0dae193d3e
swr: remove another forgotten division in DSP function.
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-22 05:20:22 +02:00
cbf21628a5
swr: remove div/mod from DSP functions.
...
Also fix a bug with resample_compensation resetting dst_incr.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-18 14:15:52 +02:00
0608bc6502
swresample/audioconvert: fix () in FMT_PAIR_FUNC()
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-18 03:13:37 +02:00
edf930472b
swr: reindent.
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-16 01:33:32 +02:00
083cd3d1f7
swr: compile mmx2 s16p functions only on x86-32.
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-15 13:34:53 +02:00
7f4dfbd080
swr: add prototypes for resample dsp functions
...
Should fix compilation failures with MSVC and any other compiler
without inline asm support.
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-15 01:33:17 +02:00
ada8f9c046
swr: remove obsolete function prototypes.
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-15 00:07:25 +02:00
7128a35f8c
swr: split out DSP functions.
...
DSP bits of swri_resample go into their own mini-DSP functions; DSP
init goes from a per-call branch in multiple_resample to a proper
DSP init routine; x86 bits go into x86/; swri_resample() moves out of
resample_template.c into resample.c because it's independent of DSP
code or sample type; multiple_resample() is simplified.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-14 20:21:39 +02:00
4411928c64
swresample/resample: replace assert by av_assert
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-14 16:33:09 +02:00
b785c62681
swr: handle initial negative sample index outside DSP function.
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-14 14:36:18 +02:00
6b9685de3a
swr: remove unnecessary assignment.
...
I don't see dst_incr/dst_incr_frac ever being changed from their
initial value (which is the inverse of this operation), so it seems
to me that this is a no-op.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-14 04:30:53 +02:00
f341340552
swr: handle 64bit overflow check in multiple_resample().
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-09 15:24:51 +02:00
88f2586adb
fix various typos
...
Signed-off-by: Lou Logan <lou@lrcd.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-03 10:58:19 -08:00
cdfd9717ed
swr: move compensation_distance handling to swri_resample caller.
...
I think there's an off-by-one in terms of the switchpoint where we
switch from dst_incr to ideal_dst_incr, I don't think that's a massive
issue, but just be aware of that. It's probably trivial to prevent but
I don't care.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
I could not reproduce any off by 1 error, results are bit exact (michael)
2014-06-02 15:06:24 +02:00
2c23f87c85
swr/resample_template: prevent end_index from overflowing and add check for delta_frac overflow
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-02 00:57:08 +02:00
9b53853756
Rewrite main resampling loop (common and linear).
...
This removes a branch at a performance-sensitive point (in the middle
of the loop). In fate-swr-resample-s32p-8000-2626, this makes the code
about 10% faster. It also simplifies the loops, allowing us to rewrite
it in yasm at some later point.
The compensation_distance != 0 code and index < 0 code are still kind
of hairy. For compensation_distance != 0, this should likely be handled
in the caller, so that it calls swri_resample twice (once until the
dst_incr switch-point, and once with the remainder of the samples). For
index < 0, the code should probably be rewritten to break out of the
loop once sample_index >= 0, and then resume (e.g. as a tail-call) to
the common or linear resampling loops.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-06-02 00:47:54 +02:00
a9bf713d35
swresample: add swri_resample_float_avx
...
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-05-16 05:27:03 +02:00
96cb4c8718
swresample: swr_close()
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-05-15 18:27:23 +02:00
1898c2f49d
inline asm: fix arrays as named constraints.
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-05-07 15:02:45 +02:00
4cdea92976
swresample/resample: add missing xmm clobbers
...
Might fix fate-swr on ICL
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-05-07 01:32:40 +02:00
68c3e6025f
Fix convertion typos
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-29 00:09:08 +02:00
cdac3ab59f
swresample: add swri_resample_double_sse2
...
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-25 16:46:07 +02:00
291d464161
swresample: fix AV_CH_LAYOUT_STEREO_DOWNMIX input
...
Fixes Ticket 3542
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-24 01:25:46 +02:00
2b58c9c945
swresample/resample_template: try to consider src_size more exactly
...
