projects / platform / upstream / libav.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: ce9ed10) | patch
fmtconvert: Add a new method, int32_to_float_fmul_array8
authorBen Avison <bavison@riscosopen.org>
Mon, 15 Jul 2013 17:28:11 +0000 (18:28 +0100)
committerMartin Storsjö <martin@martin.st>
Mon, 22 Jul 2013 07:15:33 +0000 (10:15 +0300)
commit31c6f6f65c0ed5a894e26ce44ab0c3e89c82b9a2
treef59f07d09cfaadd074431649aed42bb12b2d6b39tree | snapshot
parentce9ed10ac27b9cf32a6257e083ea2f052692d971commit | diff
fmtconvert: Add a new method, int32_to_float_fmul_array8

This is similar to int32_to_float_fmul_scalar, but
loads a new scalar multiplier every 8 input samples.
This enables the use of much larger input arrays, which
is important for pipelining on some CPUs (such as
ARMv6).

Signed-off-by: Martin Storsjö <martin@martin.st>
libavcodec/fmtconvert.c diff | blob | history
libavcodec/fmtconvert.h diff | blob | history
Domain: Multimedia / Media Content;