Statistics
| Branch: | Revision:

ffmpeg / libavcodec / arm @ 4ae3ee4a

Name Size Revision Age Author Comment
Makefile 3.43 KB 4ae3ee4a over 9 years Mans Rullgard VP8: ARM optimised decode_block_coeffs_internal...
aac.h 6.1 KB 1c6f46be almost 11 years Måns Rullgård Add missing guards and includes to arm/aac.h O...
asm-offsets.h 1.26 KB 451b4b86 over 9 years Mans Rullgard Rearrange MpegEncContext to simplify access fro...
asm.S 2.13 KB 01b75fa9 over 9 years Mans Rullgard ARM: add helper macro for declaring constant da...
dcadsp_init_arm.c 1.12 KB a8ae4e0e over 9 years Justin Ruggles Remove unneeded add bias from 3 functions. DSP...
dcadsp_neon.S 2.11 KB a8ae4e0e over 9 years Justin Ruggles Remove unneeded add bias from 3 functions. DSP...
dsputil_arm.S 24.4 KB a7e7d40c over 10 years Måns Rullgård ARM: set size of asm functions in object files ...
dsputil_arm.h 1.25 KB 4a89e0a6 over 10 years Måns Rullgård ARM: add some missing includes Originally comm...
dsputil_armv6.S 20.5 KB a7e7d40c over 10 years Måns Rullgård ARM: set size of asm functions in object files ...
dsputil_init_arm.c 5.23 KB c6c98d08 about 10 years Stefano Sabatini Move mm_support() from libavcodec to libavutil,...
dsputil_init_armv5te.c 1.59 KB 55c0e1e6 about 11 years Måns Rullgård ARM: add ff_ prefix to lots of functions Origi...
dsputil_init_armv6.c 4.93 KB 41c2bd0a over 10 years Måns Rullgård ARMv6 optimised pix_sum Originally committed a...
dsputil_init_neon.c 17.3 KB fe2ff6d2 over 9 years Justin Ruggles Separate format conversion DSP functions from D...
dsputil_init_vfp.c 1.26 KB fe2ff6d2 over 9 years Justin Ruggles Separate format conversion DSP functions from D...
dsputil_iwmmxt.c 8.72 KB 7160bb71 about 10 years Stefano Sabatini Rename FF_MM_ symbols related to CPU features f...
dsputil_iwmmxt_rnd_template.c 47 KB a2fc0f6a almost 12 years Måns Rullgård ARM: replace "armv4l" with "arm" Originally co...
dsputil_neon.S 26.4 KB fe2ff6d2 over 9 years Justin Ruggles Separate format conversion DSP functions from D...
dsputil_vfp.S 5.1 KB fe2ff6d2 over 9 years Justin Ruggles Separate format conversion DSP functions from D...
fft_init_arm.c 2.24 KB a8ae4e0e over 9 years Justin Ruggles Remove unneeded add bias from 3 functions. DSP...
fft_neon.S 15 KB 1ee076b1 over 10 years Loren Merritt more credits to D. J. Bernstein for fft Origin...
fmtconvert_init_arm.c 1.69 KB fe2ff6d2 over 9 years Justin Ruggles Separate format conversion DSP functions from D...
fmtconvert_neon.S 14.5 KB fe2ff6d2 over 9 years Justin Ruggles Separate format conversion DSP functions from D...
fmtconvert_vfp.S 2.69 KB fe2ff6d2 over 9 years Justin Ruggles Separate format conversion DSP functions from D...
h264dsp_init_arm.c 7.06 KB fa2d5d54 about 10 years Måns Rullgård ARM: NEON H264 8x8 IDCT Parts by David Conrad....
h264dsp_neon.S 61.7 KB 75148437 over 10 years Måns Rullgård ARM: NEON H264 chroma loop filter 3 cycles fast...
h264idct_neon.S 13 KB 4a6cc8fa about 10 years Måns Rullgård ARM: fix NEON h264_idct_add8 Originally commit...
h264pred_init_arm.c 3.31 KB a815602a over 10 years Ronald S. Bultje Reindent after r23716. Originally committed as...
h264pred_neon.S 11.8 KB 0fc1961e over 9 years Mans Rullgard ARM: NEON: fix overflow in h264 16x16 planar pr...
int_neon.S 4.01 KB a7e7d40c over 10 years Måns Rullgård ARM: set size of asm functions in object files ...
jrevdct_arm.S 13.6 KB 2ad4c241 about 11 years Måns Rullgård ARM: make function names all-lowercase Origina...
mathops.h 3.09 KB deb1b2b6 about 11 years Diego Biurrun Add necessary #include for config.h. Originall...
mdct_neon.S 12.2 KB 588d28ac over 10 years Måns Rullgård Remove vestiges of radix-2 FFT Patch (mostly) ...
mpegvideo_arm.c 2.04 KB 68dacb4e over 10 years Måns Rullgård ARM: check struct offsets only when they are us...
mpegvideo_arm.h 969 Bytes 4a89e0a6 over 10 years Måns Rullgård ARM: add some missing includes Originally comm...
mpegvideo_armv5te.c 2.99 KB 5bacc3ad over 10 years Måns Rullgård ARM: move mpegvideo prototypes to a header file...
mpegvideo_armv5te_s.S 3.96 KB a7e7d40c over 10 years Måns Rullgård ARM: set size of asm functions in object files ...
mpegvideo_iwmmxt.c 5.07 KB 7160bb71 about 10 years Stefano Sabatini Rename FF_MM_ symbols related to CPU features f...
mpegvideo_neon.S 3.65 KB 451b4b86 over 9 years Mans Rullgard Rearrange MpegEncContext to simplify access fro...
rdft_neon.S 6.34 KB 84368aa6 over 10 years Måns Rullgård ARM: fix build for darwin/iphone References to...
simple_idct_arm.S 21.5 KB 2ad4c241 about 11 years Måns Rullgård ARM: make function names all-lowercase Origina...
simple_idct_armv5te.S 18.8 KB a7e7d40c over 10 years Måns Rullgård ARM: set size of asm functions in object files ...
simple_idct_armv6.S 13.1 KB a7e7d40c over 10 years Måns Rullgård ARM: set size of asm functions in object files ...
simple_idct_neon.S 12.4 KB 96088566 over 10 years Måns Rullgård ARM: remove unnecessary .previous directive Or...
synth_filter_neon.S 4.14 KB a8ae4e0e over 9 years Justin Ruggles Remove unneeded add bias from 3 functions. DSP...
vp3dsp_neon.S 13.8 KB c0fda017 over 10 years David Conrad vp3: 10l Fix DC-only IDCT for C and ARM too Or...
vp56_arith.h 3.52 KB 5da7494d over 9 years Mans Rullgard ARM optimised vp56_rac_get_prob() Approximatel...
vp56dsp_init_arm.c 1.25 KB 5635985c over 10 years Måns Rullgård ARM: NEON optimised VP6 edge filter Originally...
vp56dsp_neon.S 4.78 KB 5635985c over 10 years Måns Rullgård ARM: NEON optimised VP6 edge filter Originally...
vp8.h 1.11 KB 4ae3ee4a over 9 years Mans Rullgard VP8: ARM optimised decode_block_coeffs_internal...
vp8_armv6.S 7.61 KB 4ae3ee4a over 9 years Mans Rullgard VP8: ARM optimised decode_block_coeffs_internal...
vp8dsp_init_arm.c 8.41 KB ef15d71c over 9 years Mans Rullgard VP8: ARM NEON optimisations for dsp functions ...
vp8dsp_neon.S 66.4 KB ef15d71c over 9 years Mans Rullgard VP8: ARM NEON optimisations for dsp functions ...

