Commit Graph

741 Commits

Author SHA1 Message Date
Diego Biurrun
a3cb865310 x86: dsputil: Move some ifdefs to avoid unused variable warnings 2013-04-12 09:36:47 +02:00
Diego Biurrun
2004c7c8f7 x86: dsputil: cosmetics: Remove two pointless variable indirections 2013-04-12 09:36:47 +02:00
Diego Biurrun
c51a3a5bd9 x86: dsputil: Refactor some ff_{avg|put}_pixels function declarations 2013-04-12 09:36:46 +02:00
Diego Biurrun
e027032fc6 x86: dsputil: ff_h263_*_loop_filter declarations to a more suitable place 2013-04-12 09:36:46 +02:00
Diego Biurrun
a89c05500f x86: h264qpel: int --> ptrdiff_t for some line_size parameters 2013-04-12 09:30:12 +02:00
Diego Biurrun
ac9362c5d9 Move misplaced file author information where it belongs 2013-04-11 02:42:11 +02:00
Ronald S. Bultje
b93b27edb0 dsputil: Make dsputil selectable
Signed-off-by: Martin Storsjö <martin@martin.st>
2013-04-10 11:04:05 +03:00
Ronald S. Bultje
62844c3fd6 h264: Integrate clear_blocks calls with IDCT
The non-intra-pcm branch in hl_decode_mb (simple, 8bpp) goes from 700
to 672 cycles, and the complete loop of decode_mb_cabac and hl_decode_mb
(in the decode_slice loop) goes from 1759 to 1733 cycles on the clip
tested (cathedral), i.e. almost 30 cycles per mb faster.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-04-10 11:03:06 +03:00
Ronald S. Bultje
610b18e2e3 x86: qpel: Move fullpel and l2 functions to a separate file
This way, they can be shared between mpeg4qpel and h264qpel without
requiring either one to be compiled unconditionally.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-04-08 12:38:33 +03:00
Christophe Gisquet
f4b0d12f5b x86: sbrdsp: Implement SSE neg_odd_64
Timing on Arrandale:
        C   SSE
Win32:  57   44
Win64:  47   38
Unrolling and not storing mask both save some cycles.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
2013-04-05 22:47:04 +02:00
Diego Biurrun
b6649ab503 cosmetics: Remove unnecessary extern keywords from function declarations 2013-03-27 14:21:45 +01:00
Martin Storsjö
a2acadd058 x86: vc1dsp: Fix indentation
Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-26 15:49:42 +02:00
Janne Grunau
e5c2794a71 x86: consistently use unaligned movs in the unaligned bswap
Fixes fate errors in asv1, ffvhuff and huffyuv on x86_32.
2013-03-25 12:11:11 +01:00
Martin Storsjö
285ff14413 x86: Change a missed occurrance of int to ptrdiff_t for strides
Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-24 12:06:53 +02:00
Martin Storsjö
352dbdb96c x86: Remove win64 xmm clobbering wrappers for the now removed avcodec_encode_video function
Signed-off-by: Martin Storsjö <martin@martin.st>
2013-03-23 23:37:27 +02:00
Luca Barbato
a8b6015823 dsputil: convert remaining functions to use ptrdiff_t strides
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2013-03-12 18:26:42 +01:00
Diego Biurrun
e8c52271c4 Revert "Move H264/QPEL specific asm from dsputil.asm to h264_qpel_*.asm."
This reverts commit f90ff772e7.

The code should be put back in h264_qpel_8bit.asm, but unfortunately
it is unconditionally used from dsputil_mmx.c since 71155d7.
2013-02-28 21:50:02 +01:00
Diego Biurrun
ebc701993f x86: dsputil: Drop some unused function #defines 2013-02-26 23:36:24 +01:00
Diego Biurrun
845cfc92f9 x86: dsputil: Drop aliasing of ff_put_pixels8_mmx to ff_put_pixels8_mmxext
The external assembly function uses mmxext instructions and should not be
masqueraded as an mmx-only function. Instead, use the mmx-only inline
assembly function.
2013-02-26 23:36:24 +01:00
Diego Biurrun
096cc11ec1 x86: vc1dsp: Move ff_avg_vc1_mspel_mc00_mmxext out of dsputil_mmx.c 2013-02-26 23:36:24 +01:00
Martin Storsjö
31a23a0dc6 x86: dsputil_mmx: Remove leftover inline assembly fragments
These became unused in 71155d7b.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-02-27 00:17:05 +02:00
Diego Biurrun
c242bbd8b6 Remove unnecessary dsputil.h #includes 2013-02-26 00:51:34 +01:00
Matt Wolenetz
311443f6c7 x86: h264: Don't use redzone in AVX h264_deblock on Win64
This fixes crashes in chromium on win64 on machines with AVX
(crashes that apparently aren't triggered by fate).

