2013-10-07 |
Loren Merritt | x86inc: create xm# and ym#, analagous to m#
|
commit | commitdiff | tree |
2013-10-07 |
Loren Merritt | x86inc: fix some corner cases of SWAP
|
commit | commitdiff | tree |
2013-10-07 |
Loren Merritt | build: Strip spurious labels
|
commit | commitdiff | tree |
2013-10-07 |
Loren Merritt | x86inc: activate REP_RET automatically
|
commit | commitdiff | tree |
2013-07-02 |
Loren Merritt | lls/x86: use 3-operator vaddpd in ADDPD_MEM
|
commit | commitdiff | tree |
2013-06-30 |
Loren Merritt | x86: lpc: fix a segfault in av_evaluate_lls_sse2()
|
commit | commitdiff | tree |
2013-06-29 |
Loren Merritt | lpc: use levinson for the first pass of multipass cholesky
|
commit | commitdiff | tree |
2013-06-29 |
Loren Merritt | x86: lpc: simd av_evaluate_lls
|
commit | commitdiff | tree |
2013-06-29 |
Loren Merritt | x86: lpc: simd av_update_lls
|
commit | commitdiff | tree |
2013-06-29 |
Loren Merritt | lpc: use function pointers, in preparation for asm
|
commit | commitdiff | tree |
2013-06-29 |
Loren Merritt | lpc: remove "decay" argument
|
commit | commitdiff | tree |
2013-03-13 |
Loren Merritt | hqdn3d: Fix out of array read in LOWPASS
|
commit | commitdiff | tree |
2012-08-26 |
Loren Merritt | vf_hqdn3d: x86 asm
|
commit | commitdiff | tree |
2012-08-26 |
Loren Merritt | vf_hqdn3d: support 16bit colordepth
|
commit | commitdiff | tree |
2012-07-28 |
Loren Merritt | vf_hqdn3d: support 9 and 10bit colordepth
|
commit | commitdiff | tree |
2012-07-28 |
Loren Merritt | vf_hqdn3d: reduce intermediate precision
|
commit | commitdiff | tree |
2012-07-28 |
Loren Merritt | vf_hqdn3d: simplify and optimize
|
commit | commitdiff | tree |
2012-07-28 |
Loren Merritt | factor identical ff_inplace_start_frame out of two...
|
commit | commitdiff | tree |
2012-07-28 |
Loren Merritt | vf_hqdn3d: cosmetics
|
commit | commitdiff | tree |
2012-07-28 |
Loren Merritt | x86inc: clip num_args to 7 on x86-32.
|
commit | commitdiff | tree |
2012-07-05 |
Loren Merritt | x86inc: add SPLATB_LOAD, SPLATB_REG, PSHUFLW macros
|
commit | commitdiff | tree |
2012-07-05 |
Loren Merritt | x86inc: modify ALIGN to not generate long nops on i586
|
commit | commitdiff | tree |
2012-04-19 |
Loren Merritt | x86inc: support AVX abstraction for 2-operand instructions
|
commit | commitdiff | tree |
2012-03-04 |
Loren Merritt | x86inc: don't "bake" stack_offset in named arguments.
|
commit | commitdiff | tree |
2011-11-06 |
Loren Merritt | Update x86inc.asm to latest x264 version, and add AVX...
|
commit | commitdiff | tree |
2011-07-04 |
Loren Merritt | vf_gradfun: relicense x86 asm to LGPL
|
commit | commitdiff | tree |
2011-06-02 |
Loren Merritt | Cosmetic changes to h264_idct_10bit.asm.
|
commit | commitdiff | tree |
2011-06-02 |
Loren Merritt | 2x faster h264_idct_add8_10.
|
commit | commitdiff | tree |
2011-05-22 |
Loren Merritt | dct32_sse: eliminate some spills
|
commit | commitdiff | tree |
2011-02-13 |
Loren Merritt | FFT: factor a shuffle out of the inner loop and merge...
