Commit Graph

100 Commits

Author SHA1 Message Date
Måns Rullgård
5813e05d08 ARM: NEON optimised H.264 8x8 and 16x16 qpel MC
Originally committed as revision 16149 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-12-15 22:12:47 +00:00
Måns Rullgård
ad74a0f8b3 ARM: NEON optimised H.264 loop filter
Originally committed as revision 16148 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-12-15 22:12:44 +00:00
Måns Rullgård
1cce897aad ARM: NEON optimised {put,avg}_h264_chroma_mc[48]
Originally committed as revision 16147 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-12-15 22:12:41 +00:00
Måns Rullgård
7eec43289a ARM: NEON optimised simple_idct
Originally committed as revision 16146 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-12-15 22:12:37 +00:00
Måns Rullgård
569f5a756a ARM: NEON optimised put_pixels functions
Originally committed as revision 16145 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-12-15 22:12:32 +00:00
Måns Rullgård
d8b66635e0 ARM: remove some useless comments
Originally committed as revision 16014 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-12-06 13:42:00 +00:00
Måns Rullgård
b0e8ce55ae ARM: convert VFP code to UAL syntax
Originally committed as revision 15994 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-12-03 20:16:01 +00:00
Diego Biurrun
9686df2be5 Delete unnecessary 'extern' keywords.
Originally committed as revision 15990 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-12-03 15:23:30 +00:00
Måns Rullgård
4deaa94639 Add shift argument to MULL() macro
This replaces use of FRAC_BITS in the MULL() definition with a third
argument specifying the shift amount.  All uses of this macro are
updated to pass FRAC_BITS as third argument.

Originally committed as revision 15921 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-24 09:28:55 +00:00
Måns Rullgård
d270cb4ae4 ARM: move dct_unquantize_h263_*_armv5te asm to separate file
Originally committed as revision 15917 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-23 19:11:44 +00:00
Måns Rullgård
77753f874a ARM: fix j_rev_dct_ARM
This is a bugfix for ARMv4 assembly implementation of 'j_rev_dct'
function.

The problem was in the incorrect partially empty row detection. Even
if the first two coefficients in the row were nonzero, it handled this
just like the case with only the first nonzero coefficient.

Now this function produces exactly the same output as the stripped
down reference C version of 'j_rev_dct' (with the nested checks like
'if (d6) { if (d2) { ...' always evaluated as true, avoiding shortcut
branches).

Originally committed as revision 15805 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-12 20:23:36 +00:00
Måns Rullgård
d00e8b83f1 ARM: change MUL16() macro to inline function
Originally committed as revision 15783 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-06 01:33:28 +00:00
Måns Rullgård
b911711571 ARM: prettify MAC16() macro
Originally committed as revision 15782 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-06 01:33:26 +00:00
Måns Rullgård
73c1c96b83 ARM: change MULH() macro to inline function
Originally committed as revision 15781 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-06 01:33:23 +00:00
Måns Rullgård
8f4d0d48b9 ARM: change MULL() macro to inline function
Originally committed as revision 15780 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-06 01:33:20 +00:00
Måns Rullgård
96dea17a1d ARM: mathops.h whitespace cosmetics
Originally committed as revision 15779 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-06 01:33:17 +00:00
Diego Biurrun
9f879620be Fix illegal identifier, __ are reserved for the system.
Originally committed as revision 15771 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-03 21:59:35 +00:00
Dominik Mierzejewski
82d1605fe7 Remove duplicated MM_* macros for CPU capabilities from dsputil.h.
Add missing one for FF_MM_ALTIVEC to avcodec.h.
Rename all the occurences of MM_* to the corresponding FF_MM_*.

Originally committed as revision 15770 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-11-03 18:08:00 +00:00
Diego Biurrun
01f54021dc Add missing headers to pass 'make checkheaders'.
Originally committed as revision 15745 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-10-29 23:43:55 +00:00
Diego Pettenò
782fc0c36f Rename template included sources from .h to _template.c.
There are multiple source files that are #include'd rather than
compiled, as they are used as template for generation of similar code,
like asm-optimised code. Some of these files are right now named with
a .h extension, although they are not header in any reasonable sense.

Rename them so that instead of being named with .h extension they are
named with _template.c as final part.

Originally committed as revision 15730 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-10-27 14:35:58 +00:00
Måns Rullgård
d2d398590d ARM: move add_pixels_clamped_ARM() to dsputil_arm_s.S
Originally committed as revision 15728 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-10-27 00:25:19 +00:00
Måns Rullgård
83ad74e708 ARM: move VFP DSP functions to dsputils_vfp.S
Originally committed as revision 15727 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-10-27 00:25:16 +00:00
Måns Rullgård
a648516b95 ARM: move prefetch_arm to dsputil_arm_s.S
Originally committed as revision 15726 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-10-27 00:25:12 +00:00
Måns Rullgård
c130bedc99 ARM: use new macros for assembler function labels
Originally committed as revision 15725 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-10-27 00:25:09 +00:00
Måns Rullgård
c72ab30188 ARM: add some assembler convenience macros
Originally committed as revision 15724 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-10-27 00:25:06 +00:00
Diego Pettenò
be449fca79 Convert asm keyword into __asm__.
Neither the asm() nor the __asm__() keyword is part of the C99
standard, but while GCC accepts the former in C89 syntax, it is not
accepted in C99 unless GNU extensions are turned on (with -fasm). The
latter form is accepted in any syntax as an extension (without
requiring further command-line options).

