | Remove unnecessary dsputil.h #includes | Diego Biurrun | 2013-02-26 |
| | |||
* | dsputil: Move LOCAL_ALIGNED macros to libavutil | Diego Biurrun | 2013-02-08 |
| | |||
* | dsputil: Separate h264chroma | Diego Biurrun | 2013-02-06 |
| | |||
* | Add av_cold attributes to arch-specific init functions | Diego Biurrun | 2013-02-05 |
| | |||
* | Use ptrdiff_t instead of int for {avg, put}_pixels line_size parameter. | Diego Biurrun | 2013-02-05 |
| | | | | | This avoids SIMD-optimized functions having to sign-extend their line size argument manually to be able to do pointer arithmetic. | ||
* | ppc: fmtconvert: Drop two unused variables. | Diego Biurrun | 2013-02-01 |
| | |||
* | dsputil: Separate h264 qpel | Mans Rullgard | 2013-01-24 |
| | | | | | | | | | | The sh4 optimizations are removed, because the code is 100% identical to the C code, so it is unlikely to provide any real practical benefit. Signed-off-by: Diego Biurrun <diego@biurrun.de> Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com> Signed-off-by: Luca Barbato <lu_zero@gentoo.org> | ||
* | Drop DCTELEM typedef | Diego Biurrun | 2013-01-22 |
| | | | | | | It does not help as an abstraction and adds dsputil dependencies. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com> | ||
* | floatdsp: move vector_fmul_reverse from dsputil to avfloatdsp. | Ronald S. Bultje | 2013-01-22 |
| | | | | | | Now, nellymoserenc and aacenc no longer depends on dsputil. Independent of this patch, wmaprodec also does not depend on dsputil, so I removed it from there also. | ||
* | floatdsp: move vector_fmul_add from dsputil to avfloatdsp. | Ronald S. Bultje | 2013-01-22 |
| |