har0ke/FFmpeg
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
103,871 Commits 2 Branches 401 Tags
Commit Graph

670 Commits

Author SHA1 Message Date
Andreas Rheinhardt
044a7c08dc swscale/swscale: Disable x86-specific code for other arches 
SSE2 is x86 specific, yet due to the call to av_get_cpu_flags()
compilers were unable to optimize the checks (and the call) away
on other arches.

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
 2021-09-19 23:52:37 +02:00
Andreas Rheinhardt
f440c422b7 swscale/swscale: Fix races when using unaligned strides/data 
In this case the current code tries to warn once; to do so, it uses
ordinary static ints to store whether the warning has already been
emitted. This is both a data race (and therefore undefined behaviour)
as well as a race condition, because it is really possible for multiple
threads to be the one thread to emit the warning. This is actually
common since the introduction of the new multithreaded scaling API.

This commit fixes this by using atomic integers for the state;
furthermore, these are not static anymore, but rather contained
in the user-facing SwsContext (i.e. the parent SwsContext in case
of slice-threading).

Given that these atomic variables are not intended for synchronization
at all (but only for atomicity, i.e. only to output the warning once),
the atomic operations use memory_order_relaxed.

This affected the nv12, nv21, yuv420, yuv420p10, yuv422, yuv422p10 and
yuv444 filter-overlay FATE-tests.

Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
 2021-09-19 23:52:37 +02:00
James Almer
5fe648d04a libswscale/swscale: initialize all dst plane pointers in sws_receive_slice() 
Fixes valgrind warnings about use of uninitialised values.

Signed-off-by: James Almer <jamrial@gmail.com>
 2021-09-07 09:44:58 -03:00
Anton Khirnov
d6fdc78e91 sws: implement slice threading 2021-09-06 09:17:53 +02:00
Anton Khirnov
42cd64c182 sws: add a new scaling API 2021-09-06 09:16:52 +02:00
Anton Khirnov
37c0fe49b7 sws: move updating the palette higher up 
It does not interact in any way with the code setting up the image
pointers/strides, so it should not be intermixed with it.
 2021-07-03 16:13:40 +02:00
Anton Khirnov
d6649d9a3b sws: move initializing dither_error higher up 
It does not interact in any way with the code setting up the image
pointers/strides, so it should not be intermixed with it.
 2021-07-03 16:13:10 +02:00
Anton Khirnov
e188985598 sws: move the early return for zero-sized slices higher up 
Place it right after the input parameter validation. There is no point
in performing any setup if the sws_scale() call won't do anything.
 2021-07-03 16:09:43 +02:00
Anton Khirnov
a91e6c927e sws: simplify setting sliceDir 2021-07-03 16:09:21 +02:00
Anton Khirnov
ff753f41dd sws: merge handling frame start into a single block 
Also, return an error code on failure rather than 0.
 2021-07-03 16:09:07 +02:00
Anton Khirnov
1b11a324fe sws: make checking for the start of a new frame more explicit 2021-07-03 16:07:22 +02:00
Anton Khirnov
0fb014b7bb sws: reset sliceDir at the end of sws_scale() 
Makes it more clear that resetting it does not interact with the scaling
code that it is currently intermixed with.
 2021-07-03 16:05:39 +02:00
Anton Khirnov
1f80789bf7 sws: rename SwsContext.swscale to convert_unscaled 
That function pointer is now used only for unscaled conversion.
 2021-07-03 15:57:53 +02:00
Anton Khirnov
fe490ec165 sws: separate the calls to scaled vs unscaled conversion 
Call the scaler function directly rather than through a function
pointer. Drop the now-unused return value from ff_getSwsFunc() and
rename the function to reflect its new role.

This will be useful in the following commits, where it will become
important that the amount of output is different for scaled vs unscaled
case.
 2021-07-03 15:57:13 +02:00
Anton Khirnov
0f8e0957d2 sws: do not reallocate scratch buffers for each slice 2021-07-03 15:56:16 +02:00
Anton Khirnov
2730639259 sws: group the parameters validity checks together 
Also, fail with an error code rather than 0.
 2021-07-03 15:31:18 +02:00
Anton Khirnov
c05cab34a9 sws: initialize {src,dst}Stride2 consistently with {src,dst}2 2021-07-03 15:31:08 +02:00
Anton Khirnov
d3d8e09640 sws: cosmetics 
Reindent after previous commit, rewrap long lines.
 2021-07-03 15:30:56 +02:00
Anton Khirnov
f136493d03 sws: factor out cascaded scaling 2021-07-03 15:30:34 +02:00
Anton Khirnov
a2254aedc9 sws: cosmetics 
Reindent after previous commit, split long lines.
 2021-07-03 15:30:20 +02:00
Anton Khirnov
44f12718bf sws: factor out gamma-correct scaling 2021-07-03 15:29:50 +02:00
Anton Khirnov
e355af9be9 sws: return an error code on invalid parameters to sws_scale() 2021-07-03 15:29:35 +02:00
Anton Khirnov
21a4e48f88 sws: reindent after previous commit 2021-07-03 15:29:22 +02:00
Anton Khirnov
27acca1af0 sws: factor out updating the palette 2021-07-03 15:28:46 +02:00
Anton Khirnov
f8c21ccbfc sws: remove unnecessary braces 
There used to be more code inside them, but it was removed in
6de58b49032a206985602effec91e5e46c886ea2.
 2021-07-03 15:28:36 +02:00
Anton Khirnov
e15371061d lavu/mem: move the DECLARE_ALIGNED macro family to mem_internal on next+1 bump 
They are not properly namespaced and not intended for public use.
 2021-01-01 14:14:57 +01:00
Mark Reid
b4967fc71c libswscale: add output support for AV_PIX_FMT_GBRAPF32 
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2020-05-05 20:06:58 +02:00
Ruiling Song
4700f7d6fc swscale/swscale: remove useless code 
Signed-off-by: Ruiling Song <ruiling.song@intel.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2020-04-03 00:58:07 +02:00
Michael Niedermayer
a6ca22c118 swscale/swscale: Fix several invalid shifts related to vChrDrop 
Fixes: Invalid shifts
Fixes: #8166
Fixes: filter-crop_scale_vflip FATE-test

