devs/foxbox
1
0
Fork 0
forked from len0rd/rockbox
Code Pull requests Activity

Revert r25929. Test have shown that the assembler code is more than 50% faster than the C code both on ARM7TDMI (tested on PP5002 and PP5022) and on ARM1136JF-S (tested on Gigabeat S). If it is slower on ARM9*, it should be disabled for ARM9 only.

Browse source 
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@25937 a1c6a512-1295-4272-9138-f99709370657
This commit is contained in:
Jens Arnold 2010-05-10 19:05:25 +00:00
parent ed704825d2
commit 08d3c0be37
4 changed files with 287 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

6
apps/codecs/libffmpegFLAC/decoder.c
View file

@ -44,6 +44,8 @@
#if defined(CPU_COLDFIRE)
#include "coldfire.h"
#elif defined(CPU_ARM)
#include "arm.h"
#endif
#define FFMAX(a,b) ((a) > (b) ? (a) : (b))
@ -262,6 +264,10 @@ static int decode_subframe_lpc(FLACContext *s, int32_t* decoded, int pred_order)
(void)sum;
lpc_decode_emac(s->blocksize - pred_order, qlevel, pred_order,
decoded + pred_order, coeffs);
#elif defined(CPU_ARM)
(void)sum;
lpc_decode_arm(s->blocksize - pred_order, qlevel, pred_order,
decoded + pred_order, coeffs);
#else
for (i = pred_order; i < s->blocksize; i++)
{