FS#34922 - [linux-lts] 3.0.74 xhci_hcd / kernel bug -- most USB devices won't work
Attached to Project:
Arch Linux
Opened by Kevin (palintropos) - Wednesday, 24 April 2013, 03:18 GMT
Last edited by Andreas Radke (AndyRTR) - Monday, 07 October 2013, 14:40 GMT
Opened by Kevin (palintropos) - Wednesday, 24 April 2013, 03:18 GMT
Last edited by Andreas Radke (AndyRTR) - Monday, 07 October 2013, 14:40 GMT
|
Details
Description:
I did a general pacman -Syu type upgrade today, which I usually do once or twice a week. There were a load of packages though, since I had missed the last one or two due to a dependency error in libsoup-gnome. A ton of packages updated. When I rebooted the first couple of times, all my USB devices died before I could even log in. (My keyboard is USB.) Unplugging and replugging gives NO dmesg output at all. The LED on my mouse flashes for just a second then goes out. This time I finally managed to log in before USB died and for some odd reason my keyboard stayed working. Additional info: * linux-lts 3.0.74-1-lts 64-bit * attached dmesg output from this boot (described above) * will attach other logs by request since I'm not sure at ALL what's causing this bug * Someone should really tell the kernel developers "OOPS:" is a really disheartening log line. Steps to reproduce: ? ? ? |
This task depends upon
Closed by Andreas Radke (AndyRTR)
Monday, 07 October 2013, 14:40 GMT
Reason for closing: No response
Monday, 07 October 2013, 14:40 GMT
Reason for closing: No response
Which was the last working kernel?
Could it be a hardware bug?
No. I have both linux and linux-lts package and rebooting into linux did not fix. Should have tried this last night.
Which was the last working kernel?
The last minor version # in each case, so 3.0.73 and 3.8.7 I suppose.
Could it be a hardware bug?
Sure, I suppose. I haven't narrowed this enough to rule anything out. It doesn't act like one, though. For one thing, both the front and back USBs are simultaneously and suddenly equally affected after the latest pacman -Syu. For another thing, depending on the order of boot (when the error and stack dump happens) I can continue to use my keyboard, which is USB.
I mentioned it might be related to the kernel because Googling keywords from the log file turned up lots of old kernel bugs and patch notes, and because most of the serious crashing happens after the line that says something like BUG: Kernel unable to handle null pointer dereference, which seems actually like a bug in a kernel module at a guess, but I really have no idea.
Thank you for taking a look at it with me.