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-02-21 15:02:16 +02:00
Ronald S. Bultje
e5ffffe48d h264chroma: Remove duplicate 9/10 bit functions
These functions do the same thing in 16 bit space and don't need
any depth specific clipping.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-02-19 22:33:19 +02:00
Daniel Kang
9acd23d655 x86: dsputil: Fix h263 loop filter link error in some configurations
This was caused by unconditionally referencing a conditionally compiled
table. Now the code is also compiled conditionally.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
2013-02-18 17:09:00 +01:00
Daniel Kang
7a03145ed7 x86: dsputil: int --> ptrdiff_t for ff_put_pixels16_mmxext line_size param
This avoids SIMD-optimized functions having to sign-extend their
line size argument manually to be able to do pointer arithmetic.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
2013-02-18 15:23:03 +01:00
Daniel Kang
b3f2a3fe3f x86: mpeg4qpel: Make movsxifnidn do the right thing
Fixes an instruction that does nothing by changing the
source to dword.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
2013-02-11 20:17:15 +01:00
Diego Biurrun
5d3d39c72e dsputil: Move fdct function declarations to dct.h 2013-02-09 00:08:28 +01:00
Diego Biurrun
218aefce44 dsputil: Move LOCAL_ALIGNED macros to libavutil 2013-02-08 23:13:37 +01:00
Daniel Kang
a1d3673034 dsputil: x86: Fix compile error
Accidentally prefixed ff_ with cextern.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-02-07 11:06:16 +02:00
Daniel Kang
659d4ba5af dsputil: x86: Convert h263 loop filter to yasm
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2013-02-06 15:38:27 -08:00
Martin Storsjö
a846dccb29 h264chroma: x86: Fix building with yasm disabled
Signed-off-by: Martin Storsjö <martin@martin.st>
2013-02-06 17:05:33 +02:00
Diego Biurrun
82bd04b170 rv34: Drop now unnecessary dsputil dependencies 2013-02-06 11:30:54 +01:00
Diego Biurrun
79dad2a932 dsputil: Separate h264chroma 2013-02-06 11:30:53 +01:00
Diego Biurrun
c9f933b5b6 Add av_cold attributes to arch-specific init functions 2013-02-05 17:01:05 +01:00
Diego Biurrun
25841dfe80 Use ptrdiff_t instead of int for {avg, put}_pixels line_size parameter.
This avoids SIMD-optimized functions having to sign-extend their
line size argument manually to be able to do pointer arithmetic.
2013-02-05 12:59:12 +01:00
Diego Biurrun
52acd79165 x86: hpel: Move {avg,put}_pixels16_sse2 to hpeldsp 2013-01-31 11:19:23 +01:00
Diego Biurrun
c59211b437 x86: Simplify some arch conditionals 2013-01-29 00:10:53 +01:00
Michael Niedermayer
834e9fb056 x86: hpeldsp: Fix a typo, use the right register
This makes the code actually work.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-01-28 12:49:37 +02:00
Daniel Kang
05b0998f51 dsputil: Fix error by not using redzone and register name
Signed-off-by: Derek Buitenhuis <derek.buitenhuis@gmail.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2013-01-28 07:23:20 +01:00
Daniel Kang
96753bd00d dsputil: x86: Correct the number of registers used in put_no_rnd_pixels16_l2
put_no_rnd_pixels16_l2 allocated 5 instead of 6 registers.

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2013-01-27 08:41:48 +01:00
Daniel Kang
0eedf5d74d dsputil: add missing HAVE_YASM guard
Fix compile error under
"--disable-optimizations --disable-yasm --disable-inline-asm"

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2013-01-27 08:41:46 +01:00
Daniel Kang
71155d7b41 dsputil: x86: Convert mpeg4 qpel and dsputil avg to yasm
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2013-01-27 06:45:31 +01:00
Ronald S. Bultje
f90ff772e7 Move H264/QPEL specific asm from dsputil.asm to h264_qpel_*.asm. 2013-01-26 20:35:42 -08:00
Diego Biurrun
033a86f9bb x86: h264qpel: Move stray comment to the right spot and clarify it 2013-01-26 11:19:22 +01:00
Janne Grunau
c5c2060cf5 x86: h264qpel: add cpu flag checks for init function
The code was copied from per cpu extension init function so the checks
for supported extensions was overlooked.
2013-01-24 19:03:59 +01:00
Mans Rullgard
e9d817351b dsputil: Separate h264 qpel
The sh4 optimizations are removed, because the code is
100% identical to the C code, so it is unlikely to
provide any real practical benefit.

Signed-off-by: Diego Biurrun <diego@biurrun.de>
Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2013-01-24 10:44:43 +01:00
Ronald S. Bultje
baf35bb4bc dsputil: remove one array dimension from avg_no_rnd_pixels_tab. 2013-01-22 18:41:36 -08:00
Ronald S. Bultje
32ff643228 dsputil: remove avg_no_rnd_pixels8.
This is never used.
2013-01-22 18:41:36 -08:00
Diego Biurrun
88bd7fdc82 Drop DCTELEM typedef
It does not help as an abstraction and adds dsputil dependencies.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2013-01-22 18:32:56 -08:00