Am 07.01.2015 um 14:20 schrieb Takashi Iwai:
At Wed, 07 Jan 2015 14:12:47 +0100, Stefan Seyfried wrote:
Having no real clue about the issue: wouldn't it be good to do something like this in snd_usbmidi_submit_urb()? Or to put it the other way round: will we end up putting this everywhere where snd_usbmidi_submit() is called eventually? :-)
Yeah, we need a better fix. But at least we need to know whether this is the real culprit at first. Basically this race shouldn't happen in a normal situation. Possibly some error recovery path is executed asynchronously and conflicts with the normal input start/stop sequence. If so, that's the place to fix, instead of papering over the all places.
Ok, I'll check the logs. Unfortunately the machine is powered down right now. It might have happened after resume from suspend to RAM (this machine is usually only suspended to RAM unless I explicitly request to reboot it after kernel updates etc), so some unusual recovery path might be involved here. Thanks, seife -- Stefan Seyfried "For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Richard Feynman -- To unsubscribe, e-mail: opensuse-kernel+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-kernel+owner@opensuse.org