Pierre-Louis Bossart wrote: > On 6/12/17 1:50 AM, Clemens Ladisch wrote: >> Subhransu S. Prusty wrote: >>> Add new hw_params flag to explicitly tell driver that rewinds will never >>> be used. [...] >> >>> +++ b/sound/core/pcm_native.c >>> @@ -2521,6 +2523,9 @@ static snd_pcm_sframes_t snd_pcm_playback_rewind(struct snd_pcm_substream *subst >>> if (frames == 0) >>> return 0; >>> >>> + if (runtime->no_rewinds) >>> + return 0; >> >> Doing a rewind when you've promised not to is an error. So why doesn't >> this return an error code? > > there were earlier patches where we switched from 0 to an error code > based on Takashi's feedback. I am not that convinced that applications > actually interpret signed frames as error codes and might interpret > negative values literally When the application has set SND_PCM_HW_PARAMS_NO_REWINDS, calling snd_pcm_rewind() already is an error to begin with. There's no point in trying to make it work, you can just choose how it breaks. Regards, Clemens _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
