Several tweaks and cleanups: * Use .rept instead of repeated macros for repeating...
commit486bab6453eb9df012e976ac08904a7267989c31
authoramiconn <amiconn@a1c6a512-1295-4272-9138-f99709370657>
Wed, 19 Nov 2008 21:31:33 +0000 (19 21:31 +0000)
committeramiconn <amiconn@a1c6a512-1295-4272-9138-f99709370657>
Wed, 19 Nov 2008 21:31:33 +0000 (19 21:31 +0000)
tree7acc0727874ff6b307eff293a18172a3239cd895
parent04a9de4b14e94d45c782bbfcf8bd68e803ad6479
Several tweaks and cleanups: * Use .rept instead of repeated macros for repeating blocks. * Use MUL (variant) instead of MLA (variant) in the first step of the ARM scalarproduct() if there's no loop. * Unroll ARM assembler functions to 32 where not already done, plus the generic scalarproduct().

git-svn-id: svn://svn.rockbox.org/rockbox/trunk@19144 a1c6a512-1295-4272-9138-f99709370657
apps/codecs/demac/libdemac/vector_math16_armv5te.h
apps/codecs/demac/libdemac/vector_math16_armv6.h
apps/codecs/demac/libdemac/vector_math16_cf.h
apps/codecs/demac/libdemac/vector_math32_armv4.h
apps/codecs/demac/libdemac/vector_math_generic.h