diff options
| author | Magnus Holmgren <magnushol@gmail.com> | 2008-03-09 12:36:30 +0000 |
|---|---|---|
| committer | Magnus Holmgren <magnushol@gmail.com> | 2008-03-09 12:36:30 +0000 |
| commit | ae00bd69d6c8e7ffb9a8caa1a96705e720c696ea (patch) | |
| tree | eadb8085287807504761f776484df87653ba9976 /apps/codecs/libfaad | |
| parent | 885283c424066e25faf691b604efb84b89145d71 (diff) | |
| download | rockbox-ae00bd69d6c8e7ffb9a8caa1a96705e720c696ea.zip rockbox-ae00bd69d6c8e7ffb9a8caa1a96705e720c696ea.tar.gz rockbox-ae00bd69d6c8e7ffb9a8caa1a96705e720c696ea.tar.bz2 rockbox-ae00bd69d6c8e7ffb9a8caa1a96705e720c696ea.tar.xz | |
AAC decoder: Remove suspicious use of the REAL_CONST macro. I don't know what triggers the execution of this code, but this change stops some floating point math code from being linked. It looks like the right thing to do...
git-svn-id: svn://svn.rockbox.org/rockbox/trunk@16590 a1c6a512-1295-4272-9138-f99709370657
Diffstat (limited to 'apps/codecs/libfaad')
| -rw-r--r-- | apps/codecs/libfaad/sbr_fbt.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/apps/codecs/libfaad/sbr_fbt.c b/apps/codecs/libfaad/sbr_fbt.c index 812c405..c685c71 100644 --- a/apps/codecs/libfaad/sbr_fbt.c +++ b/apps/codecs/libfaad/sbr_fbt.c @@ -687,7 +687,7 @@ restart: nOctaves = REAL_CONST(log((float)limTable[k]/(float)limTable[k-1])/log(2.0)); #else #ifdef FIXED_POINT - nOctaves = DIV_R((limTable[k]<<REAL_BITS),REAL_CONST(limTable[k-1])); + nOctaves = DIV_R((limTable[k]<<REAL_BITS),limTable[k-1]); #else nOctaves = (real_t)limTable[k]/(real_t)limTable[k-1]; #endif |