On Thu, Mar 22, 2012 at 05:19:09PM +0200, Kasatkin, Dmitry wrote:
> On Thu, Mar 22, 2012 at 5:09 PM, Al Viro wrote:
> > On Thu, Mar 22, 2012 at 10:53:04AM -0400, Mimi Zohar wrote:
> >> > BTW, you've missed several other places in mm/* doing fput(), so it
> >> > be enough to paper over your problem anyway.
> >> >
> >> > Final fput() *can* happen under mmap_sem. ??Period.
> >> I think I got that loud and clear; otherwise we wouldn't have come up
> >> with deferring the __fput(). ??We have a very real problem here -
> >> extended attributes requires taking the i_mutex.
> > Don't do it, then? ??If you _must_ write to xattr on final fput, I'd
> > starting to figure out if xattr needs its protection to be ->i_mutex -
> > might make sense to introduce a separate mutex for xattr crap. ??Or not
> "Or not" ... How to understand you?
"Or it might not make sense to go that way"
> > not familiar enough with the guts of xattr handling in individual
> > to tell if that would work (e.g. if it would need unpleasant changes to
> > ->setattr() instances)...
IOW, you'll need to do quite a bit of code review to tell if it's a
direction or not - I can't tell without doing the same amount of RTFS; look
for the places where xattrs are modified by fs code, see how far is
acquired, whether xattrs are read in the same section and whether we rely
->i_mutex to keep the xattr values unchanged between two reads or write and
To unsubscribe from this list: send the line "unsubscribe
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html