At Thu, 24 Aug 2006 08:09:26 +0200, Krzysztof Helt wrote: > > > > Dnia 23-08-2006 o godz. 20:16 Takashi Iwai napisał(a): > > At Wed, 23 Aug 2006 20:10:35 +0200, > > Krzysztof Helt wrote: > > > > > > @@ -2159,7 +2139,7 @@ static snd_pcm_uframes_t snd_dbri_pointe > > > snd_pcm_uframes_t ret; > > > > > > ret = bytes_to_frames(substream->runtime, info->offset) > > > - % substream->runtime->buffer_size; > > >+ % (substream->runtime->buffer_size + 1); > > > dprintk(D_USR, "I/O pointer: %ld frames of %ld.\n", > > > ret, substream->runtime->buffer_size); > > > return ret; > > > > What is the reason of this change? The original code looks OK > to me. > > The pointer callback must return a value between 0 and > > (buffer_size-1). > > > > > > I will recheck if recording works without this change. The last > returned pointer value is equal to buffer_size. Your email > implies that 0 means "I have full buffer filled with data". No, pointer callback returns the current processing DMA position in the ring buffer (called hw_ptr). It doesn't mean how many samples remain. The amount of remaining sample is calculated in PCM core layer. The PCM layer remembers how many samples are put in the buffer (= appl_ptr), then can easily calculate the remaining samples by (appl_ptr - hw_ptr). Takashi ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/alsa-devel
