From: Mike Snitzer <snitzer@redhat.com>
To: Tejun Heo <tj@kernel.org>
Cc: jaxboe@fusionio.com, k-ueda@ct.jp.nec.com,
j-nomura@ce.jp.nec.com, jamie@shareable.org,
linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-raid@vger.kernel.org, hch@lst.de
Subject: Re: [PATCH 4/5] dm: implement REQ_FLUSH/FUA support for request-based dm
Date: Mon, 30 Aug 2010 09:28:36 -0400 [thread overview]
Message-ID: <20100830132836.GB5283@redhat.com> (raw)
In-Reply-To: <1283162296-13650-5-git-send-email-tj@kernel.org>
On Mon, Aug 30 2010 at 5:58am -0400,
Tejun Heo <tj@kernel.org> wrote:
> This patch converts request-based dm to support the new REQ_FLUSH/FUA.
...
> This patch rips out special flush code path and deals handles
> REQ_FLUSH/FUA requests the same way as other requests. The only
> special treatment is that REQ_FLUSH requests use the block address 0
> when finding target, which is enough for now.
Looks very comparable to the patch I prepared but I have 2 observations
below (based on my findings from testing my patch).
> @@ -1562,22 +1483,15 @@ static int setup_clone(struct request *clone, struct request *rq,
> {
> int r;
>
> - if (dm_rq_is_flush_request(rq)) {
> - blk_rq_init(NULL, clone);
> - clone->cmd_type = REQ_TYPE_FS;
> - clone->cmd_flags |= (REQ_HARDBARRIER | WRITE);
> - } else {
> - r = blk_rq_prep_clone(clone, rq, tio->md->bs, GFP_ATOMIC,
> - dm_rq_bio_constructor, tio);
> - if (r)
> - return r;
> -
> - clone->cmd = rq->cmd;
> - clone->cmd_len = rq->cmd_len;
> - clone->sense = rq->sense;
> - clone->buffer = rq->buffer;
> - }
> + r = blk_rq_prep_clone(clone, rq, tio->md->bs, GFP_ATOMIC,
> + dm_rq_bio_constructor, tio);
> + if (r)
> + return r;
>
> + clone->cmd = rq->cmd;
> + clone->cmd_len = rq->cmd_len;
> + clone->sense = rq->sense;
> + clone->buffer = rq->buffer;
> clone->end_io = end_clone_request;
> clone->end_io_data = tio;
blk_rq_prep_clone() of a REQ_FLUSH request will result in a
rq_data_dir(clone) of read.
I still had the following:
if (rq->cmd_flags & REQ_FLUSH) {
blk_rq_init(NULL, clone);
clone->cmd_type = REQ_TYPE_FS;
/* without this the clone has a rq_data_dir of 0 */
clone->cmd_flags |= WRITE_FLUSH;
} else {
r = blk_rq_prep_clone(clone, rq, tio->md->bs, GFP_ATOMIC,
dm_rq_bio_constructor, tio);
...
Request-based DM's REQ_FLUSH still works without this special casing but
I figured I'd raise this to ask: what is the proper rq_data_dir() is for
a REQ_FLUSH?
> @@ -1709,15 +1621,12 @@ static void dm_request_fn(struct request_queue *q)
> if (!rq)
> goto plug_and_out;
>
> - if (unlikely(dm_rq_is_flush_request(rq))) {
> - BUG_ON(md->flush_request);
> - md->flush_request = rq;
> - blk_start_request(rq);
> - queue_work(md->wq, &md->barrier_work);
> - goto out;
> - }
> + /* always use block 0 to find the target for flushes for now */
> + pos = 0;
> + if (!(rq->cmd_flags & REQ_FLUSH))
> + pos = blk_rq_pos(rq);
>
> - ti = dm_table_find_target(map, blk_rq_pos(rq));
> + ti = dm_table_find_target(map, pos);
I added the following here: BUG_ON(!dm_target_is_valid(ti));
> if (ti->type->busy && ti->type->busy(ti))
> goto plug_and_out;
I also needed to avoid the ->busy call for REQ_FLUSH:
if (!(rq->cmd_flags & REQ_FLUSH)) {
ti = dm_table_find_target(map, blk_rq_pos(rq));
BUG_ON(!dm_target_is_valid(ti));
if (ti->type->busy && ti->type->busy(ti))
goto plug_and_out;
} else {
/* rq-based only ever has one target! leverage this for FLUSH */
ti = dm_table_get_target(map, 0);
}
If I allowed ->busy to be called for REQ_FLUSH it would result in a
deadlock. I haven't identified where/why yet.
