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

Add a more correct absolute difference function to dsp-util.

Browse source 
Differences between signed samples cover the entire unsigned 32-bit
range. "abs" will think any difference exceeding INT32_MAX is negative
which is not corrent. Test which argument is greater and subtract the
lesser from it, outputting unsigned difference.

Change-Id: I73a8e5e418d49ff73d1a7c98eeb4731946dcfe84
This commit is contained in:
Michael Sevakis 2012-04-26 15:21:43 -04:00
parent 2866063c3c
commit e5c3327cef
2 changed files with 13 additions and 7 deletions
Download patch file Download diff file Expand all files Collapse all files

13
firmware/export/dsp-util.h
View file

@ -18,8 +18,8 @@
* KIND, either express or implied.
*
****************************************************************************/
#ifndef DSP_HELPER_H
#define DSP_HELPER_H
#ifndef DSP_UTIL_H
#define DSP_UTIL_H
/** Clip sample to signed 16 bit range **/
@ -48,4 +48,11 @@ static FORCE_INLINE int32_t clip_sample_16(int32_t sample)
#undef CLIP_SAMPLE_16_DEFINED
#endif /* DSP_HELPER_H */
/* Absolute difference of signed 32-bit numbers which must be dealt with
* in the unsigned 32-bit range */
static FORCE_INLINE uint32_t ad_s32(int32_t a, int32_t b)
{
return (a >= b) ? (a - b) : (b - a);
}
#endif /* DSP_UTIL_H */