Latest revisions

# Date Author Comment
4ae3ee4a 02/12/2011 11:52 PM Mans Rullgard

VP8: ARM optimised decode_block_coeffs_internal

Approximately 5% faster on Cortex-A8.

Signed-off-by: Mans Rullgard <>
(cherry picked from commit a7878c9f73c12cfa685bd8af8f3afcca85f56a8b)

5da7494d 02/12/2011 11:52 PM Mans Rullgard

ARM optimised vp56_rac_get_prob()

Approximately 3% faster on Cortex-A8.

Signed-off-by: Mans Rullgard <>
(cherry picked from commit 7da48fd0111adf504cfcfc5ebda7fd0681968041)

ef15d71c 02/09/2011 02:31 AM Mans Rullgard

VP8: ARM NEON optimisations for dsp functions

This adds NEON optimised versions of all functions in VP8DSPContext.
Based on initial work by Rob Clark.

Signed-off-by: Mans Rullgard <>
(cherry picked from commit a1c1d3c003b0ec16fdb6574913781313fb2c7ab6)

01b75fa9 02/04/2011 02:08 AM Mans Rullgard

ARM: add helper macro for declaring constant data

Signed-off-by: Mans Rullgard <>
(cherry picked from commit b9a639ddd6e102edbf94a26b5106386792e2e4c8)

fe2ff6d2 02/04/2011 02:08 AM Justin Ruggles

Separate format conversion DSP functions from DSPContext.

This will be beneficial for use with the audio conversion API without
requiring it to depend on all of dsputil.

Signed-off-by: Mans Rullgard <>
(cherry picked from commit c73d99e672329c8f2df290736ffc474c360ac4ae)

a8ae4e0e 02/02/2011 02:40 AM Justin Ruggles

Remove unneeded add bias from 3 functions.

DSPContext.vector_fmul_window()
DCADSPContext.lfe_fir()
SynthFilterContext.synth_filter_float()

Signed-off-by: Mans Rullgard <>
(cherry picked from commit 80ba1ddb58b5923b9f36a6acd542affc4ca722eb)

451b4b86 01/30/2011 02:41 AM Mans Rullgard

Rearrange MpegEncContext to simplify access from asm

This moves the fields needed by asm near the top, before any
structs or other members which complicate the offset calculation.
Modifying other structs will no longer require updating the offsets,
and the asm code is slightly simpler due to the smaller offsets....

8afac88e 01/30/2011 02:41 AM Mans Rullgard

ARM: update MpegEncContext offsets
(cherry picked from commit 0745116c10e5650952c39ea543c66b12fba854e6)

0fc1961e 01/26/2011 02:43 AM Mans Rullgard

ARM: NEON: fix overflow in h264 16x16 planar pred

Signed-off-by: Mans Rullgard <>
(cherry picked from commit 78f318be59a8e6174f21c2d7c3403ef325c73011)

015f9f1a 01/23/2011 06:32 PM Justin Ruggles

Change DSPContext.vector_fmul() from dst=dst*src to dest=src0*src1.

Signed-off-by: Mans Rullgard <>
(cherry picked from commit 6eabb0d3ad42b91c1b4c298718c29961f7c1653a)

View revisions

Also available in: Atom