Riccardo Magliocchetti ha scritto: > Raymond Yau ha scritto: >> 2010/1/6 Riccardo Magliocchetti <riccardo.magliocchetti@xxxxxxxxx> >> >>> Hi Pierre, >>> >>> pl bossart ha scritto: >>>> You haven't specified the buffer/period size in the hw params. See >>>> alsa-lib/test/pcm.c >>>> - Pierre >>> They fail when opening default device so I had to remove them, work fine >>> when opening >>> plughw:0,0 though. >>> >>>> On Mon, Jan 4, 2010 at 11:39 AM, Riccardo Magliocchetti >>>> <riccardo.magliocchetti@xxxxxxxxx> wrote: >>>>> [the same message is waiting in moderator queue, sorry if you >>>>> receive it >>> two times] >>>>> Hello, >>>>> >>>>> i'm trying to capture audio data from the microphone, but i'm not able >>> to >>>>> read any frames. >>>>> The frames are always 0 and delay is fixed to 2730. I've tried calling >>>>> snd_pcm_avail_delay, snd_pcm_avail_update but does not make a >>> difference. >>>>> The soundcard is driven by snd_hda_intel. The code is more or less >>>>> the >>>>> same as Paul Davis' tutorial on using the alsa api. > >> >> Even 4096 bytes is x86 DMA page size , seem to be a good choice for those >> PCI sound cards, however you cannot assume all sound cards/plugins >> support >> this buffer size >> >> If you did not set period_size, buffer_size or periods (period_time, >> buffer_time or periods) , you should call >> snd_pcm_hw_params_get_buffer_size() and >> snd_pcm_hw_params_get_period_size() >> after snd_pcm_hw_params() > > Thank you both Pierre and Raymond, using the get functions after > snd_pcm_hw_params() is working, now i have some concurrency issues to > resolve before i can say everything is working properly. The init works fine but snd_pcm_avail() always returns 0, full code is here http://pastebin.com/f687bf37a thanks, riccardo _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
