On Fri, May 18, 2012 at 11:46:09AM -0700, Greg KH wrote: > On Fri, May 18, 2012 at 10:44:22AM +0000, Chen Peter-B29397 wrote: > > > > > On Thu, May 17, 2012 at 02:40:03PM -0400, Alan Stern wrote: > > > > On Thu, 17 May 2012, Greg KH wrote: > > > > > > > > > > Sometimes the keyboard gets locked up and I move the mouse around > > > to > > > > > > unstick it. In that case it seems like there is normally one > > > > > > character in the buffer. A couple times I was backspacing and it > > > > > > just started chewing through my text until I touched the mouse. > > > > > > > > > > Alan pointed me at this older thread, sorry I missed it first time > > > > > around. > > > > > > > > > > But I too can duplicate this here, on my usb-next branch, so it's at > > > > > least localized there so far. If I get time this weekend, I'll try > > > to > > > > > bisect things down to find the root problem... > > > > > > > > Assuming you have CONFIG_USB_DEBUG enabled, another thing you can do is > > > > get a copy of the files under /sys/kernel/debug/usb/ehci/0000:00:1a.0/ > > > > at a time while the problem is occurring. > > > > > > > > (Of course, this means you'll have to use a non-USB keyboard or a > > > > network login to copy the files.) > > > > > > > > I have no idea if it will help, but it might show something useful. > > > > > > Ok, I just duplicated this, and logged into it remotely to get the > > > information and not touch the keyboard. > > > > > > Attached is the dmesg, and the ehci debugfs files. > > > > > > The trackball isn't showing any movements on the screen, yet if I move > > > it, the ball lights up so it is trying to do something (i.e. it's > > > getting power.) > > > > > > I'll leave it like this for now for a few hours, have to go out at the > > > moment, so I can try to do any needed tests. > > > > > > > It may be similar with the problem I met with usb-hub (periods transfer abnormal) > > at linux-next tree. > > > > http://marc.info/?t=133727396500042&r=1&w=2 > > Thanks for the pointer, I've reverted that patch and let's see if it > solves the issue for me or not... Ok, this seems to have solved the problem for me. Alan, I'm going to revert 3d9545cc375d117554a9b35dfddadf9189c62775 (EHCI: maintain the ehci->command value properly), your as1543 patch, if you don't mind, as we are running out of time for the 3.5 merge window (I'd like to close my tree by the end of today given the immanent 3.4 release.) 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
