generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
42,688 Commits 24 Branches 241 Tags 173 MiB
2008f76054
Commit Graph

29 Commits

Author SHA1 Message Date
Alexandra Hájková
2008f76054 dca: remove unused decode_hf function and quant_d tables 
They were superseded with their integer equivalents. Rename integer
decode_hf to decode_hf.
 2015-12-24 13:58:18 +01:00
Janne Grunau
cc29d96d5a arm64: fix inverted register order in transpose_4x4H 
Fix related register order issue in ff_h264_idct_add_neon.

Found-by: zjh8890 <243186085@qq.com>
 2015-12-21 13:44:20 +01:00
Janne Grunau
a0fc780a20 arm64: int32_to_float_fmul neon asm 
3% faster dts decoding on a cortex-a57.

                                 cortex-a57   cortex-a53
int32_to_float_fmul_array8_c:    1270.9       4475.6
int32_to_float_fmul_array8_neon:  328.6        569.2
int32_to_float_fmul_scalar_c:     928.5       4119.6
int32_to_float_fmul_scalar_neon:  309.1        524.1
 2015-12-14 16:45:02 +01:00
Janne Grunau
705f5e5e15 arm64: port synth_filter_float_neon from arm 
~25% faster dts decoding overall. The checkasm CPU cycles numbers are
not that useful since synth_filter_float() calls FFTContext.imdct_half().

                         cortex-a57   cortex-a53
synth_filter_float_c:    1866.2       3490.9
synth_filter_float_neon:  915.0       1531.5

With fftc.imdct_half forced to imdct_half_neon:
                         cortex-a57   cortex-a53
synth_filter_float_c:    1718.4       3025.3
synth_filter_float_neon:  926.2       1530.1
 2015-12-14 16:45:01 +01:00
Janne Grunau
c33c1fa8af arm64: convert dcadsp neon asm from arm 
~2% faster dts decoding overall.

                    cortex-a57   cortex-a53
dca_decode_hf_c:    474.8        1659.9
dca_decode_hf_neon: 225.2         301.1
dca_lfe_fir0_c:     913.2        1537.7
dca_lfe_fir0_neon:  286.8         451.9
dca_lfe_fir1_c:     848.7        1711.5
dca_lfe_fir1_neon:  387.1         506.4
 2015-12-14 16:45:01 +01:00
Janne Grunau
f56d8d8dd7 h264: aarch64: intra prediction optimisations 2015-07-20 23:10:29 +02:00
Janne Grunau
c2de2cf0d2 arm64: constify src in h264qpel dsp function definitions 2015-06-24 08:41:32 +02:00
Diego Biurrun
3d5d46233c opus: Factor out imdct15 into a standalone component 
It will be reused by the AAC decoder.
 2015-02-02 16:07:33 +01:00
Martin Storsjö
780cd20b00 aarch64: Use .data.rel.ro for const data with relocations 
This reverts commit c00365b46d
in addition to using a different section.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2014-12-09 11:43:31 +02:00
Martin Storsjö
c00365b46d aarch64: Make the function pointer tables position independent 
This allows running the code on android, where 64 bit binaries with
text relocations aren't allowed to be loaded.

Signed-off-by: Martin Storsjö <martin@martin.st>
 2014-11-16 01:07:24 +02:00
Janne Grunau
ac6b95dbc0 aarch64: add ',' between assembler macro arguments where missing 
llvm's integrated assembler does not accept spaces as macro argument
delimiter when targeting darwin. Using a explicit delimiter is a good
idea in principle since it makes case like 'macro 4 -2' vs 'macro 4 - 2'
clear.
 2014-08-04 00:17:21 +02:00
Janne Grunau
f23d26a686 h264: avoid using uninitialized memory in NEON chroma mc 
Adapt commit 982b596ea6 for the arm and
aarch64 NEON asm. 5-10% faster on Cortex-A9.
 2014-06-23 16:32:15 +02:00
Janne Grunau
d3f5b94762 aarch64: opus NEON iMDCT and FFT 
Opus celt decoding 11% faster and the iMDCT over 2.5 times faster on
Apple's A7.
 2014-05-15 18:17:02 +02:00
Janne Grunau
9aa4592076 aarch64: assembler in clang-3.4 ignores the division by two 
Values are positive powers of two, so just replace it with right shift.
 2014-05-13 19:44:09 +02:00
Janne Grunau
3956a5e0ea aarch64: NEON vorbis_inverse_coupling 
From the ARMv7 NEON version. 16 times faster as the C version, overall
more than 12% faster vorbis decoding on Apple's A7.
 2014-04-22 22:01:45 +02:00
Janne Grunau
8f9fe6ae34 aarch64: NEON fixed/floating point MPADSP apply_window 
30%/25% (fixed/float) faster mp3 decoding on Apple's A7. The floating
point decoder is approximately 7% faster.
 2014-04-22 22:01:45 +02:00
Janne Grunau
ee2bc5974f aarch64: NEON float (i)MDCT 
Approximately as fast as the ARM NEON version on Apple's A7.
 2014-04-22 19:35:41 +02:00
Janne Grunau
650c4300d9 aarch64: NEON float FFT 
Approximately as fast as the ARM NEON version on Apple's A7.
 2014-04-22 19:35:40 +02:00
Janne Grunau
d3789eeeed aarch64: implement videodsp.prefetch 
8% faster h264 decoding on Apple A7.
 2014-04-06 21:18:49 +02:00
Diego Biurrun
0e083d7e43 build: Group general components separate from de/encoders in arch Makefiles 
This is in line with how the top-level libavcodec Makefile is structured.
 2014-03-20 05:03:23 -07:00
Janne Grunau
dfe224f377 aarch64: get_cabac inline asm 
Based on the x86 branchless get_cabac asm. get_cabac_noinline() gets
approximately 20% faster (no cycle counts available) compared to clang
from Xcode 5.1 beta5. More than 6% faster overall. A part of the overall
speedup might be explained by additional inlining of get_cabac().
 2014-03-09 00:45:33 +01:00
Janne Grunau
9c029f67ca aarch64: use EXTERN_ASM consistently for exported symbols 
Based on e3fec3f095 for arm.
 2014-02-20 15:24:35 +01:00
Janne Grunau
fe96769bed aarch64: port neon clobber test from arm 2014-01-15 12:31:07 +01:00
Janne Grunau
f896bca03f aarch64: h264 (bi)weight NEON optimizations 
Ported from ARMv7 NEON.
 2014-01-15 12:31:07 +01:00
Janne Grunau
36e3b1f2fd aarch64: h264 loop filter NEON optimizations 
Ported from ARMv7 NEON.
 2014-01-15 12:31:04 +01:00
Janne Grunau
c65d67ef50 aarch64: hpeldsp NEON optimizations 
Ported from ARMv7 NEON.
 2014-01-15 12:30:24 +01:00
Janne Grunau
d5dd8c7bf0 aarch64: h264 qpel NEON optimizations 
Ported from ARMv7 NEON.
 2014-01-15 12:17:49 +01:00
Janne Grunau
8438b3f09f aarch64: h264 idct NEON assembler optimizations 
Ported from ARMv7 NEON.
 2014-01-15 12:13:41 +01:00
Janne Grunau
71617884a2 aarch64: h264 chroma motion compensation NEON optimizations 
Since RV40 and VC-1 use almost the same algorithm so optimizations for
those two decoders are easy to do and included.
 2014-01-15 12:07:18 +01:00