Re: Sleeping function called from invalid context in 2.6.37xhci_hcd

[Sarah Sharp]

On Wed, Jan 05, 2011 at 06:59:59PM -0600, Robert Hancock wrote:
> On Wed, Jan 5, 2011 at 6:30 PM, Greg KH <gregkh@xxxxxxx> wrote:
> > On Wed, Jan 05, 2011 at 06:22:03PM -0600, Robert Hancock wrote:
> >> Just saw this on booting up on the latest Fedora Rawhide kernel
> >> based on 2.6.37. Looks like we were trying to do a runtime suspend
> >> on the xHCI USB controller and it disabled MSI-X while interrupts
> >> were disabled?
> >
> > Known issue, patches came in too late to make it for .37, hopefully will
> > show up in .38-rc and then get backported to .37.  They are on the
> > linux-usb mailing list if you need them now.
> 
> Not sure if it's related to this issue or not, but I also noticed the
> USB 3.0 card wasn't detecting any connected devices until I disabled
> USB autosuspend on the controller through sysfs (which is enabled by
> default in that Fedora Rawhide kernel)..

There's a couple more patches to deal with some issues with
disconnecting a USB device (or safely removing a USB mass storage
device) when the bus is suspended.  See the patches starting in this
thread:

http://marc.info/?l=linux-usb&m=129421308020494&w=2

Try applying the first five patches and let me know if you still see any
issues.  The fourth patch has a known compile warning (it will get fixed
shortly) but you shouldn't run into that particular error path.

Sarah Sharp
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/