Re: BUG: unable to handle kernel paging request at virtual addressfeededed (was: Re: Linux v2.6.17-rc5)

[Linus Torvalds]

On Sun, 28 May 2006, Tomasz Torcz wrote:
>
>   After 2 days and few hours uptime, during updatedb run I got:
> 
> BUG: unable to handle kernel paging request at virtual address feededed

Looks like one of the magic numbers ("0xfee1dead", "0xfeedbeef", 
0xfeedface"), but that's not it.

>  It never happened before. d_splice_alias in bt is very strange, as I don't
> think anything on my system uses splice(). It's too new, and my system is
> Slackware -current (which seems to return ENOSUPORTED even for old stuff
> like posix_fadvise()).

No, d_splice_alias() is a different kind of splicing: it splices a dentry 
entry into the alias list. Nothing to do with the new splice() system 
call, except that the naming comes from the same english word ("splice: to 
join two ropes by interweaving strands").

I don't see anything suspicious anywhere, and this doesn't ring a bell. 
It is probably a good idea to open a bugzilla entry on it, so that it 
doesn't get lost. And perhaps cc the reiserfs people (there's been a few 
reiserfs changes since 2.6.16, but none of them looks suspicious to me: 
however, maybe this makes somebody else go "Aaah!").

Try Jan Kara <jack@xxxxxxx>, Jeff Mahoney <jeffm@xxxxxxxx> and 
Alexander Zarochentzev <zam@xxxxxxxxxxx>.

			Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/