Commit Graph

24940 Commits

Author SHA1 Message Date
Michael Niedermayer
5f312139df Merge commit '45ee556d51ef04d79d52bf6b0b7f28a4d231cb0c'
* commit '45ee556d51ef04d79d52bf6b0b7f28a4d231cb0c':
  qdm2: Whitespace cosmetics
  flac: use meaningful return values

Conflicts:
	libavcodec/flacdec.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-29 11:34:07 +02:00
Michael Niedermayer
688331c51c Merge commit 'e13a929314e4a9a09d3146648f876f162f46cc32'
* commit 'e13a929314e4a9a09d3146648f876f162f46cc32':
  xl: K&R formatting cosmetics

Conflicts:
	libavcodec/xl.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-29 11:08:02 +02:00
Paul B Mahol
19da86f3ff targaenc: remove unused TargaContext
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-28 22:42:40 +00:00
Paul B Mahol
e999f2339a truemotion2: make code independent of sizeof(AVFrame)
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-28 22:42:40 +00:00
Paul B Mahol
a5155294e5 pnm: do not set avctx->coded_frame
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-28 22:27:36 +00:00
Paul B Mahol
329221eee7 pnm: fix 255 < maxval < 65535 for pam with depth 1
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-28 22:27:36 +00:00
Carl Eugen Hoyos
2c046c718a Revert "Avoid overflows when reading pgm files with maxval != 255 and != 65535."
The patch worked on little endian because pgm decoding was broken
but it was not correct in any way.

This reverts commit 7651c0e49b.
2013-07-29 00:05:43 +02:00
Paul B Mahol
768e40b451 Revert "pnm: remove nonsense code"
Breaks decoding pgms with 256 < maxval < 65535.

Found-by: Carl Eugen Hoyos <cehoyos@ag.or.at>.

This reverts commit a0348d0966.
2013-07-28 21:45:37 +00:00
Luca Barbato
3ca1dd2502 xl: Make sure the width is valid
And undo the wrong commit f1cb490d6d

CC: libav-stable@libav.org
2013-07-28 19:36:44 +02:00
Carl Eugen Hoyos
7651c0e49b Avoid overflows when reading pgm files with maxval != 255 and != 65535.
libjpeg v6b apparently does not initialize the most significant bits
to zero when writing 12bit pgm's with maxval 4095.
2013-07-28 19:07:06 +02:00
Carl Eugen Hoyos
7e8e8ba9cc Assume gray8 if 1 < maxval <= 255 in pgm. 2013-07-28 18:12:23 +02:00
Carl Eugen Hoyos
ae4dc0b37a Set bits_per_raw_sample when decoding mjpeg / ljpeg.
The ljpeg decoder also supports gray images with
8 - 15 bits per component.
2013-07-28 17:13:01 +02:00
Paul B Mahol
b09d86c636 utvideoenc: use av_image_copy_plane()
Signed-off-by: Paul B Mahol <onemda@gmail.com>
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
2013-07-28 10:57:51 -04:00
Luca Barbato
62b1e3b103 aasc: Check minimum buffer size
Prevent some overreads.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
2013-07-28 15:41:59 +02:00
Luca Barbato
45ee556d51 qdm2: Whitespace cosmetics 2013-07-28 14:58:19 +02:00
Luca Barbato
0e78ef0f94 flac: use meaningful return values 2013-07-28 14:58:18 +02:00
Michael Niedermayer
b0edd59994 Merge commit 'bf4b0ed1d5d323050a87c9f0ad1dd40860eb3da2'
* commit 'bf4b0ed1d5d323050a87c9f0ad1dd40860eb3da2':
  Add missing deprecation attributes

Conflicts:
	libavfilter/avfilter.h

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-28 13:44:34 +02:00
Luca Barbato
e13a929314 xl: K&R formatting cosmetics 2013-07-28 12:45:05 +02:00
Luca Barbato
f1cb490d6d xl: Fix the buffer size check
Also make it the first check.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
2013-07-28 12:45:05 +02:00
Michael Niedermayer
102397d2c1 Merge commit '16c22122c788b5e54a2f2e224bd0106429f0714c'
* commit '16c22122c788b5e54a2f2e224bd0106429f0714c':
  h264: K&R formatting cosmetics

Conflicts:
	libavcodec/h264.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-28 11:40:49 +02:00
Marton Balint
63c0113588 lavc: do not override format if neither text nor bitmap codec prop is set
Some decoders may output both, this way the decoder can set it.

