On Wed, Mar 08, 2017 at 09:07:01AM -0800, Darrick J. Wong wrote:
> > + if (WARN_ON_ONCE(args.fsbno == NULLFSBLOCK)) {
> > + xfs_iroot_realloc(ip, -1, whichfork);
> > + xfs_btree_del_cursor(cur, XFS_BTREE_ERROR);
> > + return -ENOMEM;
>
> ENOSPC?
Yeah.
> > + }
> > /*
> > * Allocation can't fail, the space was reserved.
> > */
>
> Can we get rid of the ASSERT(args.fsbno != NULLFSBLOCK); just below here
> now that we jump out above?
Sure.
>
> Conceptually I guess it's ok for now until we separate out the uses of
> *firstblock to stay ahead of locking rules vs. *firstblock to remap
> things. Hmm, I'll try to make a first stab at that today.
I've been working on that for a while - the problem is that it goes
up a few layers, including xfs_bmapi_write and the da_args structures.
Give me a little more time and I should have a series for you.
--
To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
