Commit Graph

10 Commits

Author SHA1 Message Date
Christophe Gisquet
4cb6964244 dcadec: simplify decoding of VQ high frequencies
The vector dequantization has a test in a loop preventing effective SIMD
implementation. By moving it out of the loop, this loop can be DSPized.

Therefore, modify the current DSP implementation. In particular, the
DSP implementation no longer has to handle null loop sizes.

The decode_hf implementations have following timings:

For x86 Arrandale:
        C  SSE SSE2 SSE4
win32: 260 162  119  104
win64: 242 N/A   89   72

The arm NEON optimizations follow in a later patch as external asm. The
now unused check for the y modifier in arm inline asm is removed from
configure.
2014-02-28 13:03:22 +01:00
Christophe Gisquet
87ec849fe9 dcadec: remove scaling in lfe_interpolation_fir
The scaling factor is constant so it is faster to scale the
FIR coefficients in the tables during compilation.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
2014-02-28 13:00:47 +01:00
Christophe Gisquet
5fdbfcb5b7 dcadsp: split lfe_dir cases
The x86 runs short on registers because numerous elements are not static.
In addition, splitting them allows more optimized code, at least for x86.

Arm asm changes by Janne Grunau.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
2014-02-07 22:54:18 +01:00
Christophe Gisquet
5b59a9fc61 x86: dcadsp: implement int8x8_fmul_int32
For the callable function (as opposed to the inline one):
         C  SSE  SSE2  SSE4
Win32:  47   42   29    26
Win64:  30   33   25    23
The SSE version is neither compiled nor set for ARCH_X86_64, as the
inlinable function takes over.

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
2014-02-07 22:52:40 +01:00
Christophe Gisquet
2bd44cb705 dcadsp: add int8x8_fmul_int32 to dsp context
It is currently declared as a macro who is set to inlinable functions,
among which a Neon and a default C implementations.

Add a DSP parameter to each inline function, unused except by the
default C implementation which calls a function from the DSP context.

On an Arrandale CPU, gain for an inlined SSE2 function vs. a call:
- Win32: 29 to 26 cycles
- Win64: 25 to 23 cycles

Signed-off-by: Janne Grunau <janne-libav@jannau.net>
2014-02-07 22:51:59 +01:00
Ben Avison
800ffab48a dcadsp: Add a new method, qmf_32_subbands
This does most of the work formerly carried out by
the static function qmf_32_subbands() in dcadec.c.

Signed-off-by: Martin Storsjö <martin@martin.st>
2013-07-22 10:15:42 +03:00
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
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
08255107cf DCA: ARM/NEON optimised lfe_fir
Originally committed as revision 22863 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-04-12 20:45:33 +00:00
Måns Rullgård
309d16a4a0 DCA: break out lfe_interpolation_fir() inner loops to a function
This enables SIMD optimisations of this function.

Originally committed as revision 22861 to svn://svn.ffmpeg.org/ffmpeg/trunk
2010-04-12 20:45:25 +00:00