On Wed, 11 Sep 2013, Daniel Mack wrote: > On 11.09.2013 14:53, Ankit Agarwal wrote: > > > > On Sun, Sep 8, 2013 at 8:27 PM, Daniel Mack <zonque@xxxxxxxxx > > <mailto:zonque@xxxxxxxxx>> wrote: > > > > On 02.09.2013 12:51, Ankit Agarwal wrote: > > > I am using alsa pcm player to listen to audio. The alsa pcm > > crashes with > > > the error: ALSA sound/usb/urb.c:179 cannot submit urb (err = -27). > > I am > > > able to hear the data initially but when other transaction begins > > on HID > > > report endpoint, alsa crashes with the mentioned error. > > > > -27 == -EFBIG, which seems to relate to a scheduling problem on the bus, > > and hence probably rather to a host controller bug than an issue with > > snd-usb. > > > > > I am working on imx6q board with linux kernel-3.0.35. I am using Apple > > > device to get the digital audio. > > > > Please try with a more recent kernel, and see if you can find the > > location where EFBIG is initially reported. > > > > > > I have found that EFBIG is being reported by ehci-host controller. It is > > coming from file ehci-sched.c under iso_stream_schedule() function. The > > error is thrown when the controller tries to schedule the urb too far > > into the future, ie, there is overflow of the scheduled URBs. I might be > > happening because the URBs are not scheduled properly. > > Ok. What about trying a newer kernel? The issue might be fixed already. -EFBIG is still returned under the same circumstances as in earlier kernels. You can get more information about those errors if you enable CONFIG_USB_DEBUG. Starting in the 3.11 kernel, however, -EXDEV is no longer returned for USB submission. It would be a good idea to test with 3.11. It also would be a good idea to get a usbmon trace showing what happens when the errors occur. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
