Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | 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> | ||
* | vorbisdsp: convert x86 simd functions from inline asm to yasm. | Ronald S. Bultje | 2013-01-22 |
| | |||
* | floatdsp: move scalarproduct_float from dsputil to avfloatdsp. | Ronald S. Bultje | 2013-01-22 |
| | | | | This makes the aac decoder and all voice codecs independent of dsputil. | ||
* | 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 |
| | |||
* | x86: ac3: Fix HAVE_MMXEXT condition to only refer to external assembly | Diego Biurrun | 2013-01-21 |
| | | | | CC: libav-stable@libav.org | ||
* | dsputilenc: x86: Convert pixel inline asm to yasm | Daniel Kang | 2013-01-21 |
| | | | | Signed-off-by: Luca Barbato <lu_zero@gentoo.org> | ||
* | vorbisdsp: change block_size type from int to intptr_t. | Ronald S. Bultje | 2013-01-20 |
| | | | | This saves one instruction in the x86-64 assembly. | ||
* | dsputil: remove butterflies_float_interleave. | Ronald S. Bultje | 2013-01-20 |
| | | | | The function is unused. | ||
* | dsputil: drop non-compliant "fast" qpel mc functions | Mans Rullgard | 2013-01-20 |
| | | | | Signed-off-by: Diego Biurrun <diego@biurrun.de> | ||
* | Move vorbis_inverse_coupling from dsputil to vorbisdspcontext. | Ronald S. Bultje | 2013-01-19 |
| | | | | | Conveniently (together with Justin's earlier patches), this makes our vorbis decoder entirely independent of dsputil. | ||
* | vp3: integrate clear_blocks with idct of previous block. | Ronald S. Bultje | 2013-01-19 |
| | | | | | | | | | This is identical to what e.g. vp8 does, and prevents the function call overhead (plus dependency on dsputil for this particular function). Arm asm updated by Janne Grunau <janne-libav@jannau.net>. Signed-off-by: Janne Grunau <janne-libav@jannau.net> | ||
* | x86: dsputil: Drop some unused macro definitions | Diego Biurrun | 2013-01-18 |
| | |||
* | lavc: Move vector_fmul_window to AVFloatDSPContext | Justin Ruggles | 2013-01-16 |
| | | | | Signed-off-by: Luca Barbato <lu_zero@gentoo.org> | ||
* | x86: Add PAVGB macro to abstract pavgb/pavgusb instruction via cpuflags | Diego Biurrun | 2013-01-15 |
| | |||
* | x86: ABS2: port to cpuflags | Diego Biurrun | 2013-01-14 |
| | |||
* | Drop Snow codec | Diego Biurrun | 2013-01-06 |
| | | | | Snow is a toy codec with no real-world use and horrible code. | ||
* | x86: sbrdsp: Implement SSE qmf_post_shuffle | Christophe Gisquet | 2013-01-06 |
| | | | | | | 255 to 174 cycles on Arrandale / Win64. Unrolling yields no gain. Signed-off-by: Diego Biurrun <diego@biurrun.de> | ||
* | x86: sbrdsp: Implement SSE sum64x5 | Christophe Gisquet | 2013-01-06 |
| | | | | | | 698 to 174 cycles on Arrandale. Unrolling is a 6 cycles gain. Signed-off-by: Diego Biurrun <diego@biurrun.de> | ||
* | x86: ABS1: port to cpuflags | Diego Biurrun | 2013-01-06 |
| | |||
* | lavc: introduce VideoDSPContext | Ronald S. Bultje | 2012-12-20 |
| | | | | | | | | Move some functions from dsputil. The idea is that videodsp contains functions that are useful for a large and varied set of video decoders. Currently, it contains emulated_edge_mc() and prefetch(). Signed-off-by: Luca Barbato <lu_zero@gentoo.org> | ||
* | x86inc: support stack mem allocation and re-alignment in PROLOGUE | Ronald S. Bultje | 2012-12-12 |
| | | | | | | | Use this in VP8/H264-8bit loopfilter functions so they can be used if there is no aligned stack (e.g. MSVC 32bit or ICC 10.x). Signed-off-by: Luca Barbato <lu_zero@gentoo.org> | ||
* | ac3dec: make downmix() take array of pointers to channel data | Mans Rullgard | 2012-12-09 |
