| |||||||||||||
|
| |||||||||||||
The 2.6.27 kernel is outThe 2.6.27 kernel is outPosted Oct 10, 2008 12:19 UTC (Fri) by shlomif (subscriber, #11299)Parent article: The 2.6.27 kernel is out
Congrats! Question: was the e1000 problem fixed?
(Log in to post comments)
The 2.6.27 kernel is out Posted Oct 10, 2008 13:00 UTC (Fri) by Tuxie (guest, #47191) [Link]
No, but a layer of protection was added so it will print some debug data in your syslog instead of actually destroying the e1000 when the bug occurs.
The 2.6.27 kernel is out Posted Oct 10, 2008 15:37 UTC (Fri) by iabervon (subscriber, #722) [Link]
The source of the problematic writes could be in a wide variety of places, many of which aren't in the kernel. 2.6.27 has a fix to make the hardware not break when they happen. The kernel fix makes e1000e behave like other hardware in not getting broken when those writes happen, so finding the real problem is less of an emergency, and seems likely to be some sort of X probing issue.
Device register violation Posted Oct 13, 2008 21:29 UTC (Mon) by ncm (subscriber, #165) [Link]
I see what might be a similar sort of problem on my "nv" laptop: running "lshw" as superuser clobbers the display device registers. I have to reboot to get back control of the display.
|
|
||||||||||||