On Fri, 5 May 2000, Alexander Viro wrote:
> On Fri, 5 May 2000, Tigran Aivazian wrote:
>
> > Hi Alexander,
> >
> > I am almost sure that you are aware of this fact but in the unlikely case
> > you are not - the mnt->mnt_count behaves incorrectly which causes
> > filesystems to be not-umountable when they should be. E.g. if you mount a
> > ramfs filesystem and then do a bit of io on it (chdir, creat, readdir,
> > unlink) so that at the end the state of the filesystem should be exactly
> > the same as in the beginning, it will fail to umount because
> > mnt->mnt_count > 2 (it is 3 actually).
> >
> > I will be finding out what causes it next.
>
> Can you reproduce it _not_ on a ramfs?
actually I first reproduced it on nullfs (and then on ramfs) which is only
1 hour old so it is too raw to be shown yet.
I need to methodically study all causes of this so, although it will be
very slow, I promise I will get to the bottom of it (I can't proceed with
my work unless I figure this one out). I will let you know all I find.
Regards,
Tigran
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun May 07 2000 - 21:00:17 EST