lkml.org 
[lkml]   [2004]   [Jan]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: 2.6.1-mm2: BUG in kswapd? / Oops in kobject_put during rsync
Maneesh Soni wrote:
> Hi Walt,
>
> Earlier you had BUG in kswapd while running rsync and reverting this patch
> solved the problem.
>
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.1/2.6.1-mm3/broken-out/sysfs_remove_dir-vs-dcache_readdir-race-fix.patch
>
> Did you get the kswapd BUG hit again on 2.6.1-mm3 with the debug patch I sent
> you earlier? If you have got any logs with this then please send them to me.
>
> If I am not wrong the kobject_put() oops is the new one you are seeing now
> and reverting this sysfs-pin-kobjects.patch solves this oops. The call trace
> for kobject_put oops seems some what surprising to me because on -mm kernels
> I don't think we can have sysfs dentries on the LRU list, sysfs dentries are
> not pruned due to any memory pressure. sysfs dentries are pinned in memory
> so there is no question of them being pruned due to memory pressure.
>
> Now this is somewhat different in case of -mjb kernels as it has sysfs backing
> store patches and with that I can think of sysfs dentires coming to LRU list
> and getting pruned due to memory pressure.
>
> Are you using -mjb tree or the sysfs backing store patches? I would like
> to take a look at the kobject_put() oops you are having.
>
> Thanks
> Maneesh
>

Hi Maneesh,

Good memory. You are correct, this oops is new. Reverting the
sysfs-pin-kobjects patch fixes it for me. I could reliably (like last
time that is) trigger an oops during rsync.

I'm running 2.6.2-rc1 with the -mm3 patchset and the mppe/mppc patches
located at: http://www.polbox.com/h/hs001/
No -mjb patches are used.

I did patch 2.6.1-mm3 with the debug patch you sent earlier, but alas,
it didn't BUG on me when I did the rsync with it. I don't believe that I
gave it much opportunity though, because -mm4 was released shortly
thereafter. Unfortunately, the BUG was nasty, in that it completely
stopped the system forcing an emergency sync, unmount reboot situation,
so I was anxious to get away from it ;)

-Walt

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

\
 
 \ /
  Last update: 2005-03-22 14:00    [W:0.045 / U:0.224 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site