| | |||
* | SBR DSP x86: implement SSE sbr_hf_gen | Christophe Gisquet | 2012-12-07 |
| | | | | | | | | | | | | Start and end index are multiple of 2, therefore guaranteeing aligned access. Also, this allows to generate 4 floats per loop, keeping the alignment all along. Timing: - 32 bits: 326c -> 172c - 64 bits: 323c -> 156c Signed-off-by: Diego Biurrun <diego@biurrun.de> | ||
* | x86: dsputilenc: port to cpuflags | Diego Biurrun | 2012-11-28 |
| | |||
* | x86: h264dsp: Fix linking with yasm and optimizations disabled | Diego Biurrun | 2012-11-28 |
| | | | | | Some optimized functions reference optimized symbols, so the functions must be explicitly disabled when those symbols are unavailable. | ||
* | x86: h264_idct: port to cpuflags | Diego Biurrun | 2012-11-28 |
| | |||
* | x86: h264_weight: port to cpuflags | Diego Biurrun | 2012-11-27 |
| | |||
* | x86: fix build without inline asm | Diego Biurrun | 2012-11-26 |
| | | | | | | | | The qpel functions referenced here are not related to h264 and should thus never have been under CONFIG_H264QPEL. Signed-off-by: Mans Rullgard <mans@mansr.com> Signed-off-by: Diego Biurrun <diego@biurrun.de> | ||
* | x86: h264 qpel: use the correct number of utilized xmm regs in cglobal | Justin Ruggles | 2012-11-25 |
| | | | | Fixes xmm register clobbering on win64. | ||
* | x86: h264: Convert 8-bit QPEL inline assembly to YASM | Daniel Kang | 2012-11-25 |
| | | | | Signed-off-by: Diego Biurrun <diego@biurrun.de> | ||
* | x86: h264: Remove 3dnow QPEL code | Daniel Kang | 2012-11-25 |
| | | | | | | | The only CPUs that have 3dnow and don't have mmxext are 12 years old. Moreover, AMD has dropped 3dnow extensions from newer CPUs. Signed-off-by: Diego Biurrun <diego@biurrun.de> | ||
* | x86: h264_chromamc: port to cpuflags | Diego Biurrun | 2012-11-25 |
| | |||
* | x86: h264_intrapred: Fix C function names in comments | Diego Biurrun | 2012-11-18 |
| | | | | | Function names changed after switching to declaration with PRED4x4/8x8/8x8L/16x16 macros in the C code. | ||
* | x86: SPLATD: port to cpuflags | Diego Biurrun | 2012-11-18 |
| | |||
* | x86: dsputil: port to cpuflags | Diego Biurrun | 2012-11-16 |
| | |||
* | x86: mmx2 ---> mmxext in asm constructs | Diego Biurrun | 2012-11-14 |
| | |||
* | x86: h264_weight_10bit: port to cpuflags | Diego Biurrun | 2012-11-13 |
| | |||
* | build: Drop AVX assembly ifdefs | Diego Biurrun | 2012-11-11 |
| | | | | An assembler able to cope with AVX instructions is now required. | ||
* | x86: h264_qpel_10bit: drop unused parameter from MC10/MC20/MC30 macros | Diego Biurrun | 2012-11-10 |
| | |||
* | x86: PALIGNR: port to cpuflags | Diego Biurrun | 2012-11-09 |
| | |||
* | x86: h264_qpel_10bit: port to cpuflags | Diego Biurrun | 2012-11-09 |
| | |||
* | x86: h264_intrapred: port to cpuflags | Diego Biurrun | 2012-11-08 |
| | |||
* | x86: h264qpel: Only define mmxext QPEL functions if H264QPEL is enabled | Diego Biurrun | 2012-11-05 |
| | | | | This fixes compilation with --disable-everything and components enabled. | ||
* | x86: PABSW: port to cpuflags | Diego Biurrun | 2012-11-05 |
| | |||
* | x86: vc1dsp: port to cpuflags | Diego Biurrun | 2012-11-05 |
| | |||
* | x86: Refactor PSWAPD fallback implementations and port to cpuflags | Diego Biurrun | 2012-11-02 |
| | |||
* | x86: PMINUB: port to cpuflags | Diego Biurrun | 2012-11-02 |
| | |||
* | x86: ac3dsp: port to cpuflags | Diego Biurrun | 2012-11-02 |
| | |||
* | x86: Move optimization suffix to end of function names | Diego Biurrun | 2012-10-31 |
| | | | | This simplifies cpuflags porting. |