Reviewed-by: Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2020-01-22 18:41:46 +01:00
Limin Wang
cde1d70a39 swscale/swscale: cosmetics 
Signed-off-by: Limin Wang <lance.lmwang@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2019-09-27 10:58:30 +02:00
Limin Wang
29bde4b3b6 swscale/swscale: delete unwanted assignments 
Signed-off-by: Limin Wang <lance.lmwang@gmail.com>
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2019-09-09 18:16:06 +02:00
Martin Vignali
bdd6754648 swscale/swscale : small cosmetic 2018-08-22 11:36:15 +02:00
Martin Vignali
3af1c4ea7d swscale : treat float input data as uint 16bpc 
Currently float are converted to 16b uint in input part
using src depth (32 bits) in hScale16To19 and hScale16to15,
make an invalid shift for the data

So shift the value when using float input
like 16 bpc uint.
 2018-08-22 11:36:09 +02:00
Lou Logan
183fd30e0f Fix several typos 
"apix_fmts" found by Marc Péchaud.
"speedloss" found by Mikhail V.

Signed-off-by: Lou Logan <lou@lrcd.com>
 2017-09-21 16:17:02 -08:00
Michael Niedermayer
03ce71e4a1 swscale/swscale: Fix dereference of stride array before null check 
Fixes: CID1396263
Fixes: CID1396271

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2016-12-23 21:47:47 +01:00
Michael Niedermayer
d736b52a04 swscale: Drop is9_OR_10BPS() use, its name is not correct 
Found-by: Luca Barbato
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2016-11-10 00:33:12 +01:00
Pedro Arthur
8433d953e4 swscale: fix for sliced scaling artifacts 
Signed-off-by: Pedro Arthur <bygrandao@gmail.com>
 2016-09-16 11:25:28 -03:00
Michael Niedermayer
ba7be8c083 swscale: Fix "warning: ISO C90 forbids mixed declarations and code" 
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2016-09-07 01:44:16 +02:00
Michael Niedermayer
89ec11e096 swscale/swscale: Try to fix rgb48Toxyz12() with slices 
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2016-09-02 21:53:53 +02:00
Michael Niedermayer
f580719b1a swscale/swscale: Factor bottom to top handling 
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2016-09-02 21:53:53 +02:00
Ronald S. Bultje
70d418c7e6 Revert "PPC64: Add versions of functions in libswscale/input.c optimized for POWER8 VSX SIMD." 
This reverts commit 1df908f33f658979b32599489ca6f1a39821013c. The expected
performance improvements are essentially non-existent.
 2016-07-11 10:15:05 -04:00
Dan Parrot
1df908f33f PPC64: Add versions of functions in libswscale/input.c optimized for POWER8 VSX SIMD. 
This patch addresses Trac ticket #5570. The optimized functions are in file
libswscale/ppc/input_vsx.c. Each optimized function name is a concatenation of the
corresponding name in libswscale/input.c with suffix _vsx.

Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2016-06-30 12:02:08 +02:00
Pedro Arthur
e616e9a4b8 swscale: fix ring buffer size when scaling slices of a frame 
The ring buffer size should be able to store input lines
when there is not enough lines to output a single line.
 2016-06-10 09:45:24 -03:00
Matthieu Bouron
7abc8e7ae3 swscale/arm: add ff_hscale_8_to_15_neon 2016-04-09 00:16:56 +02:00
Pedro Arthur
6de58b4903 swscale: cleanup unused code 
Removed previous swscale code under '#ifndef NEW_FILTER'
and removed unused fields of SwsContext
 2016-03-31 16:36:16 -03:00
Clément Bœsch
263eb76bdf sws/aarch64: add ff_hscale_8_to_15_neon 
./ffmpeg -nostats -f lavfi -i testsrc2=4k:d=2 -vf bench=start,scale=1024x1024,bench=stop -f null -

    before: t:0.489726 avg:0.489883 max:0.491852 min:0.489482
    after:  t:0.256515 avg:0.256458 max:0.256999 min:0.253755
 2016-03-31 10:12:55 +02:00
Derek Buitenhuis
21f9468402 avutil: Rename FF_CEIL_COMPAT to AV_CEIL_COMPAT 
Libav, for some reason, merged this as a public API function. This will
aid in future merges.

A define is left for backwards compat, just in case some person
used it, since it is in a public header.

Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
 2016-01-27 16:36:46 +00:00
Michael Niedermayer
321e85e176 swscale/swscale: Add some sanity checks for srcSlice* parameters 
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
 2016-01-17 20:35:36 +01:00
Pedro Arthur
5bd62a1b3c swscale: fix ticket #4881 
When scaling only a slice of a frame the output was written always
in the first lines leaving the rest of the frame black.
 2015-10-13 13:43:39 -03:00
Pedro Arthur
a8602dde5e swscale: fix ticket #4877 2015-09-25 12:18:58 -03:00