Peter Meerwald
578d2ce5c8
remap: fix check when to apply specialized code path
...
the specialized code path just duplicate samples, so are only
applicable if the volume in map_table is == 1.0 (or == 0x10000);
don't use them for volumes >= 1.0
compare the integer version of the volume stored in map_table;
comparing floats is ugly (als leads to compiler warnings)
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
2013-03-29 10:48:26 +02:00
Peter Meerwald
2c5d3d79ad
remap_sse: More specific logging: SSE -> SSE2
...
Signed-off-by: Peter Meerwald <p.meerwald@bct-electronic.com>
2013-02-01 07:26:16 +02:00
Maarten Bosmans
dd9265ac78
Remove unnecessary #includes
2011-06-22 23:12:20 +01:00
Maarten Bosmans
ecf09f2cd6
Fix up according to Coding Style
...
Only whitespace changes in here
2011-03-11 11:49:35 +00:00
Lennart Poettering
71e066c873
simd: be more precise which SIMD optimizations we activate
2009-09-09 04:28:22 +02:00
Wim Taymans
b5ac3839e1
x86: only install some functions when SSE2
...
Remap and volume functions use SSE2 instructions so only install them when SSE2
is present.
2009-09-07 17:21:21 +02:00
Lennart Poettering
6f396c89ab
remap: build sse code only on x86
2009-09-03 01:48:30 +02:00
Lennart Poettering
9f4f374a19
remap_sse: fix inner loop increment on SSE
...
In each iteration we can process 2^4 S16NE samples and 2^5 F32NE
samples, that's twice as much as in MMX, hence correct the increments.
2009-09-03 00:17:28 +02:00
Lennart Poettering
51423cae52
remap_sse: reindent macro so that diff to MMX is nicer
...
Completely useless, but diff -u remap_mmx.c remap_sse.c is much nicer
this way.
2009-09-03 00:16:40 +02:00
Wim Taymans
dc221f204b
remap: fix counters for mmx and sse remap
...
Take the size of the sample into account when calculating the amount of samples
we process in parallel.
2009-08-31 21:50:56 +02:00
Wim Taymans
59070892ed
remap: cleanup assembler a little
2009-08-27 10:44:52 +02:00
Wim Taymans
509d9f04e5
remap: add sse optimized mono to stereo
2009-08-27 10:44:52 +02:00
