On Tue, Feb 22, 2011 at 08:05:29AM -0800, Sarah Sharp wrote: > On Thu, Feb 17, 2011 at 05:06:15PM -0800, Greg KH wrote: > > On Thu, Feb 17, 2011 at 04:57:00PM -0800, Sarah Sharp wrote: > > > Greetings, > > > > > > This is a note to let Dmitry, Paul, and Luben know that while testing > > > their patches against Greg's usb-linus branch, my kernel crashed when I > > > plugged in a USB 3.0 hard drive. This was the first device to be > > > plugged into the host after driver load, so it should be fairly easy to > > > reproduce. > > > > > > At this point, I'm not sure whether one of the patches are to blame, or > > > if the problem lies somewhere in Greg's tree. I'm compiling a kernel to > > > do diagnosis now. Until I find the underlying cause, I'm not going to > > > be able to send any of your patches off to Greg. Thanks for your > > > patience. > > > > No problem, good luck debugging. > > Hi Greg, > > I haven't been able to reproduce my crash in a couple days of testing, > so I'm inclined to just send these bug fixes off. > > What happened with the original crash was a flaky USB 3.0 drive > disconnected sometime in the night, while I had a bash shell open to a > directory in the mount point. When I noticed the drive had > disconnected, and re-plugged it in and tried to play a movie from the > bash shell, the graphical session froze. I'm not sure if it was a > kernel crash, because the keyboard lights weren't blinking, and the > system responded to syscontrol keys to sync, unmount, and reboot. I > think I couldn't ssh into the box, although I can't clearly remember > that bit. The system log file was corrupted just after it showed the > device connecting, and I wasn't running netconsole at the time. > > I haven't seen this issue show up again with the flaky USB3 drive. I'm > inclined to think the original hang wasn't caused by the bug fix > patches, unless maybe Paul's ring patches caused a failure when the > drive re-connected? But ring math issues usually have the effect of > just causing the xHCI host controller to stop responding to transfers, > so my gut feeling is that Paul's patches aren't the cause either. > > Should I just send the patches off? Please do. thanks, greg k-h -- 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
