From: Bart Van Assche <bvanassche@acm.org>
To: Damien Le Moal <Damien.LeMoal@wdc.com>, Jens Axboe <axboe@kernel.dk>
Cc: "linux-block@vger.kernel.org" <linux-block@vger.kernel.org>,
Christoph Hellwig <hch@lst.de>, Jaegeuk Kim <jaegeuk@kernel.org>,
Adam Manzanares <Adam.Manzanares@wdc.com>,
Ming Lei <ming.lei@redhat.com>
Subject: Re: [PATCH 8/9] block/mq-deadline: Add I/O priority support
Date: Thu, 27 May 2021 13:12:12 -0700 [thread overview]
Message-ID: <d1cf99a8-df7a-f33d-cee0-2906813256ce@acm.org> (raw)
In-Reply-To: <DM6PR04MB7081661E924716B91E4E0899E7239@DM6PR04MB7081.namprd04.prod.outlook.com>
On 5/26/21 8:48 PM, Damien Le Moal wrote:
> On 2021/05/27 10:02, Bart Van Assche wrote:
>> + if (next->elv.priv[0]) {
>> + atomic_inc(&dd->merged[prio]);
>> + } else {
>> + WARN_ON_ONCE(true);
>> + }
>
> I do not think you need the curly braces here.
These are present because patch 9/9 adds more code in this if-statement
and to improve readability of patch 9/9.
>> @@ -392,9 +453,14 @@ static struct request *dd_dispatch_request(struct blk_mq_hw_ctx *hctx)
>> {
>> struct deadline_data *dd = hctx->queue->elevator->elevator_data;
>> struct request *rq;
>> + enum dd_prio prio;
>>
>> spin_lock(&dd->lock);
>> - rq = __dd_dispatch_request(dd);
>> + for (prio = 0; prio <= DD_PRIO_MAX; prio++) {
>> + rq = __dd_dispatch_request(dd, prio);
>> + if (rq || dd_queued(dd, prio))
>> + break;
>> + }
>> spin_unlock(&dd->lock);
>
> Unless I missed something, this means that the aging (fifo list expire)
> mechanism is per prio list but there is no aging between the prio classes. This
> means that an application doing lots of RT direct IOs will completely starve
> other prio classes and hang the applications using them.
>
> I think we need aging between the prio classes too to avoid that.
OK, I will add an aging mechanism.
Bart.
next prev parent reply other threads:[~2021-05-27 20:12 UTC|newest]
Thread overview: 65+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-27 1:01 [PATCH 0/9] Improve I/O priority support Bart Van Assche
2021-05-27 1:01 ` [PATCH 1/9] block/mq-deadline: Add several comments Bart Van Assche
2021-05-27 3:03 ` Damien Le Moal
2021-05-27 6:45 ` Hannes Reinecke
2021-05-27 19:30 ` Bart Van Assche
2021-05-27 8:43 ` Johannes Thumshirn
2021-05-27 15:13 ` Himanshu Madhani
2021-05-27 1:01 ` [PATCH 2/9] block/mq-deadline: Add two lockdep_assert_held() statements Bart Van Assche
2021-05-27 2:25 ` Chaitanya Kulkarni
2021-05-27 3:09 ` Damien Le Moal
2021-05-27 6:46 ` Hannes Reinecke
2021-05-27 8:44 ` Johannes Thumshirn
2021-05-27 15:14 ` Himanshu Madhani
2021-05-27 1:01 ` [PATCH 3/9] block/mq-deadline: Remove two local variables Bart Van Assche
2021-05-27 2:26 ` Chaitanya Kulkarni
2021-05-27 3:11 ` Damien Le Moal
2021-05-27 6:46 ` Hannes Reinecke
2021-05-27 8:44 ` Johannes Thumshirn
2021-05-27 15:15 ` Himanshu Madhani
2021-05-27 1:01 ` [PATCH 4/9] block/mq-deadline: Rename dd_init_queue() and dd_exit_queue() Bart Van Assche
2021-05-27 2:27 ` Chaitanya Kulkarni
2021-05-27 3:13 ` Damien Le Moal
2021-05-27 19:33 ` Bart Van Assche
2021-05-27 6:47 ` Hannes Reinecke
2021-05-27 8:44 ` Johannes Thumshirn
2021-05-27 15:16 ` Himanshu Madhani
2021-05-27 1:01 ` [PATCH 5/9] block/mq-deadline: Improve compile-time argument checking Bart Van Assche
2021-05-27 2:28 ` Chaitanya Kulkarni
2021-05-27 3:24 ` Damien Le Moal
2021-05-27 19:38 ` Bart Van Assche
2021-05-28 1:42 ` Damien Le Moal
2021-05-27 6:48 ` Hannes Reinecke
2021-05-27 8:49 ` Johannes Thumshirn
2021-05-27 15:19 ` Himanshu Madhani
2021-05-27 1:01 ` [PATCH 6/9] block/mq-deadline: Reduce the read expiry time for non-rotational media Bart Van Assche
2021-05-27 2:30 ` Chaitanya Kulkarni
2021-05-27 3:27 ` Damien Le Moal
2021-05-27 19:43 ` Bart Van Assche
2021-05-27 6:52 ` Hannes Reinecke
2021-05-27 15:20 ` Himanshu Madhani
2021-05-27 1:01 ` [PATCH 7/9] block/mq-deadline: Reserve 25% of tags for synchronous requests Bart Van Assche
2021-05-27 3:33 ` Damien Le Moal
2021-05-27 20:00 ` Bart Van Assche
2021-05-27 6:54 ` Hannes Reinecke
2021-05-27 19:55 ` Bart Van Assche
2021-05-27 1:01 ` [PATCH 8/9] block/mq-deadline: Add I/O priority support Bart Van Assche
2021-05-27 3:48 ` Damien Le Moal
2021-05-27 20:12 ` Bart Van Assche [this message]
2021-05-27 7:07 ` Hannes Reinecke
2021-05-27 20:23 ` Bart Van Assche
2021-05-28 1:40 ` Damien Le Moal
2021-05-27 1:01 ` [PATCH 9/9] block/mq-deadline: Add cgroup support Bart Van Assche
2021-05-27 7:09 ` Hannes Reinecke
2021-05-27 6:25 ` [PATCH 0/9] Improve I/O priority support Wang Jianchao
2021-05-27 8:05 ` Wang Jianchao
2021-05-27 18:40 ` Bart Van Assche
2021-05-28 2:05 ` Wang Jianchao
2021-05-28 8:43 ` Paolo Valente
2021-05-28 16:28 ` Bart Van Assche
2021-05-27 8:56 ` Johannes Thumshirn
2021-05-27 17:23 ` Chaitanya Kulkarni
2021-05-27 19:00 ` Bart Van Assche
2021-05-27 17:58 ` Adam Manzanares
2021-05-27 18:53 ` Bart Van Assche
2021-05-27 22:45 ` Adam Manzanares
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=d1cf99a8-df7a-f33d-cee0-2906813256ce@acm.org \
--to=bvanassche@acm.org \
--cc=Adam.Manzanares@wdc.com \
--cc=Damien.LeMoal@wdc.com \
--cc=axboe@kernel.dk \
--cc=hch@lst.de \
--cc=jaegeuk@kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=ming.lei@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 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).