Signed-off-by: Marton Balint <cus@passwd.hu>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-27 23:03:00 +02:00
Paul B Mahol
94ad38e2e8 dpx: use av_image_copy_plane()
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 20:51:14 +00:00
Paul B Mahol
6a57494204 dpxenc: cosmetics: reindent
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 20:45:48 +00:00
Paul B Mahol
9d099b9ae4 targa_y216dec: remove empty function
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 20:27:15 +00:00
Paul B Mahol
0909f3edbd zmbv: return meaningful error code
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 20:21:33 +00:00
Paul B Mahol
60a7fac61b utvideoenc: use av_image_copy_plane()
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 20:07:48 +00:00
Tim.Nicholson
84e345b38e Forward interlaced field information from mov to v210 decoder.
Reported and tested by Dave Rice.

Signed-off-by: Carl Eugen Hoyos <cehoyos@ag.or.at>
2013-07-27 18:57:28 +02:00
Diego Biurrun
bf4b0ed1d5 Add missing deprecation attributes 2013-07-27 16:08:49 +02:00
Michael Niedermayer
3adf054b22 avcodec/vorbisenc: change 6 asserts to av_asserts()
speed relevant ones use av_assert2()

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-27 15:23:13 +02:00
Michael Niedermayer
5e763920a4 avcodec/mpegvideo: update disabled assert() to av_assert0()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-27 15:23:13 +02:00
Paul B Mahol
88b071a473 exr: set pict_type
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 12:02:19 +00:00
Paul B Mahol
d9954ccff0 targa: set pict_type
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 12:02:18 +00:00
Paul B Mahol
3ea5d01a12 escape130: make "chroma_vals" static
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 12:02:18 +00:00
Paul B Mahol
6dbcecd78e lclenc: allocate FF_INPUT_BUFFER_PADDING_SIZE for extradata
Also check return value.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 12:02:18 +00:00
Paul B Mahol
573018c61f lclenc: cosmetics: reindent
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 12:00:25 +00:00
Paul B Mahol
86e722ab97 sgidec: safer check for buffer overflow
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 12:00:25 +00:00
Paul B Mahol
997e2b59e3 sgidec: return meaningful error codes
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 12:00:25 +00:00
Paul B Mahol
5efa5103b0 sgidec: simplify return path
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-27 12:00:25 +00:00
Diego Biurrun
16c22122c7 h264: K&R formatting cosmetics 2013-07-27 10:47:38 +02:00
Michael Niedermayer
205f423e28 libavcodec/mpegvideo: change some asserts to av_assert0()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-26 22:29:55 +02:00
Michael Niedermayer
c88bbc01eb avcodec/mpegvideo: replace assert about interlaced EC by debug av_log()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-26 22:29:47 +02:00
Paul B Mahol
1fa0284c57 qcelpdec: change asserts to av_asserts
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-26 18:32:27 +00:00
Paul B Mahol
fdf9296df7 sgienc: remove unused SgiContext
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-26 18:21:44 +00:00
Paul B Mahol
2868035d95 pcxenc: get rid of unused PCXContext
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-26 14:00:40 +00:00
Paul B Mahol
3838637bbb yop: check return value of av_frame_alloc()
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-26 13:51:26 +00:00
Paul B Mahol
9437259276 vmnc: check return values of reallocation
Also avoid memory leaks.

Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-26 13:49:36 +00:00
Paul B Mahol
cf5ab8b6f7 vmnc: make code independent of sizeof(AVFrame)
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-07-26 13:39:40 +00:00
Michael Niedermayer
46ad2d9e44 Merge remote-tracking branch 'qatar/master'
* qatar/master:
  miscellaneous typo fixes

Conflicts:
	configure
	libavformat/avisynth.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-26 11:12:11 +02:00
Michael Niedermayer
5dc2c99055 Merge commit '93a51984a27f3ba84d4e6f13d0c704ee9891603e'
* commit '93a51984a27f3ba84d4e6f13d0c704ee9891603e':
  mpeg12: Ignore slice threading if hwaccel is active

Conflicts:
	libavcodec/mpeg12dec.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-26 10:57:51 +02:00
Michael Niedermayer
bbc7d33a0d Merge commit '582963a8156522582e55466be4a59974a8d909a5'
* commit '582963a8156522582e55466be4a59974a8d909a5':
  vdpau: Fix VC-1 interlaced mode

Conflicts:
	libavcodec/vdpau_vc1.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-26 10:51:42 +02:00
Michael Niedermayer
af7949fdea Merge commit '4a2ef39442bf7f0150db07a1fbfcf8286e4d44a3'
* commit '4a2ef39442bf7f0150db07a1fbfcf8286e4d44a3':
  cosmetics: Add '0' to float constants ending in '.'.

