Commit Graph

16 Commits

Author SHA1 Message Date
Ben Avison
87552d54d3 armv6: Accelerate ff_fft_calc for general case (nbits != 4)
The previous implementation targeted DTS Coherent Acoustics, which only
requires nbits == 4 (fft16()). This case was (and still is) linked directly
rather than being indirected through ff_fft_calc_vfp(), but now the full
range from radix-4 up to radix-65536 is available. This benefits other codecs
such as AAC and AC3.

The implementaion is based upon the C version, with each routine larger than
radix-16 calling a hierarchy of smaller FFT functions, then performing a
post-processing pass. This pass benefits a lot from loop unrolling to
counter the long pipelines in the VFP. A relaxed calling standard also
reduces the overhead of the call hierarchy, and avoiding the excessive
inlining performed by GCC probably helps with I-cache utilisation too.

I benchmarked the result by measuring the number of gperftools samples that
hit anywhere in the AAC decoder (starting from aac_decode_frame()) or
specifically in the FFT routines (fft4() to fft512() and pass()) for the
same sample AAC stream:

              Before          After
              Mean   StdDev   Mean   StdDev  Confidence  Change
Audio decode  2245.5 53.1     1599.6 43.8    100.0%      +40.4%
FFT routines  940.6  22.0     348.1  20.8    100.0%      +170.2%

Signed-off-by: Martin Storsjö <martin@martin.st>
2014-07-18 01:34:23 +03:00
Diego Biurrun
bd549cbaac arm: dcadsp: Move synth filter initialization to dcadsp file 2013-08-29 11:24:14 +02:00
Martin Storsjö
b63bb251ea arm: Add VFP-accelerated version of imdct_half
Before           After
               Mean    StdDev   Mean    StdDev  Change
This function   2653.0  28.5     1108.8  51.4   +139.3%
Overall        17049.5 408.2    15973.0 223.2     +6.7%

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:37 +03:00
Ben Avison
41ef1d360b arm: Add VFP-accelerated version of synth_filter_float
Before           After
               Mean    StdDev   Mean    StdDev  Change
This function   9295.0 114.9     4853.2 83.5    +91.5%
Overall        23699.8 397.6    19285.5 292.0   +22.9%

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:17 +03:00
Mans Rullgard
d526c5338d ARM: allow runtime masking of CPU features
This allows masking CPU features with the -cpuflags avconv option
which is useful for testing different optimisations without rebuilding.

Signed-off-by: Mans Rullgard <mans@mansr.com>
2012-04-22 12:30:45 +01:00
Felipe Contreras
c3d5e290ca ARM: fix build with FFT enabled and MDCT disabled
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
Signed-off-by: Mans Rullgard <mans@mansr.com>
2012-01-20 16:14:01 +00:00
Mans Rullgard
0aded9484d Move dct and rdft definitions to separate files
This leaves fft.h with only the core FFT and MDCT definitions
thus making it more managable.

Signed-off-by: Mans Rullgard <mans@mansr.com>
2011-03-20 17:15:33 +00:00
Mans Rullgard
2912e87a6c Replace FFmpeg with Libav in licence headers
Signed-off-by: Mans Rullgard <mans@mansr.com>
2011-03-19 13:33:20 +00:00
Loren Merritt
e6b1ed693a FFT: factor a shuffle out of the inner loop and merge it into fft_permute.
6% faster SSE FFT on Conroe, 2.5% on Penryn.

Signed-off-by: Janne Grunau <janne-ffmpeg@jannau.net>
2011-02-13 15:36:39 +01:00
Justin Ruggles
80ba1ddb58 Remove unneeded add bias from 3 functions.
DSPContext.vector_fmul_window()
DCADSPContext.lfe_fir()
SynthFilterContext.synth_filter_float()

Signed-off-by: Mans Rullgard <mans@mansr.com>
2011-01-31 20:28:42 +00:00
Måns Rullgård
e73d1a5efc ARM: NEON optimised synth_filter_float
2.7x faster DCA decoding on Cortex-A8

Originally committed as revision 22828 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-04-10 16:27:56 +00:00
Måns Rullgård
a8bb9ea532 ARM: NEON optimised RDFT
Originally committed as revision 22641 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-23 03:35:02 +00:00
Måns Rullgård
1429224b04 Move FFT parts from dsputil.h to fft.h
Originally committed as revision 22235 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-03-06 14:34:46 +00:00
Måns Rullgård
f7a3b6030c ARM: interleave cos/sin tables for improved NEON MDCT
Originally committed as revision 19940 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-09-21 02:56:09 +00:00
Måns Rullgård
01b2214758 Merge FFTContext and MDCTContext
Originally committed as revision 19931 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-09-20 17:30:20 +00:00
Måns Rullgård
f486321395 Move per-arch fft init bits into the corresponding subdirs
Originally committed as revision 19864 to svn://svn.ffmpeg.org/ffmpeg/trunk
2009-09-15 21:14:14 +00:00