> I'll stick to test11 for now and pray that the ext2 corruption will not
> happen. On that subject, I read the archives of linux-kernel, and it
> appears confusing. What exactly are the symptoms, and how was the bug
> fixed? It it's a narrow patch, I can apply it to my -test11 tree.
IIRC it was due to the VM, dirty pages which were unallocated weren't
handled correctly. I believe the symptoms were that incorrect data was
written to disk, if you were lucky it only corrupted your files, but of
course when it hit directories, or superblocks...
The patch was in one of the earliest test12pre ones, I believe various fixes
went in before it was fully fixed.
Of course there is also suspicion of corruption with the Highpoint, and on
the IDE front there are probably problems with large disks, which have long
seeks which timeout. There's an extract from Andre Hedrik in the summary at
kt.linuxcare.com.
Don't get me wrong, I am not an expert on this, I know only because I was
considering using a 2.4 test kernel. The sheer weight of email, and the
current lack of a digest for linux-kernel, means keeping current is fairly
painful.
Good luck, I'd like to help the test effort for 2.4, but I cannot afford the
risks at present.
Rob
--
=- To unsubscribe, email [EMAIL PROTECTED] with the -=
=- body of "unsubscribe linux-abit". -=
