1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-11-26 19:01:44 +02:00
Commit Graph

73 Commits

Author SHA1 Message Date
Ronald S. Bultje
9e66b892e8 swscale: add missing colons to x86 assembly yuv2planeX.
This fixes assembling using "nasm".
2011-10-23 09:44:03 -07:00
Ronald S. Bultje
6cacecdca3 swscale: make yuv2yuvX_10_sse2/avx 8/9/16-bits aware.
Also implement MMX/MMX2 versions and SSE4 versions.
2011-10-22 10:35:14 -07:00
Kieran Kunhya
7fbbf95293 yuv2planeX10 SIMD
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-10-22 10:35:14 -07:00
Kieran Kunhya
34e8d147b3 Split out yuv2yuv1 luma and chroma in order to make them generic DSP functions
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-10-22 10:35:13 -07:00
Ronald S. Bultje
6aa3cac6bf swscale: use aligned move for storage into temporary buffer.
The intermediate buffer is always aligned.
2011-10-11 07:50:48 -07:00
Ronald S. Bultje
e0c3e07387 sws: implement MMX/SSE2/SSSE3/SSE4 versions for horizontal scaling.
Speed: from 3.9x to 9.6x speed improvement over C, and some small
(up to 15%) speed improvements over existing MMX code (particularly
for bigger filters).
2011-09-13 09:53:42 -07:00
Ronald S. Bultje
3f04ab4fcd swscale: split hScale() function pointer into h[cy]Scale().
This allows using more specific implementations for chroma/luma, e.g.
we can make assumptions on filterSize being constant, thus avoiding
that test at runtime.
2011-08-17 20:56:06 -07:00
Ronald S. Bultje
28c1115a91 swscale: use 15-bit intermediates for 9/10-bit scaling. 2011-08-12 11:54:25 -07:00
Ronald S. Bultje
5c391a161a swscale: rename uv_off/uv_off2 to uv_off_px/byte. 2011-07-08 15:15:31 -07:00
Ronald S. Bultje
4e3e333a79 swscale: error dithering for 16/9/10-bit to 8-bit.
Based on a somewhat similar idea in FFmpeg's swscale copy.
2011-07-08 15:15:30 -07:00
Ronald S. Bultje
42d622fab3 swscale: fix 16-bit scaling when output is 8-bits.
We would use the second half of the U plane buffer, rather than the
V plane buffer, to output the V plane pixels.
2011-07-08 15:15:30 -07:00
Ronald S. Bultje
8a8d0ce208 swscale: for >8bit scaling, read in native bit-depth.
For 9/10bit, it means we don't have to upscale to 16bit before
actual scaling or pixel format conversion, and thus a performance
gain.
2011-07-01 09:08:26 -07:00
Ronald S. Bultje
ef1ee362b3 swscale: implement >8bit scaling support.
This means that precision is retained when scaling between sample
formats with >8 bits per component (48bit RGB, 16bit grayscale,
9/10/16bit YUV).
2011-06-29 09:45:52 -07:00
Ronald S. Bultje
13a099799e swscale: change prototypes of scaled YUV output functions.
Remove unused variables "flags" and "dstFormat" in yuv2packed1,
merge source rows per plane for yuv2packed[12], and make every
source argument int16_t (some where invalidly set to uint16_t).
This prevents stack pollution and is part of the Great Evil Plan
to simplify swscale.
2011-06-27 21:24:44 -07:00
Ronald S. Bultje
0d994b2f45 swscale: don't use planar output functions to write to NV12/21.
This prevents a crash when converting to NV12/21 without the bitexact
flags enabled.
2011-06-27 17:15:41 -07:00
Ronald S. Bultje
97535ffb97 swscale: remove unused xInc/srcW arguments from hScale(). 2011-06-26 14:41:44 -07:00
Diego Biurrun
a60466dbc3 swscale: Remove HAVE_MMX from files that are only compiled with MMX enabled. 2011-06-15 01:18:10 +02:00
Diego Biurrun
97e057ff81 swscale: Fix compilation with --disable-mmx2.
Some MMX2 functions were being referenced without proper #ifdefs.
2011-06-15 01:17:59 +02:00
Ronald S. Bultje
a27db4c349 swscale: fix function declaration keywords in x86/swscale_template.c.
Remove inline keyword for functions that are only called through
their function pointers (and thus cannot be inlined); add av_cold
keyword to init function, and use av_always_inline instead of
inline for functions that must be inlined for performance reasons.
2011-06-09 09:25:59 -04:00
Ronald S. Bultje
9bcbb250e2 swscale: fix types of assembly arguments.
This prevents the following compiler warnings: "warning:
initialization from incompatible pointer type". Since the
variables are only ever used in inline assembly, their type
is actually irrelevant (so the part where it was wrong did
not invoke any buggy behaviour).
2011-06-08 11:39:26 -04:00
Ronald S. Bultje
496d95c34c swscale: move two macros that are only used once into caller.
This way, they look like regular code, which is easier to
understand.
2011-06-08 11:33:46 -04:00
Ronald S. Bultje
0fb5193156 swscale: reindent x86 init code. 2011-06-07 10:05:50 -04:00
Ronald S. Bultje
ca364a5b43 swscale: extract SWS_FULL_CHR_H_INT conditional into init code. 2011-06-07 10:03:47 -04:00
Ronald S. Bultje
bda9b20fa4 swscale: un-special-case yuv2yuvX16_c().
Make yuv2yuvX16_c a function pointer for yuv2yuvX(), so that the
function pointer becomes bitdepth-independent.