Conflicts:
	libavcodec/ra288.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-26 10:43:12 +02:00
Michael Niedermayer
e28fabb4cb Merge commit '6c145ecf785dc3d26ba3fed3ea9892cc80244625'
* commit '6c145ecf785dc3d26ba3fed3ea9892cc80244625':
  twinvq: K&R formatting cosmetics

Conflicts:
	libavcodec/twinvq.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-26 10:30:47 +02:00
Diego Biurrun
03039f4c8c miscellaneous typo fixes 2013-07-25 19:43:32 +02:00
Rémi Denis-Courmont
93a51984a2 mpeg12: Ignore slice threading if hwaccel is active
Slice threading does not work with hardware acceleration, as decoding
is per-picture.  This fixes Bugzilla #542.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
2013-07-25 19:43:32 +02:00
Carl Eugen Hoyos
9375879d57 Show max bitrate for mpeg2 video streams in avcodec_string().
See also 25b7aa9
2013-07-25 19:08:19 +02:00
Rainer Hochecker
582963a815 vdpau: Fix VC-1 interlaced mode
VDPAU expects the bitstream value (0, 2 or 3).
libavcodec uses an enum (0, 1 or 2).

Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
Signed-off-by: Diego Biurrun <diego@biurrun.de>
2013-07-25 14:19:17 +02:00
Diego Biurrun
4a2ef39442 cosmetics: Add '0' to float constants ending in '.'. 2013-07-25 11:33:23 +02:00
Diego Biurrun
6c145ecf78 twinvq: K&R formatting cosmetics 2013-07-25 11:17:06 +02:00
Michael Niedermayer
0815b23062 Merge remote-tracking branch 'qatar/master'
* qatar/master:
  arm: Comment out unused labels in simple_idct_arm

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-25 10:15:08 +02:00
Michael Niedermayer
6db57fa8bc avcodec/mjpegdec: Support RGBA lpegs
Fixes Ticket896

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-25 00:32:44 +02:00
Martin Storsjö
54ba52077c arm: Comment out unused labels in simple_idct_arm
When building for iOS in thumb mode, gas-preprocessor.pl doesn't
mark unused labels as thumb functions (as it does for other
local labels, where it can figure out that they are functions
due to being referenced in branch instructions). This leads to
linker warnings for some of those local labels, such as:

ld: warning: ARM function not 4-byte aligned: __a_evaluation from
libavcodec/libavcodec.a(simple_idct_arm.o)

Therefore, comment them out since they don't have any function.
They do still have a value in documenting key points in the
assembly source though.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-24 22:43:21 +03:00
Michael Niedermayer
101c7ea906 avcodec/mpeg12dec: remove commented out assert
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-24 13:00:04 +02:00
Michael Niedermayer
800637709c alsdec: minor simplification in read_block()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-24 12:36:52 +02:00
Michael Niedermayer
7dec8512be Merge remote-tracking branch 'qatar/master'
* qatar/master:
  8bps: Make the bound-checks consistent

Conflicts:
	libavcodec/8bps.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-24 12:01:29 +02:00
Michael Niedermayer
da0f67b177 Merge commit 'bd7b4da0f4627bb6c4a7c2575da83fe6b261a21c'
* commit 'bd7b4da0f4627bb6c4a7c2575da83fe6b261a21c':
  8bps: Bound-check the input buffer

Conflicts:
	libavcodec/8bps.c

See: 66ff90f4a3
Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-24 11:45:54 +02:00
Michael Niedermayer
82654172ce Merge commit '2f034f255c49050e894ab9b88087c09ebe249f3f'
* commit '2f034f255c49050e894ab9b88087c09ebe249f3f':
  4xm: Reject not a multiple of 16 dimension

