Re: [PATCH] iomap: Move page_done callback under the folio lock

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Dec 14, 2022 at 10:07 AM Christoph Hellwig <hch@xxxxxxxxxxxxx> wrote:
> On Tue, Dec 13, 2022 at 12:03:41PM -0800, Darrick J. Wong wrote:
> > On Tue, Dec 13, 2022 at 08:48:33PM +0100, Andreas Gruenbacher wrote:
> > > Hi Darrick,
> > >
> > > I'd like to get the following iomap change into this merge window.  This
> > > only affects gfs2, so I can push it as part of the gfs2 updates if you
> > > don't mind, provided that I'll get your Reviewed-by confirmation.
> > > Otherwise, if you'd prefer to pass this through the xfs tree, could you
> > > please take it?
> >
> > I don't mind you pushing changes to ->page_done through the gfs2 tree,
> > but don't you need to move the other callsite at the bottom of
> > iomap_write_begin?
>
> Yes.

I assume you mean yes to the former, because the ->page_done() call in
iomap_write_begin() really doesn't need to be moved.

> And if we touch this anyway it really should switch to passing
> a folio, which also nicely breaks any in progress code (if there is any)
> and makes them notice the change.

Okay.

> That being said, do you mean 6.2 with "this window"?  Unless the gfs2
> changes are a critical bug fix, I don't think Linux will take them if
> applied after 6.1 was released.

Yes, I really mean the merge window that is currently open.

Thanks,
Andreas




[Index of Archives]     [XFS Filesystem Development (older mail)]     [Linux Filesystem Development]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux RAID]     [Linux SCSI]


  Powered by Linux