2011-06-07 10:03:12 -04:00
Ronald S. Bultje
983260b0a4 swscale: split out x86/swscale_template.c from swscale.c. 2011-06-03 09:52:09 -07:00
Anton Khirnov
b8e893399f sws: replace all long with int.
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-05-28 10:03:37 -04:00
Ronald S. Bultje
009f829dde swscale: fix crash in bilinear scaling. 2011-05-28 08:59:56 -04:00
Diego Biurrun
5a5a0f1613 swscale: Remove unused variables in x86 code.
libswscale/x86/swscale_template.c:2072: warning: unused variable ‘canMMX2BeUsed’
libswscale/x86/swscale_template.c:2145: warning: unused variable ‘canMMX2BeUsed’
libswscale/x86/swscale_template.c:2209: warning: unused variable ‘chrVPixBuf’
libswscale/x86/swscale_template.c:2237: warning: unused variable ‘chrVSrcPtr’
2011-05-27 21:25:57 +02:00
Ronald S. Bultje
f327bfa6dc swscale: fix build with --disable-swscale-alpha. 2011-05-27 13:22:27 -04:00
Ronald S. Bultje
9f5d45025e swscale: fix non-bitexact yuv2yuv[X2]() MMX/MMX2 functions. 2011-05-27 09:41:48 -04:00
Ronald S. Bultje
e758573a88 swscale: fix compile on x86-32. 2011-05-26 10:41:27 -04:00
Ronald S. Bultje
b4a224c5e4 swscale: split chroma buffers into separate U/V planes.
Preparatory step to implement support for sizes > VOFW.
2011-05-26 09:31:27 -04:00
Ronald S. Bultje
522d65ba25 rgb2rgb: remove duplicate mmx/mmx2/3dnow/sse2 functions.
Many functions have such a prefix, but do not actually use any
instructions or features from that set, thus giving the false
impression that swscale is highly optimized for a particular
system, whereas in reality it is not.
2011-05-26 09:31:02 -04:00
Ronald S. Bultje
836b82e3c9 swscale: reindent h[cy]scale_fast() and updateDitherTables(). 2011-05-26 09:31:02 -04:00
Ronald S. Bultje
e2bad983b5 swscale: reformat x86/swscale_template.c.
Interleave macros and code so that it's easier to find the
actual code that belongs to a function. Also reindent where
appropriate and remove dead code.
2011-05-26 09:30:15 -04:00
Ronald S. Bultje
71d9c33c86 swscale: remove duplicate mmx/mmx2 functions if they are identical. 2011-05-26 09:30:15 -04:00
Ronald S. Bultje
acb96bc268 swscale: remove if (c->dstFormat) branch from yuv2packed[12X]().
This allows cutting up the function in much smaller and easier-
to-maintain chunks.
2011-05-26 09:30:14 -04:00
Ronald S. Bultje
b6cac9b3bf swscale: remove if(full_chr_int) from yuv2packed1().
If that flag is set, swScale() already proxies the call to
yuv2rgbXinC_full(). Therefore, this flag is never set when
yuv2packed1() is called.
2011-05-26 09:30:14 -04:00
Ronald S. Bultje
2159a24573 swscale: remove if(accurate_rnd) branch from functions. 2011-05-26 09:30:14 -04:00
Diego Biurrun
0093199184 swscale: Remove commented-out printf cruft. 2011-05-26 03:45:58 +02:00
Michael Niedermayer
7dc303a60e swscale: Eliminate rgb24toyv12_c() duplication.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Signed-off-by: Diego Biurrun <diego@biurrun.de>
2011-05-26 00:56:06 +02:00
Ronald S. Bultje
1bb0f0c925 swscale: remove if(bitexact) branch from functions.
Instead, only set the function pointers if bitexact flag is
not set during initialization. Since a change in flags triggers
a re-init anyway, this doesn't situations where flag values
change during runtime.
2011-05-24 14:50:20 -04:00
Ronald S. Bultje
566b5fbbb3 swscale: remove if(canMMX2BeUsed) conditional.
Instead, set function pointers conditionally during init. This
patch also reveals a whole branch of dead assembly code that is
therefore also removed.
2011-05-24 14:50:19 -04:00
Ronald S. Bultje
aaca69c130 swscale: remove swScale_{c,MMX,MMX2} duplication. 2011-05-24 14:50:15 -04:00
Ronald S. Bultje
c4fd283a46 swscale: use emms_c(). 2011-05-24 14:48:49 -04:00
Ronald S. Bultje
b9eb2136af swscale: remove dead macro WRITEBGR24OLD. 2011-05-24 10:46:40 -04:00
Ronald S. Bultje
1dd4f4be5a swscale: remove AMD3DNOW "optimizations".
The functions are identical to their MMX counterparts. Thus,
pretending that swscale is highly optimized for AMD3DNOW
extensions is a poorly executed practical joke at best.
2011-05-24 10:19:41 -04:00
Ronald S. Bultje
11ffefefdb swscale: remove duplicated x86/ functions. 2011-05-24 10:11:26 -04:00
Ronald S. Bultje
e66149e714 swscale: force --enable-runtime-cpudetect and remove SWS_CPU_CAPS_*. 2011-05-24 10:03:26 -04:00
Ronald S. Bultje
c8f487deae swscale: fix YUV420P 9/10bit support.
Fix handling of input if not in native endianness, and add support for
9/10-bit output. This allows us to force endianness of YUV420P 9/10bit
in the H264/10bit fate tests, which should fix them on big-endian
systems.
2011-05-11 19:15:14 -04:00