The newly added checks are redundant and thus replaced by asserts

See: db5b487551
Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-24 11:31:14 +02:00
Michael Niedermayer
a550926705 Merge commit 'ca488ad480360dfafcb5766f7bfbb567a0638979'
* commit 'ca488ad480360dfafcb5766f7bfbb567a0638979':
  alsdec: Clean up error paths

Conflicts:
	libavcodec/alsdec.c

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-24 10:31:10 +02:00
Michael Niedermayer
865f34823c Merge commit '70ecc175c7b513a153ac87d1c5d219556ca55070'
* commit '70ecc175c7b513a153ac87d1c5d219556ca55070':
  alsdec: Fix the clipping range

Conflicts:
	libavcodec/alsdec.c

See: feaff427c0

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-24 10:15:50 +02:00
Michael Niedermayer
274a50ab57 avcodec/utils: Warn the user about the lack of a lock manager if insufficient locks are detected
A lock manager is not the only possibility to avoid open/close locking
issues but its easier and more robust than maintaining a lot of lock/unlock
calls.

Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-24 01:11:09 +02:00
Luca Barbato
6fd221e5f8 8bps: Make the bound-checks consistent 2013-07-23 23:03:37 +02:00
Luca Barbato
bd7b4da0f4 8bps: Bound-check the input buffer
Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
2013-07-23 23:03:37 +02:00
Luca Barbato
2f034f255c 4xm: Reject not a multiple of 16 dimension
Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
2013-07-23 23:03:37 +02:00
Luca Barbato
ca488ad480 alsdec: Clean up error paths
Fix at least a memory leak.

CC: libav-stable@libav.org
2013-07-23 23:03:37 +02:00
Luca Barbato
70ecc175c7 alsdec: Fix the clipping range
mcc_weightings is only 32 elements.

Reported-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org
2013-07-23 23:03:37 +02:00
Michael Niedermayer
55db06af64 avcodec/utils: use av_assert0() to check validity of input pointers for start code search
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-23 21:23:09 +02:00
Michael Niedermayer
abc8110f7e avcodec: add avcodec_chroma_pos_to_enum()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-23 03:11:28 +02:00
Michael Niedermayer
b7397857e2 avcodec: add avcodec_enum_to_chroma_pos()
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-23 03:07:30 +02:00
clook
4443213d2e avcodec/libopenjpegen: XYZ 12 bit support for libopenjpeg encoder 2013-07-23 01:02:41 +02:00
Martin Storsjö
69e6702c01 arm: Mangle external symbols properly in new vfp assembly files
Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 14:48:30 +03:00
Martin Storsjö
47d57f24e3 arm: Mangle external symbols properly in new vfp assembly files
Reviewed-by: Kostya Shishkov
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 13:16:21 +02:00
Michael Niedermayer
23f250d2bc Merge remote-tracking branch 'qatar/master'
* qatar/master:
  arm: Add VFP-accelerated version of qmf_32_subbands

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 12:11:23 +02:00
Michael Niedermayer
366e415836 Merge commit '800ffab48a7844dd5dc0a33b8f6b8e5ed718cf2e'
* commit '800ffab48a7844dd5dc0a33b8f6b8e5ed718cf2e':
  dcadsp: Add a new method, qmf_32_subbands

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 12:10:59 +02:00
Michael Niedermayer
c30eb74d18 Merge commit '8b9eba664edaddf9a304d3acbf0388b5c520781d'
* commit '8b9eba664edaddf9a304d3acbf0388b5c520781d':
  arm: Add VFP-accelerated version of fft16

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 12:05:39 +02:00
Michael Niedermayer
bacba8e5d1 Merge commit 'ba6836c966debc56314ce2ef133c7f0c1fdfdeac'
* commit 'ba6836c966debc56314ce2ef133c7f0c1fdfdeac':
  arm: Add VFP-accelerated version of dca_lfe_fir

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 12:04:28 +02:00
Michael Niedermayer
2305a6775d Merge commit 'b63bb251ea6d6ba23295294e37a92625c0192206'
* commit 'b63bb251ea6d6ba23295294e37a92625c0192206':
  arm: Add VFP-accelerated version of imdct_half

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 11:57:05 +02:00
Michael Niedermayer
3cab228005 Merge commit 'd6e4f5fef0d811e180fd7541941e07dca9e11dc0'
* commit 'd6e4f5fef0d811e180fd7541941e07dca9e11dc0':
  arm: Add VFP-accelerated version of int32_to_float_fmul_array8

