From: Jens Axboe <jens.axboe@oracle.com>
To: Vasily Tarasov <vtaras@openvz.org>
Cc: LKML <linux-kernel@vger.kernel.org>, OVZDL <devel@openvz.org>
Subject: Re: [PATCH] cfq: get rid of cfqq hash
Date: Wed, 25 Apr 2007 12:27:49 +0200 [thread overview]
Message-ID: <20070425102748.GT9715@kernel.dk> (raw)
In-Reply-To: <20070425075725.GK9715@kernel.dk>
On Wed, Apr 25 2007, Jens Axboe wrote:
> On Wed, Apr 25 2007, Vasily Tarasov wrote:
> > >> @@ -1806,7 +1765,11 @@ static int cfq_may_queue(request_queue_t
> > >> * so just lookup a possibly existing queue, or return 'may queue'
> > >> * if that fails
> > >> */
> > >> - cfqq = cfq_find_cfq_hash(cfqd, key, tsk->ioprio);
> > >> + cic = cfq_get_io_context_noalloc(cfqd, tsk);
> > >> + if (!cic)
> > >> + return ELV_MQUEUE_MAY;
> > >> +
> > >> + cfqq = cic->cfqq[rw & REQ_RW_SYNC];
> > >> if (cfqq) {
> > >> cfq_init_prio_data(cfqq);
> > >> cfq_prio_boost(cfqq);
> > >
> > > Ahem, how well did you test this patch?
> >
> > Ugh, again: bio_sync() returns not only 0/1
> > Sorry for giving so much trouble...
>
> Right, and REQ_RW_SYNC isn't 1 either, so it returns a large number if
> set.
This is also needed, you can't just rely on bio_sync(), a check for a
READ is needed as well.
diff --git a/block/cfq-iosched.c b/block/cfq-iosched.c
index 17e4660..b793817 100644
--- a/block/cfq-iosched.c
+++ b/block/cfq-iosched.c
@@ -215,6 +215,18 @@ static inline void cic_set_cfqq(struct cfq_io_context *cic,
}
/*
+ * We regard a request as SYNC, if it's either a read or has the SYNC bit
+ * set (in which case it could also be direct WRITE).
+ */
+static inline int cfq_bio_sync(struct bio *bio)
+{
+ if (bio_data_dir(bio) == READ || bio_sync(bio))
+ return 1;
+
+ return 0;
+}
+
+/*
* scheduler run of queue, if there are requests pending and no one in the
* driver that will restart queueing
*/
@@ -594,7 +606,7 @@ cfq_find_rq_fmerge(struct cfq_data *cfqd, struct bio *bio)
if (!cic)
return NULL;
- cfqq = cic_to_cfqq(cic, bio_sync(bio));
+ cfqq = cic_to_cfqq(cic, cfq_bio_sync(bio));
if (cfqq) {
sector_t sector = bio->bi_sector + bio_sectors(bio);
@@ -694,7 +706,7 @@ static int cfq_allow_merge(request_queue_t *q, struct request *rq,
/*
* Disallow merge of a sync bio into an async request.
*/
- if ((bio_data_dir(bio) == READ || bio_sync(bio)) && !rq_is_sync(rq))
+ if (cfq_bio_sync(bio) && !rq_is_sync(rq))
return 0;
/*
@@ -705,7 +717,7 @@ static int cfq_allow_merge(request_queue_t *q, struct request *rq,
if (!cic)
return 0;
- cfqq = cic_to_cfqq(cic, bio_sync(bio));
+ cfqq = cic_to_cfqq(cic, cfq_bio_sync(bio));
if (cfqq == RQ_CFQQ(rq))
return 1;
--
Jens Axboe
next prev parent reply other threads:[~2007-04-25 10:28 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-04-24 13:53 [PATCH] cfq: get rid of cfqq hash Vasily Tarasov
2007-04-24 17:02 ` Jens Axboe
2007-04-24 17:22 ` Jens Axboe
2007-04-24 18:29 ` Jens Axboe
2007-04-24 18:33 ` Jens Axboe
2007-04-25 11:31 ` Vasily Tarasov
2007-04-25 6:58 ` Jens Axboe
2007-04-25 7:27 ` Jens Axboe
2007-04-25 11:50 ` Vasily Tarasov
2007-04-25 7:57 ` Jens Axboe
2007-04-25 10:27 ` Jens Axboe [this message]
-- strict thread matches above, loose matches on Subject: below --
2007-04-24 11:53 Vasily Tarasov
2007-04-24 12:27 ` Jens Axboe
2007-04-24 12:51 ` Vasily Tarasov
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=20070425102748.GT9715@kernel.dk \
--to=jens.axboe@oracle.com \
--cc=devel@openvz.org \
--cc=linux-kernel@vger.kernel.org \
--cc=vtaras@openvz.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