Re: [PATCH 2/3] xfs: evict all inodes involved with log redo item recovery

["Darrick J. Wong" <darrick.wong@oracle.com>]

On Fri, Aug 11, 2017 at 04:22:03AM -0700, Christoph Hellwig wrote:
> On Thu, Aug 10, 2017 at 10:18:51AM -0700, Darrick J. Wong wrote:
> > I felt there's less messing around with vfs internals by sticking to
> > igrab and iput (and letting them deal with the internal inode state)
> > since that's what xfs uses elsewhere to manage inode life cycles.
> 
> I'd feel much safer calling evict_inodes rather than reimplementing
> it poorly.  In fact we might want to just call it in the VFS after
> a failed mount?

I don't think we can call it in the vfs after a failed mount since by
that time we've pulled down all the fs-specific stuff by then, right?

In any case I think it's only xfs that needs this, since (afaict) the
ones I looked at don't set MS_ACTIVE in fill_super and have a small
set of inodes that they explicitly iget and iput.

> Either way please also add Al to the Cc for the next version.

Ok.

--D

> --
> To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html