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 11:55:40 +02:00
Michael Niedermayer
1573274b91 Merge commit '26ffcc7de1d98f77400a2ebe4e75055515c54587'
* commit '26ffcc7de1d98f77400a2ebe4e75055515c54587':
  dcadec: Use int32_to_float_fmul_array8

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 11:48:43 +02:00
Michael Niedermayer
0d3400ec53 Merge commit '31c6f6f65c0ed5a894e26ce44ab0c3e89c82b9a2'
* commit '31c6f6f65c0ed5a894e26ce44ab0c3e89c82b9a2':
  fmtconvert: Add a new method, int32_to_float_fmul_array8

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 11:48:06 +02:00
Michael Niedermayer
877bbe05ce Merge commit 'ce9ed10ac27b9cf32a6257e083ea2f052692d971'
* commit 'ce9ed10ac27b9cf32a6257e083ea2f052692d971':
  arm: Add VFP-accelerated version of int32_to_float_fmul_scalar

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 11:42:19 +02:00
Michael Niedermayer
0129026c4e Merge commit '41ef1d360bac65032aa32f6b43ae137666507ae5'
* commit '41ef1d360bac65032aa32f6b43ae137666507ae5':
  arm: Add VFP-accelerated version of synth_filter_float

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 11:41:37 +02:00
Michael Niedermayer
b909428dab Merge commit '4719040cd1b54381796b22056a9ae37bd2b4668d'
* commit '4719040cd1b54381796b22056a9ae37bd2b4668d':
  libx264: Define X264_API_IMPORTS on MSVC/ICL

Merged-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 10:23:46 +02:00
Ben Avison
ff30d12159 arm: Add VFP-accelerated version of qmf_32_subbands
Before           After
               Mean    StdDev   Mean    StdDev  Change
This function   1323.0  98.0      746.2  60.6   +77.3%
Overall        15400.0 336.4    14147.5 288.4    +8.9%

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:44 +03:00
Ben Avison
800ffab48a dcadsp: Add a new method, qmf_32_subbands
This does most of the work formerly carried out by
the static function qmf_32_subbands() in dcadec.c.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:42 +03:00
Martin Storsjö
8b9eba664e arm: Add VFP-accelerated version of fft16
Before           After
               Mean    StdDev   Mean    StdDev  Change
This function   1389.3  4.2       967.8  35.1   +43.6%
Overall        15577.5 83.2     15400.0 336.4    +1.2%

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:41 +03:00
Martin Storsjö
ba6836c966 arm: Add VFP-accelerated version of dca_lfe_fir
Before           After
               Mean    StdDev   Mean    StdDev  Change
This function    868.2  33.5      436.0  27.0   +99.1%
Overall        15973.0 223.2    15577.5  83.2    +2.5%

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:39 +03:00
Martin Storsjö
b63bb251ea arm: Add VFP-accelerated version of imdct_half
Before           After
               Mean    StdDev   Mean    StdDev  Change
This function   2653.0  28.5     1108.8  51.4   +139.3%
Overall        17049.5 408.2    15973.0 223.2     +6.7%

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:37 +03:00
Ben Avison
d6e4f5fef0 arm: Add VFP-accelerated version of int32_to_float_fmul_array8
Before           After
               Mean    StdDev   Mean    StdDev  Change
This function    366.2  18.3      277.8  13.7   +31.9%
Overall        18420.5 489.1    17049.5 408.2    +8.0%

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:36 +03:00
Ben Avison
26ffcc7de1 dcadec: Use int32_to_float_fmul_array8
Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:35 +03:00
Ben Avison
31c6f6f65c fmtconvert: Add a new method, int32_to_float_fmul_array8
This is similar to int32_to_float_fmul_scalar, but
loads a new scalar multiplier every 8 input samples.
This enables the use of much larger input arrays, which
is important for pipelining on some CPUs (such as
ARMv6).

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:33 +03:00
Ben Avison
ce9ed10ac2 arm: Add VFP-accelerated version of int32_to_float_fmul_scalar
Before           After
               Mean    StdDev   Mean    StdDev  Change
This function   1175.0   4.4      366.2  18.3   +220.8%
Overall        19285.5 292.0    18420.5 489.1     +4.7%

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:30 +03:00