On 27.01.2015 13:11, Heinz Diehl wrote: > Hi, > > I'm not sure if this is the right place to report this, and would be glad if somebody could > point me to the right one if I should be wrong. > > Anyway, since the main error message comes from xhci_hcd: > When resuming from suspend to RAM, sometimes (= about every 4. or 5. suspend) my external USB3 harddisks filesystem (XFS) gets corrupted > with I/O block errors. This is what dmesg says: > > [....] > [ 274.103912] nouveau [ CLK][0000:01:00.0] --: core 500 MHz shader 1250 MHz memory 333 MHz > [ 274.104561] nouveau [ DRM] resuming client object trees... > [ 274.104923] nouveau [ DRM] resuming display... > [ 274.155454] nouveau [ DRM] resuming console... > [ 274.156877] xhci_hcd 0000:02:00.0: Host took too long to start, waited 16000 microseconds. > [ 274.156896] xhci_hcd 0000:02:00.0: PCI post-resume error -19! > [ 274.156897] xhci_hcd 0000:02:00.0: HC died; cleaning up Looks like xhci tries to re-initialize after resume, and fails to get it back running. It should only re-initializes in resume if it - was hibernated (S4) - xhci failed normal restore in resume - a special RESET_ON_RESUME quirk is set. Does resume work without the usb drive connected? Does resume work with older kernels? -Mathias -- 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