Sun Studio C99 compiler also does not accept asm() while accepting
__asm__(), albeit reporting warnings that it's not valid C99 syntax.

Originally committed as revision 15627 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-10-16 13:34:09 +00:00
Stefano Sabatini
987903826b Globally rename the header inclusion guard names.
Consistently apply this rule: the guard name is obtained from the
filename by stripping the leading "lib", converting '/' and '.'  to
'_' and uppercasing the resulting name. Guard names in the root
directory have to be prefixed by "FFMPEG_".

Originally committed as revision 15120 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-31 07:39:47 +00:00
Måns Rullgård
94cdcece91 ARM: add prefetch function using ARMv5 PLD instruction
Originally committed as revision 14962 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-08-25 02:29:04 +00:00
Måns Rullgård
b0367ef00a ARM: optimised MAC64 and MLS64
Originally committed as revision 14245 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-07-15 19:06:32 +00:00
Måns Rullgård
6865e04620 ARM: optimised MUL64
Originally committed as revision 14244 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-07-15 19:06:29 +00:00
Måns Rullgård
7995962d4b ARM: ARMv6 optimised MULH
Originally committed as revision 14243 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-07-15 19:06:25 +00:00
Benoit Fouet
0abf0c5398 Simplify mm_support() for ARM (by Måns).
Originally committed as revision 14033 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-30 11:16:47 +00:00
Matthieu Castet
37dcd091da Define mm_support() for ARM.
Patch by Matthieu Castet <castet matthieu free fr>

Originally committed as revision 14032 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-30 07:31:41 +00:00
Guillaume Poirier
28215b3700 convert DOS-style carriage return to Unix-style
Originally committed as revision 13799 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-06-17 14:42:51 +00:00
Siarhei Siamashka
e6303192c7 Add some initial optimizations for ARM VFP (floating
point coprocessor available in some ARM11 cores).
Patch by Siarhei Siamashka % siarhei.siamashka@gmail.com %
Original thread:
date: Apr 20, 2008 5:41 PM
subject: [FFmpeg-devel] [PATCH] Some ARM VFP optimizations (vector_fmul, vector_fmul_reverse, float_to_int16)

Originally committed as revision 13128 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-05-11 20:06:03 +00:00
Diego Biurrun
245976da2a Use full path for #includes from another directory.
Originally committed as revision 13098 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-05-09 11:56:36 +00:00
Diego Biurrun
a1fe4604d4 Revert previous mistaken commit.
Originally committed as revision 13094 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-05-09 10:18:58 +00:00
Diego Biurrun
81f46d5ac9 ipp.h is a system header, use <> when #including it.
Originally committed as revision 13093 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-05-09 10:06:51 +00:00
Diego Biurrun
c88c253d8b cosmetics: __asm__ __volatile__ --> asm volatile
Originally committed as revision 12885 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-04-17 21:57:52 +00:00
Matthieu Castet
b4e806b2b9 pld instruction are used in libavcodec/armv4l/dsputil_arm_s.S which can be used
on armv4 that doesn't support this instruction. Futhermore pld is a nop on some
armv5 processor like arm926. Detect if pld is supported and have the
preprocessor remove it when it's not supported.
Fixes issue 393.
patch by matthieu castet, castet.matthieu free fr

Originally committed as revision 12569 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-03-24 11:49:59 +00:00
Reimar Döffinger
78d3d94f14 __asm __volatile -> asm volatile, improves code consistency and works
(as far as that is possible) with the Sun C compiler.

Originally committed as revision 12188 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-02-24 14:46:22 +00:00
Guillaume Poirier
d761f08987 fix indentation that was messed up by r11628
Originally committed as revision 11629 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-01-27 08:37:28 +00:00
Siarhei Siamashka
13686bc2cd fix lowres decoding support on ARM CPUs
Original thread:
date: Jan 27, 2008 12:03 AM
subject: [FFmpeg-devel] [PATCH] Fix for lowres decoding support on ARM
Patch by Siarhei Siamashka $ siarhei P siamashka A gmail P com $

Originally committed as revision 11628 to svn://svn.ffmpeg.org/ffmpeg/trunk
2008-01-27 08:36:50 +00:00
Diego Biurrun
22170ec0b2 Remove mistakenly added multiple inclusion guards.
Originally committed as revision 11332 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-12-28 01:42:58 +00:00
Diego Biurrun
56cc85a068 Misc spelling fixes, prefer American over British English.
Originally committed as revision 11126 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-12-02 15:43:08 +00:00
Vitor Sessak
52b541ad79 spelling
Originally committed as revision 11122 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-12-01 22:21:04 +00:00
Diego Biurrun
5b21bdabe4 Add FFMPEG_ prefix to all multiple inclusion guards.
Originally committed as revision 10765 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-10-17 09:37:46 +00:00
Diego Biurrun
31b2c1446f Add missing multiple inclusion guards.
Originally committed as revision 10763 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-10-17 09:31:06 +00:00
Tao Huang
005b683744 Fix put_*_xy2_arm bug and enable put_pixels16_arm and put_pixels8_y2_arm.
Speed up > 4%.
patch by Tao Huang, ulysses.huang gmail com

Originally committed as revision 10402 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-09-05 11:05:33 +00:00
Guillaume Poirier
14c2634bcc Fix links to ARM documentation
Originally committed as revision 10399 to svn://svn.ffmpeg.org/ffmpeg/trunk
2007-09-05 07:49:01 +00:00