Bug 1752 - Kernel Crash
: Kernel Crash
Product: Desktop Bugs
Classification: ROSA Desktop
Component: Main Packages
: Fresh
: x86_64 Linux
: Normal normal
: ---
Assigned To: Eugene Shatokhin
: Stanislav Fomin
Depends on:
  Show dependency treegraph
Reported: 2013-03-05 17:14 MSK by Stanislav Fomin
Modified: 2013-03-06 22:32 MSK (History)
0 users

See Also:
RPM Package: kernel-nrj-desktop-3.6.10-1rosa-1-1-rosa2012.1.x86_64
Bad POT generating:

Diagnostic information (18.77 KB, text/plain)
2013-03-05 17:39 MSK, Eugene Shatokhin

Note You need to log in before you can comment on or make changes to this bug.
Description Stanislav Fomin 2013-03-05 17:14:01 MSK
Description of problem:
Kernel Crash.

No user actions was made.
Laptop was without user interaction.

Comment 1 Eugene Shatokhin 2013-03-05 17:39:58 MSK
Created attachment 1250 [details]
Diagnostic information
Comment 2 Eugene Shatokhin 2013-03-06 18:49:08 MSK
I have looked through the C and binary code of task_active_pid_ns() kernel function (the one where the crash happened).

Looks like a hardware failure. The function has read an address of a structure from memory which differs from the valid address in a single bit only: 

0xff7f88022e6a4500 VS

The crash happens when the function tries to access the memory at that "bad" address.

I would recommend running Memtest86+ on that machine for several hours at least. Let's see if it finds something.

It is possible, of course, that something else may have corrupted memory in such a peculiar way... Let's see what Memtest86+ shows first.
Comment 3 Stanislav Fomin 2013-03-06 22:32:50 MSK
Yes, extensive memory testing confirm memory error (one bit?)

Sorry for disturbance. 

I also request for memtest in grub2 loader.