|
commit | commitdiff | tree |
2010-08-28 |
Loren Merritt | cosmetics in imdct_sse
|
commit | commitdiff | tree |
2010-07-22 |
Loren Merritt | relicense h264 deblock sse2 to lgpl
|
commit | commitdiff | tree |
2010-07-21 |
Loren Merritt | sync yasm macros from x264
|
commit | commitdiff | tree |
2010-07-18 |
Loren Merritt | more credits to D. J. Bernstein for fft
|
commit | commitdiff | tree |
2010-07-01 |
Loren Merritt | PPC: Altivec IMDCT Patch by Loren Merritt
|
commit | commitdiff | tree |
2010-05-19 |
Loren Merritt | 13% faster yuv420 to rgb15 mmx.
|
commit | commitdiff | tree |
2010-05-19 |
Loren Merritt | 40% faster yuv420 to rgb24 mmx.
|
commit | commitdiff | tree |
2010-05-12 |
Loren Merritt | change a variable-length array to a malloc.
|
commit | commitdiff | tree |
2010-03-29 |
Loren Merritt | indent
|
commit | commitdiff | tree |
2010-03-29 |
Loren Merritt | optimize init_vlc().
|
commit | commitdiff | tree |
2010-01-26 |
Loren Merritt | optimize h264_loop_filter_strength_mmx2
|
commit | commitdiff | tree |
2009-12-09 |
Loren Merritt | revert r20775, because equivalent code has been committed...
|
commit | commitdiff | tree |
2009-12-08 |
Loren Merritt | fix a crash in ape decoding on x86_32 sse2
|
commit | commitdiff | tree |
2009-12-05 |
Loren Merritt | slightly faster scalarproduct_and_madd_int16_ssse3...
|
commit | commitdiff | tree |
2009-12-05 |
Loren Merritt | r20739 broke compilation on systems without yasm
|
commit | commitdiff | tree |
2009-12-05 |
Loren Merritt | refactor and optimize scalarproduct
|
commit | commitdiff | tree |
2009-12-04 |
Loren Merritt | 1-13% faster apply_filter, 1-3% faster ape decoding...
|
commit | commitdiff | tree |
2009-12-03 |
Loren Merritt | port ape dsp functions from sse2 to mmx
|
commit | commitdiff | tree |
2009-12-03 |
Loren Merritt | avoid an unpredictable branch
|
commit | commitdiff | tree |
2009-12-03 |
Loren Merritt | inline to allow constant propagation
|
commit | commitdiff | tree |
2009-10-25 |
Loren Merritt | s/movdqa/movaps/ in sse1 fft. (regression in r20293)
|
commit | commitdiff | tree |
2009-10-18 |
Loren Merritt | fix linking on systems with a function name prefix...
|
commit | commitdiff | tree |
2009-10-18 |
Loren Merritt | sync yasm macros to x264
|
commit | commitdiff | tree |
2009-10-18 |
Loren Merritt | huffyuv: add some const qualifiers
|
commit | commitdiff | tree |
2009-10-18 |
Loren Merritt | simd add_hfyu_left_prediction
|
commit | commitdiff | tree |
2009-09-21 |
Loren Merritt | extend ff_inverse[], and fix its documentation
|
commit | commitdiff | tree |
2009-08-27 |
Loren Merritt | indent
|
commit | commitdiff | tree |
2009-08-27 |
Loren Merritt | 1.5x faster ff_vorbis_floor1_render_list, 5% faster...
|
commit | commitdiff | tree |
2009-08-22 |
Loren Merritt | bring back some randomness in fft-test. (regression...