This should avoid slight differences in the output causes by input
size alignment differences between archs
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-15 06:35:35 +02:00
5e379cd3ee
swresample/resample: simplify index/consumed calculation for the filter = 1 case
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-14 02:23:04 +02:00
6c8ee74af2
swresample/resample: Fix fractional part of index in the filter_size = 1 filters = 1 case
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-14 02:22:17 +02:00
5027f39712
swresample/resample: use av_malloc_array() where appropriate
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-08 00:29:26 +02:00
a5290cb1ac
swresample/dither: use av_malloc_array()
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-08 00:29:20 +02:00
f9158b01d0
swresample/resample: Limit filter length
...
Related to CID1197063
The limit choosen is arbitrary and much larger than what makes sense.
It avoids the need for checking arithmetic operations with the length for overflow
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-04-08 00:25:49 +02:00
63dbba655e
swresample/resample: sse float linear interpolation
...
About two times faster
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-24 02:34:02 +01:00
fa25c4c400
swresample/resample: mmx2/sse2 int16 linear interpolation
...
About three times faster
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-24 02:33:16 +01:00
32291ba6ea
swresample: add swri_resample_float_sse
...
At least two times faster than the C version.
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-20 06:01:06 +01:00
8236747511
Automatically change MANGLE() into named inline asm operands when direct symbol reference in inline asm are not supported.
...
This is part of the patch-set for intel C inline asm on windows support
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-18 23:39:30 +01:00
3d48cbc56c
swresample: reuse COMMON_CORE asm where possible
...
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-18 16:08:34 +01:00
7c8bf09edd
swresample: change COMMON_CORE_INT16 asm from SSSE3 to SSE2
...
pshuf+paddd is slightly faster than phaddd.
The real gain is in pre-ssse3 processors like AMD K8 and K10, which get
a big boost in performance compared to the mmxext version
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-03-18 15:00:50 +01:00
6c6e4dd139
swr: check that the context for swr_convert() has been initialized
...
Reviewed-by: ubitux
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-24 08:23:42 +01:00
a66be60888
swresample: add swr_is_initialized()
...
Idea-from/based-on: 7e86c27b4e
2014-02-24 08:23:22 +01:00
f284e2a58a
swresample: factorize clear_context() out
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-02-22 21:36:30 +01:00
e535897fad
Fix libswresample compilation with Apple Neon assembler.
...
Signed-off-by: Carl Eugen Hoyos <cehoyos@ag.or.at>
2014-02-17 17:40:10 +01:00
3dd04cbcf7
swresample: Add arm&x86 clobber tests
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-01-18 18:38:57 +01:00
cbeaf67888
Avoid using empty macro arguments.
...
These are not supported by all compilers (gcc 2.95 but also older SPARC
compilers, see gcc bug #33304 for example), and there is no real need for them.
One use of this feature remains in libavdevice/v4l2.c which can't be
replaced quite as easily.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
2013-12-31 12:19:59 +01:00
334e2e2363
lavu,lavc,lswr: do not hardcode AV_SAMPLE_FMT_NB value when setting sample format max value
...
The constant may change in libavutil but the library may be compiled
against an older version, thus rejecting a value which is otherwise
supported by the new libavutil.
INT_MAX is used here to denote the max allowed value for a sample format.
The opt-test code is changed to provide a valid reference example.
2013-12-26 11:35:27 +01:00
56572787ae
Add Windows resource file support for shared libraries
...
Originally written by James Almer <jamrial@gmail.com>
With the following contributions by Timothy Gu <timothygu99@gmail.com>
* Use descriptions of libraries from the pkg-config file generation function
* Use "FFmpeg Project" as CompanyName (suggested by Alexander Strasser)
* Use "FFmpeg" for ProductName as MSDN says "name of the product with which the
file is distributed" [1].
* Use FFmpeg's version (N-xxxxx-gxxxxxxx) for ProductVersion per MSDN [1].
* Only build the .rc files when --enable-small is not enabled.
[1] http://msdn.microsoft.com/en-us/library/windows/desktop/aa381058.aspx
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-12-05 23:42:07 +01:00
a6af5da7a2
swresample: use the internal buffer for resampling the last few samples
...
Fixes out of array read
Fixes Ticket3193
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-12-04 20:40:42 +01:00
4a640a6ac8
lswr: fix assert failure on unknown layouts.