Other than these remaining issues this patch looks good.
Thanks,
Mike
next prev parent reply other threads:[~2010-08-30 13:28 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-08-30 9:58 [PATCHSET 2.6.36-rc2] block, dm: finish REQ_FLUSH/FUA conversion, take#2 Tejun Heo
2010-08-30 9:58 ` [PATCH 1/5] block: make __blk_rq_prep_clone() copy most command flags Tejun Heo
2010-09-01 15:30 ` Christoph Hellwig
2010-08-30 9:58 ` [PATCH 2/5] dm: implement REQ_FLUSH/FUA support for bio-based dm Tejun Heo
2010-09-01 13:43 ` Mike Snitzer
2010-09-01 13:50 ` Tejun Heo
2010-09-01 13:54 ` Mike Snitzer
2010-09-01 13:56 ` Tejun Heo
2010-08-30 9:58 ` [PATCH 3/5] dm: relax ordering of bio-based flush implementation Tejun Heo
2010-09-01 13:51 ` Mike Snitzer
2010-09-01 13:56 ` Tejun Heo
2010-09-03 6:04 ` Kiyoshi Ueda
2010-09-03 9:42 ` Tejun Heo
2010-08-30 9:58 ` [PATCH 4/5] dm: implement REQ_FLUSH/FUA support for request-based dm Tejun Heo
2010-08-30 13:28 ` Mike Snitzer [this message]
2010-08-30 13:59 ` Tejun Heo
2010-08-30 15:07 ` Tejun Heo
2010-08-30 19:08 ` Mike Snitzer
2010-08-30 21:28 ` Mike Snitzer
2010-08-31 10:29 ` Tejun Heo
2010-08-31 13:02 ` Mike Snitzer
2010-08-31 13:14 ` Tejun Heo
2010-08-30 15:42 ` [PATCH] block: initialize flush request with WRITE_FLUSH instead of REQ_FLUSH Tejun Heo
2010-08-30 15:45 ` [PATCH UPDATED 4/5] dm: implement REQ_FLUSH/FUA support for request-based dm Tejun Heo
2010-08-30 19:18 ` Mike Snitzer
2010-09-01 7:15 ` Kiyoshi Ueda
2010-09-01 12:25 ` Mike Snitzer
2010-09-02 13:22 ` Tejun Heo
2010-09-02 13:32 ` Tejun Heo
2010-09-03 5:46 ` Kiyoshi Ueda
2010-09-02 17:43 ` [PATCH] block: make sure FSEQ_DATA request has the same rq_disk as the original Tejun Heo
2010-09-03 5:47 ` Kiyoshi Ueda
2010-09-03 9:33 ` Tejun Heo
2010-09-03 10:28 ` Kiyoshi Ueda
2010-09-03 11:42 ` Tejun Heo
2010-09-03 11:51 ` Kiyoshi Ueda
2010-08-30 9:58 ` [PATCH 5/5] block: remove the WRITE_BARRIER flag Tejun Heo
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=20100830132836.GB5283@redhat.com \
--to=snitzer@redhat.com \
--cc=hch@lst.de \
--cc=j-nomura@ce.jp.nec.com \
--cc=jamie@shareable.org \
--cc=jaxboe@fusionio.com \
--cc=k-ueda@ct.jp.nec.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-raid@vger.kernel.org \
--cc=tj@kernel.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).