|
commit | commitdiff | tree |
2009-08-11 |
Loren Merritt | fix a crash in SVQ1 with cmp!=sad
|
commit | commitdiff | tree |
2009-07-04 |
Loren Merritt | fix a buffer overrun on corrupt huffyuv streams
|
commit | commitdiff | tree |
2009-02-15 |
Loren Merritt | Add my GPG fingerprint
|
commit | commitdiff | tree |
2009-02-08 |
Loren Merritt | ff_add_hfyu_median_prediction_mmx2
|
commit | commitdiff | tree |
2009-02-08 |
Loren Merritt | re-enable mid_pred asm on x86_64. (broke in r16681)
|
commit | commitdiff | tree |
2009-01-28 |
Loren Merritt | fix an overflow in vc1 intra overlap filter
|
commit | commitdiff | tree |
2008-12-25 |
Loren Merritt | flatten an array, since gcc fails at optimizing multidimensi...
|
commit | commitdiff | tree |
2008-12-18 |
Loren Merritt | 1.6x faster pred*_vertical_add_c
|
commit | commitdiff | tree |
2008-12-10 |
Loren Merritt | clear_block mmx
|
commit | commitdiff | tree |
2008-12-10 |
Loren Merritt | fix progressive jpeg:
|
commit | commitdiff | tree |
2008-08-24 |
Loren Merritt | document some dsp alignments
|
commit | commitdiff | tree |
2008-08-15 |
Loren Merritt | document some dsp alignments
|
commit | commitdiff | tree |
2008-08-14 |
Loren Merritt | file which should have been added in r14749
|
commit | commitdiff | tree |
2008-08-14 |
Loren Merritt | missing prototype
|
commit | commitdiff | tree |
2008-08-14 |
Loren Merritt | gcc chokes on the 7 registers needed for float_to_int16_inte...
|
commit | commitdiff | tree |
2008-08-14 |
Loren Merritt | gcc chokes on xmm constraints, so pessimize int32_to_float_f...
|
commit | commitdiff | tree |
2008-08-13 |
Loren Merritt | special case 6 channel version of float_to_int16_interleave
|
commit | commitdiff | tree |
2008-08-13 |
Loren Merritt | simd int->float
|
commit | commitdiff | tree |
2008-08-13 |
Loren Merritt | simd downmix
|
commit | commitdiff | tree |
2008-08-13 |
Loren Merritt | don't mark the delayed samples for upmixing if they...
|
commit | commitdiff | tree |
2008-08-13 |
Loren Merritt | oops, non-normalized downmix to mono contaminated add_bias...
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | optimize ac3_downmix.
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | align the audio decoding buffer, since some codecs...
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | use float_to_int16_interleave in ac3
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | use imdct_half in ac3
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | cosmetics in dsp init
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | mdct wrapper function to match fft
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | remove mdct tmp buffer
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | remove a float->double->float conversion.
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | optimize imdct_half:
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | indent
|
commit | commitdiff | tree |
2008-08-12 |
Loren Merritt | split-radix FFT
|
commit | commitdiff | tree |
2008-08-11 |
Loren Merritt | import yasm macros from x264
|
commit | commitdiff | tree |
2008-08-11 |
Loren Merritt | yasm buildsystem
|
commit | commitdiff | tree |
2008-07-29 |
Loren Merritt | r14205 broke vorbis float_to_int on ppc
|
commit | commitdiff | tree |
2008-07-16 |
Loren Merritt | float_to_int16_interleave: change src to an array of...
|
commit | commitdiff | tree |
2008-07-15 |
Loren Merritt | vorbis: reduce memory by reusing buffers
|
commit | commitdiff | tree |
2008-07-15 |
Loren Merritt | unroll another common case in vorbis_residue_decode
|
commit | commitdiff | tree |
2008-07-15 |
Loren Merritt | 10l, float_to_int16_interleave_sse/3dnow wrote the...
|
commit | commitdiff | tree |
2008-07-13 |
Loren Merritt | exploit mdct symmetry
|
commit | commitdiff | tree |
2008-07-13 |
Loren Merritt | simplify vorbis windowing
|
commit | commitdiff | tree |
next |