Hi Alasdair,
On Mon, 18 Feb 2008 12:34:03 +0000, Alasdair G Kergon wrote:
> On Mon, Feb 18, 2008 at 12:21:48PM +0000, Alasdair G Kergon wrote:
> > On Fri, Feb 15, 2008 at 05:55:05PM -0500, Kiyoshi Ueda wrote:
> > > This patch removes dead codes for the noflush suspend.
> > I've been sent (unfinished) patches for 2.6.26 (or even 2.6.25 if people
> > consider the reinstatement of flushing support important enough) that
> > build on this one.
>
> That was ambiguous:-( I meant that code has only just got added to the
> kernel and there are patches waiting to be finished that use it - it's
> not dead code.
You mean that the DM_WQ_FLUSH_ALL will be used, but the code
in dm_suspend() below won't be used, right?
If so, I'd like to remove only that part, since nobody won't run
the 'flush_and_out' part.
Is it OK for you?
> @@ -1472,7 +1468,7 @@ int dm_suspend(struct mapped_device *md,
> if (!md->suspended_bdev) {
> DMWARN("bdget failed in dm_suspend");
> r = -ENOMEM;
> - goto flush_and_out;
> + goto out;
> }
>
> /*
> @@ -1523,14 +1519,6 @@ int dm_suspend(struct mapped_device *md,
>
> set_bit(DMF_SUSPENDED, &md->flags);
>
> -flush_and_out:
> - if (r && noflush)
> - /*
> - * Because there may be already I/Os in the pushback list,
> - * flush them before return.
> - */
> - dm_queue_flush(md, DM_WQ_FLUSH_ALL, NULL);
> -
> out:
> if (r && md->suspended_bdev) {
> bdput(md->suspended_bdev);
Thanks,
Kiyoshi Ueda
--
dm-devel mailing list
dm-devel@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/dm-devel
