From: Mike Snitzer <snitzer@redhat.com>
To: Damien Le Moal <damien.lemoal@wdc.com>
Cc: Christoph Hellwig <hch@infradead.org>,
Bart Van Assche <bart.vanassche@sandisk.com>,
dm-devel@redhat.com, Mikulas Patocka <mpatocka@redhat.com>,
Alasdair Kergon <agk@redhat.com>
Subject: Re: dm-zoned: Avoid metadata flush writeback throttling
Date: Mon, 11 Sep 2017 11:05:55 -0400 [thread overview]
Message-ID: <20170911150555.GA15818@redhat.com> (raw)
In-Reply-To: <0ee663a1-f497-1d10-f252-91fee072322a@wdc.com>
On Mon, Jul 24 2017 at 5:16am -0400,
Damien Le Moal <damien.lemoal@wdc.com> wrote:
>
>
> On 7/24/17 17:03, Christoph Hellwig wrote:
> > On Mon, Jul 24, 2017 at 04:44:22PM +0900, Damien Le Moal wrote:
> >> Avoid metadata flush to be blocked by the writeback throttling code in
> >> wbt_wait(). Otherwise, we can end up with a deadlock under heavy write
> >> load with all chunk works active. In such situation, a deadlock can
> >> happen if the flush work is blocked by the throttling code while holding
> >> the metadata lock: as the chunk works will wait for the metadata lock
> >> too, no progress can be made.
> >>
> >> Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
> >> ---
> >> drivers/md/dm-zoned-metadata.c | 3 ++-
> >> 1 file changed, 2 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/drivers/md/dm-zoned-metadata.c b/drivers/md/dm-zoned-metadata.c
> >> index 884ff7c170a0..f694fb98b002 100644
> >> --- a/drivers/md/dm-zoned-metadata.c
> >> +++ b/drivers/md/dm-zoned-metadata.c
> >> @@ -567,7 +567,8 @@ static void dmz_write_mblock(struct dmz_metadata *zmd, struct dmz_mblock *mblk,
> >> bio->bi_bdev = zmd->dev->bdev;
> >> bio->bi_private = mblk;
> >> bio->bi_end_io = dmz_mblock_bio_end_io;
> >> - bio_set_op_attrs(bio, REQ_OP_WRITE, REQ_META | REQ_PRIO);
> >> + bio_set_op_attrs(bio, REQ_OP_WRITE,
> >> + REQ_META | REQ_PRIO | REQ_SYNC | REQ_IDLE);
> >
> > Please just assign bi_opf directly instea dof using bio_set_op_attrs
> > in new code.
>
> OK. Will do.
>
> > Also what do you need REQ_IDLE for?
>
> It is not really needed here, but the wbt_should_throttle() code test is:
>
> if ((bio->bi_opf & (REQ_SYNC | REQ_IDLE)) == (REQ_SYNC | REQ_IDLE))
> return false;
>
> So if REQ_IDLE is not set, wbt will trigger.
>
> Having a req flag that says "no wb throttling please" would be a lot
> cleaner...
Did you have a new version of this patch?
Thanks,
Mike
next prev parent reply other threads:[~2017-09-11 15:05 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-07-24 7:44 [PATCH] dm-zoned: Avoid metadata flush writeback throttling Damien Le Moal
2017-07-24 8:03 ` Christoph Hellwig
2017-07-24 9:16 ` Damien Le Moal
2017-09-11 15:05 ` Mike Snitzer [this message]
2017-09-12 5:46 ` Damien Le Moal
2017-09-12 6:12 ` Mikulas Patocka
2017-09-12 8:01 ` Damien Le Moal
2017-09-12 10:32 ` Mikulas Patocka
2017-09-13 0:38 ` Damien Le Moal
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170911150555.GA15818@redhat.com \
--to=snitzer@redhat.com \
--cc=agk@redhat.com \
--cc=bart.vanassche@sandisk.com \
--cc=damien.lemoal@wdc.com \
--cc=dm-devel@redhat.com \
--cc=hch@infradead.org \
--cc=mpatocka@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.