diff options
| author | Jens Arnold <amiconn@rockbox.org> | 2008-10-03 09:33:36 +0000 |
|---|---|---|
| committer | Jens Arnold <amiconn@rockbox.org> | 2008-10-03 09:33:36 +0000 |
| commit | 46bf6bd49301cdfe254253725b7a104b32da1061 (patch) | |
| tree | 00ed95eb0782527df6be4503df4b1b3076783f3b /apps/codecs/demac/libdemac/vector_math16_armv6.h | |
| parent | 6fcf2765dd01703e940dce5d891a16e28180a508 (diff) | |
| download | rockbox-46bf6bd49301cdfe254253725b7a104b32da1061.tar.gz rockbox-46bf6bd49301cdfe254253725b7a104b32da1061.zip | |
Add preliminary ARMv5te optimisations (verified working, but can probably be sped up further), and fix a comment in the ARMv6 code.
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@18693 a1c6a512-1295-4272-9138-f99709370657
Diffstat (limited to 'apps/codecs/demac/libdemac/vector_math16_armv6.h')
| -rw-r--r-- | apps/codecs/demac/libdemac/vector_math16_armv6.h | 4 |
1 files changed, 1 insertions, 3 deletions
diff --git a/apps/codecs/demac/libdemac/vector_math16_armv6.h b/apps/codecs/demac/libdemac/vector_math16_armv6.h index 7ecf372462..592e9dc7e6 100644 --- a/apps/codecs/demac/libdemac/vector_math16_armv6.h +++ b/apps/codecs/demac/libdemac/vector_math16_armv6.h @@ -205,9 +205,7 @@ static inline void vector_sub(int16_t* v1, int16_t* v2) /* This version fetches data as 32 bit words, and *requires* v1 to be * 32 bit aligned, otherwise it will result either in a data abort, or - * incorrect results (if ARM aligncheck is disabled). It is optimised - * for ARM7TDMI. Using it for ARM9 or higher results in worse performance - * than the C version. */ + * incorrect results (if ARM aligncheck is disabled). */ static inline int32_t scalarproduct(int16_t* v1, int16_t* v2) { int res = 0; |