2013-11-03 10:30:37 +01:00
904c89ac1b
lswr/swresample: convert ocl and icl options to AV_OPT_TYPE_CHANNEL_LAYOUT
...
Improve interface flexibility. In particular makes it possible to set
channel layout through symbolic names using the av_opt_set() interface.
2013-10-17 18:09:54 +02:00
ad75d2b590
x86: Fix compilation with nasm on PPC & OS/2
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-10-08 12:36:19 +02:00
a89e9ad1c8
swr: fix example code and doxy indent.
2013-09-29 23:37:46 +02:00
9078b40d35
swresample: replace 2 av_free() by av_freep()
...
avoids leaving stale pointers in memory
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-09-16 22:44:15 +02:00
c56d4dab03
swr/rematrix: Fix handling of AV_CH_LAYOUT_STEREO_DOWNMIX output
...
Fixes Ticket2859
Note, testcases related to the downmix channels are welcome.
(id like to make sure this is working correctly now, as obviously it didnt
work before ...)
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-08-20 16:56:51 +02:00
6dfffe9200
swr: clean layouts before checking sanity
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-08-20 16:49:31 +02:00
7ac12599fe
swresample: fix negative rematrix volumns
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 03:24:22 +02:00
e2b718464e
swresample: Make rematrix maxvalue user settable
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-07-22 03:23:54 +02:00
a9bbf59be7
cosmetics: Fix "dont" "wont" "doesnt" typos
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-29 02:31:16 +02:00
ca2818b881
swresample/x86/audio_convert: add emms to CONV
...
Might fix Ticket1874
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-18 02:26:36 +02:00
d62030ffca
swresample/rematrix_template: Fix integer overflow in mix8to2
...
Fixes: CID1030352
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-10 03:03:17 +02:00
b04bbe6b86
swresample/rematrix_template: Fix integer overflow in mix6to2
...
Fixes: CID1030351
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-10 03:03:12 +02:00
3289670142
swr: dont treat 32 and 24 as equal in simple copy check
...
Fixes dithering from 32 to 24 bit
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-05 03:06:34 +02:00
6bc4e36ba7
swr: set scale for 32->32/24 dither
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-05 03:06:34 +02:00
0804171104
swresample: set flags & description and add documentation for output_sample_bits
...
Allows users to actually use such option.
Signed-off-by: Paul B Mahol <onemda@gmail.com>
2013-06-04 21:57:25 +00:00
495bf1b685
swr: Fix dithering of s16 with SIMD
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-04 23:50:45 +02:00
4cfc92081d
swr: add native_simd_one
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-04 23:50:45 +02:00
b84de51f54
swresample/rematrix: add S32 mixing code
...
Needed for dithering S32
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-04 15:12:11 +02:00
f3c638eac1
wresample/rematrix_template: fix internal types
...
Should make no difference, but future commits will need it
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-06-04 15:12:11 +02:00
6e29f39252
swr/rematrix: use av_calloc()
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-04-15 20:31:06 +02:00
03148fd174
buildsys: only include log2_tab per library for shared builds
...
Fix linking failures with -all_load due to multiple log2_tabs
Signed-off-by: Carl Eugen Hoyos <cehoyos@ag.or.at>
2013-03-04 12:25:02 +01:00
dc666d360b
soxr: libsoxr 0.1.1 support
...
Libsoxr 0.1.1 will be out very soon; no changes planned beyond what's currently in git.
It includes a couple of fixes (not affecting FFmpeg's current usage) and a minor API change (but remains ABI compatible).
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-03-01 14:38:02 +01:00
35aaa306ac
swr: make the default of nopts for first_pts actually work
...
Fixes Ticket2210
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-02-25 04:39:26 +01:00
8b3affda87
swr: support a seperate output sample bits.
...
This avoids user apps having to mangle dither scale. for pcm24
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-02-24 20:31:56 +01:00
2abbe6d07a
swr: add duplicate cutoff for compatibility
...
Based on
commit 50f4337a2f
2013-02-24 20:31:56 +01:00
21cd905cd4
swr: check channel layouts before using them.
...
