| commit | 4bd6d85ab1ff9377e2c0c98d70e67b47391729d0 | |
| author | amiconn <amiconn@a1c6a512-1295-4272-9138-f99709370657> | |
| Wed, 10 Feb 2010 23:23:17 +0000 (10 23:23 +0000) | ||
| committer | amiconn <amiconn@a1c6a512-1295-4272-9138-f99709370657> | |
| Wed, 10 Feb 2010 23:23:17 +0000 (10 23:23 +0000) | ||
| tree | 4c7d6df8d4d43b07a8cf17b6eedaf950fbd38d1b | treesnapshot (tar.gz zip) |
| parent | 54ed89ba8a6cddfed5ef591bc9f8d253b934d673 | commitdiff |
APE: Fused vector math for the filters on ARMv5te. Speedup on Cowon D2 is ~4% for -c2000..-c4000 (less for -c5000). Thanks to Frank Gevaerts for testing.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24590 a1c6a512-1295-4272-9138-f99709370657
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@24590 a1c6a512-1295-4272-9138-f99709370657
| apps/codecs/demac/libdemac/vector_math16_armv5te.h | diffblobblamehistory |