Re: [PATCH] Useless inode semaphore locking in 2.4.0-test8

From: Alexander Viro (viro@math.psu.edu)
Date: Fri Sep 15 2000 - 07:31:43 EST


On Fri, 15 Sep 2000, Andi Kleen wrote:

> On Fri, Sep 15, 2000 at 02:00:38PM +0200, Eric PAIRE wrote:
> > Hi,
> >
> > In open.c:do_truncate(), the call to notify_change() is protected by
> > the inode->i_sem, which seems to me useless, and thus can be removed.

Look better. "thus can be removed" would go with s/seems to/is provably/
(preferably with proof in bargain). Hint: it _is_ needed.

> > BTW, I also removed the useless inode pointer and error local variables.
>
> Please don't. You opened race country with writes in most file systemsnow.

... and with truncate() itself.

> In reiserfs it is also needed to lock against tail merging.
>
> Also truncate inode locking is needed to get a halfway reliable loopback
> device (unlike the current one)

?
I'm afraid that I've lost you here - what do you mean?

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



This archive was generated by hypermail 2b29 : Fri Sep 15 2000 - 21:00:25 EST