Fixes out of array accesses
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-02-13 17:50:24 +01:00
b5f544a0f9
swr: Dont use floats for S32->S32 when possible
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-02-04 04:11:46 +01:00
be989f9236
swr: reorder init code to make rematrix status available earlier
...
This also makes some other fields available earlier which may
be usefull one day
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-02-04 04:05:59 +01:00
24244cec2f
swr-test: Fix clip to 32bit
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-02-04 04:05:59 +01:00
b8c55590d5
swr/resample: fix integer overflow, add missing cast
...
The effects of this are limited to numeric errors in the output
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-02-04 04:05:59 +01:00
91debec976
swr/resample: fix filter rounding and cliping for s32
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-02-04 04:05:59 +01:00
f9abeecd94
swr/ build_filter: use av_calloc() fix buffer overflow
...
Fixes integer & buffer overflow
Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-27 05:58:47 +01:00
167af704ea
swr: limit phase_shift to a less insane value.
...
This prevents potential crashes and out of memory issues if
the user manually sets such insane values.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-27 05:54:52 +01:00
00cae86754
swr: support first_pts
...
Trolled-by: Daemon404
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-21 00:52:14 +01:00
cc4a41727e
swr: fix "may be used uninitialized in this function" warnings
...
The uninitialized use implicates 0 channels, which is unlikely
to reach this function but fixed anyway in this commit.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-20 17:09:48 +01:00
55d0528669
swr/dither: fix division by 0
...
Found-by: Daemon404
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-14 02:57:00 +01:00
d676598f87
swr: fix handling of timestamps that cause multiple drops or silence injections
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-13 18:39:06 +01:00
a34aee4646
swr: limit buffer size for silence injection
...
This reduces memory usage for unreasonable large silence injections
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-13 16:04:41 +01:00
b481d09bd9
swr: limit buffer size for discarding.
...
This prevents insane memory usage in case of insane input values.
Untested due to lack of a testcase that causes such insane allocation
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-13 16:04:41 +01:00
db4e0ecaf0
swr: move silence buffer to context to avoid per use malloc/free
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-13 15:28:00 +01:00
dc6588421e
swr: move buffer used to discard sample to context
...
This avoids the need to allocate & free to repeatly
Fixes Ticket2122
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-13 15:28:00 +01:00
c8737d348b
swr: work with 4 noise shaping coeffs at a time
...
63->38 kcycles
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-12 19:17:08 +01:00
93bc0f0180
swr/noise_shaping_data: pad coeffs to multiple of 4 when they are 1 below
...
This allows using code working with 4 at a time
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-12 19:09:54 +01:00
ef7fdc8cff
swr: use a local variable for ns_coeffs
...
maybe a very tiny bit faster
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-12 17:43:29 +01:00
68ff7d265f
swr: use local variable for ns_errors
...
71 -> 63 kcycles
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-12 17:43:29 +01:00
b5d9e5d06c
swr: minor simplification for the noise shaping pos update
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-12 17:43:29 +01:00
a74548521b
swr: check that allocation of in/out_convert succeeded
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-10 20:27:07 +01:00
e5a736261b
swr: Use a temporary buffer for dither/Noise shaping when the input is read only
...
This avoids copying the input
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-10 18:51:30 +01:00
2eec98125e
swr: mark noise data for noise shaping as const
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-10 18:07:32 +01:00
558aa6cab7
swr: add seperate in/out pointers to the noise shaping code
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-10 18:06:19 +01:00
fca51256d4
swr: rename dither_pos to noise_pos
...
THis more closely matches what the field represents
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-10 18:01:26 +01:00
2672b2c1d2
swr: scale data down in noise shaping to avoid cliping
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-10 01:44:15 +01:00
0b60fc94aa
swr: reorder the operations in the noise shaping code
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-10 00:27:04 +01:00
ead3a2a37e
swr: add noise_scale field to seperate it from the noise shaping scales
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-10 00:25:09 +01:00
6908ded20e
swr: fix the noise sample type for noise shaping
...
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2013-01-09 23:28:09 +01:00
Ronald S. Bultje
Michael Niedermayer
James Almer
Lou Logan
Matt Oliver
Reimar Döffinger
Martin Storsjö
Stefano Sabatini
Nicolas George
Clément Bœsch
Timothy Gu
Paul B Mahol
Rob Sykes