Commit Graph

40 Commits

Author SHA1 Message Date
Justin Ruggles
10e645e9cb lavr: handle clipping in the float to s32 conversion
We cannot clip to INT_MAX because that value cannot be exactly
represented by a float value and ends up overflowing during conversion
anyway. We need to use a slightly smaller float value, which ends up
with slightly inaccurate results for samples which clip or nearly clip,
but it is close enough. Using doubles as intermediates in the conversion
would be more accurate, but it takes about twice as much time.

Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2012-10-13 12:34:34 +02:00
Diego Biurrun
e0c6cce447 x86: Replace checks for CPU extensions and flags by convenience macros
This separates code relying on inline from that relying on external
assembly and fixes instances where the coalesced check was incorrect.
2012-09-08 18:18:34 +02:00
Diego Biurrun
17337f54c0 x86: Split inline and external assembly #ifdefs 2012-08-31 01:53:25 +02:00
Diego Biurrun
a886b279a0 x86: cosmetics: Comment some #endifs for better readability 2012-08-30 18:50:33 +02:00
Justin Ruggles
06e751a40f lavr: x86: optimized 6-channel flt to fltp conversion 2012-08-23 20:10:57 -04:00
Justin Ruggles
e07c9705c8 lavr: x86: optimized 2-channel flt to fltp conversion 2012-08-23 20:10:57 -04:00
Justin Ruggles
5245c9f3ad lavr: x86: optimized 6-channel flt to s16p conversion 2012-08-23 20:10:57 -04:00
Justin Ruggles
31d0d7181d lavr: x86: optimized 2-channel flt to s16p conversion 2012-08-23 20:10:57 -04:00
Justin Ruggles
6092dafb5a lavr: x86: optimized 6-channel s16 to fltp conversion 2012-08-23 20:10:57 -04:00
Justin Ruggles
91851a7b37 lavr: x86: optimized 2-channel s16 to fltp conversion 2012-08-23 20:10:57 -04:00
Justin Ruggles
205ace8843 lavr: x86: optimized 6-channel s16 to s16p conversion 2012-08-23 20:10:57 -04:00
Justin Ruggles
8eeffa8ada lavr: x86: optimized 2-channel s16 to s16p conversion 2012-08-23 20:10:57 -04:00
Justin Ruggles
b66e20d2aa lavr: x86: optimized 2-channel fltp to flt conversion 2012-08-23 20:10:56 -04:00
Justin Ruggles
d5b4e50c47 lavr: x86: optimized 6-channel fltp to s16 conversion 2012-08-23 20:10:56 -04:00
Justin Ruggles
a58a013980 lavr: x86: optimized 2-channel fltp to s16 conversion 2012-08-23 20:10:56 -04:00
Justin Ruggles
90cc27f813 lavr: x86: optimized 6-channel s16p to flt conversion 2012-08-23 20:10:56 -04:00
Justin Ruggles
46f929adad lavr: x86: optimized 2-channel s16p to flt conversion 2012-08-23 20:10:56 -04:00
Justin Ruggles
13df7d2d40 lavr: x86: optimized 6-channel s16p to s16 conversion 2012-08-23 20:10:56 -04:00
Justin Ruggles
c0e12535aa lavr: x86: optimized 2-channel s16p to s16 conversion 2012-08-23 20:10:56 -04:00
Mans Rullgard
a3df4781f4 x86: add colons after labels
nasm prints a warning if the colon is missing.

Signed-off-by: Mans Rullgard <mans@mansr.com>
2012-08-07 15:20:56 +01:00
Justin Ruggles
e9da9a3111 lavr: x86: improve non-SSE4 version of S16_TO_S32_SX macro
Removes a false dependency on existing contents of the 2nd dst register,
giving better performance for OOE.
2012-07-27 14:21:32 -04:00
Justin Ruggles
2f096bb10e lavr: add x86-optimized mixing functions
Adds optimized functions for mixing 3 through 8 input channels to 1 and 2
output channels in fltp or s16p format with flt coeffs.
2012-07-27 11:25:48 -04:00
Ronald S. Bultje
30b45d9c38 x86inc: automatically insert vzeroupper for YMM functions. 2012-07-26 13:43:16 -07:00
Justin Ruggles
0dadf9d1e9 lavr: x86: add missing vzeroupper in ff_mix_1_to_2_fltp_flt() 2012-07-25 15:41:25 -04:00
Justin Ruggles
acd9948e74 lavr: x86: fix ff_conv_fltp_to_flt_6ch function prototypes
Changed to match the number of parameters in conv_func_interleave(), which is
how they are called. The change isn't strictly necessary because the 4th
parameter is not used, but the code is clearer if they match.
2012-06-26 12:29:35 -04:00
Justin Ruggles
14a34d90ad lavr: x86: merge some branches 2012-06-25 13:49:18 -04:00
Justin Ruggles
4e4dd71730 lavr: Add x86-optimized function for flt to s32 conversion 2012-06-18 16:16:59 -04:00
Justin Ruggles
6c63cbfe7a lavr: Add x86-optimized function for flt to s16 conversion 2012-06-18 16:16:59 -04:00
Justin Ruggles
97ce1ba867 lavr: Add x86-optimized functions for s32 to flt conversion 2012-06-18 16:16:59 -04:00
Justin Ruggles
5904f25b9f lavr: Add x86-optimized functions for s32 to s16 conversion 2012-06-18 16:16:59 -04:00
Justin Ruggles
d721f67d0a lavr: Add x86-optimized functions for s16 to flt conversion 2012-06-18 16:16:59 -04:00
Justin Ruggles
1168e29df1 lavr: Add x86-optimized function for s16 to s32 conversion 2012-06-18 16:16:59 -04:00
Justin Ruggles
f61ce90caa lavr: add x86-optimized functions for mixing 1-to-2 s16p with flt coeffs 2012-06-18 11:24:10 -04:00
Justin Ruggles
29f7490c46 lavr: add x86-optimized functions for mixing 1-to-2 fltp with flt coeffs 2012-06-18 11:24:10 -04:00
Justin Ruggles
b75726cb79 lavr: add x86-optimized function for mixing 2 to 1 s16p with q8 coeffs 2012-05-29 15:33:25 -04:00
Justin Ruggles
c140fb2cbc lavr: add x86-optimized functions for mixing 2 to 1 s16p with float coeffs 2012-05-29 15:33:18 -04:00
Justin Ruggles
5cc6d5244d lavr: replace the SSE version of ff_conv_fltp_to_flt_6ch() with SSE4 and AVX
The current SSE version is slower than the MMX version on Athlon64 and Sandy
Bridge, but the SSE4 and AVX versions are faster on Sandy Bridge.
2012-05-09 16:17:59 -04:00
Diego Biurrun
dbe6ba55a3 build: cosmetics: Add missing end-of-line backslashes to item lists. 2012-05-07 14:17:40 +02:00
Reimar Döffinger
e5b7d7773a Fix compilation with YASM/NASM without AVX support.
Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
Signed-off-by: Justin Ruggles <justin.ruggles@gmail.com>
2012-05-01 13:37:25 -04:00
Justin Ruggles
c8af852b97 Add libavresample
This is a new library for audio sample format, channel layout, and sample rate
conversion.
2012-04-24 21:28:27 -04:00