mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Tejun Heo <tj@kernel.org>
To: Christoph Hellwig <hch@infradead.org>
Cc: Tao Cui <cui.tao@linux.dev>, Josef Bacik <josef@toxicpanda.com>,
	josef@toxicopanda.com, Jens Axboe <axboe@kernel.dk>,
	cgroups@vger.kernel.org, linux-block@vger.kernel.org,
	linux-kernel@vger.kernel.org, cuitao@kylinos.cn
Subject: Re: [PATCH v2 1/4] blk-iocost: charge flushes as pageless random writes
Date: Mon, 21 Sep 2026 06:58:03 -1000	[thread overview]
Message-ID: <arFiG4sJJGjyoHT3@slm.duckdns.org> (raw)
In-Reply-To: <arDXYwNm7rGqB7pg@infradead.org>

On Mon, Sep 21, 2026 at 12:06:11AM -0700, Christoph Hellwig wrote:
> On Sat, Sep 19, 2026 at 04:59:51AM -1000, Tejun Heo wrote:
> > Hello, Tao.
> > Pricing a flush as a random write, or as a page where that comes out to
> > zero, is as arbitrary as pricing it at zero,
> 
> No, it absolutely is not.  For any flush that actually makes here it
> is extremely expensive, and 0 is the wrongest possible value.  Even
> batching does not change that, as batching is not actually effective
> in any normal file system workload.

I see. I don't think the default value matters all that much given how crude
a tool the default model is. We can set it to a page. I don't think we
should impose it on existing users that are configuring custom models.

Thanks.

-- 
tejun

  reply	other threads:[~2026-09-21 16:58 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-16  8:53 [PATCH v2 0/4] blk-iocost: charge flushes and zone appends Tao Cui
2026-09-16  8:53 ` [PATCH v2 1/4] blk-iocost: charge flushes as pageless random writes Tao Cui
2026-09-18 13:21   ` Christoph Hellwig
2026-09-19 14:59   ` Tejun Heo
2026-09-20  7:08     ` Tao Cui
2026-09-20 12:19       ` Tejun Heo
2026-09-21  7:06     ` Christoph Hellwig
2026-09-21 16:58       ` Tejun Heo [this message]
2026-09-16  8:53 ` [PATCH v2 2/4] blk-iocost: charge zone appends as page-counted sequential writes Tao Cui
2026-09-18 13:22   ` Christoph Hellwig
2026-09-20  7:11     ` Tao Cui
2026-09-19 15:00   ` Tejun Heo
2026-09-20  7:13     ` Tao Cui
2026-09-16  8:53 ` [PATCH v2 3/4] blk-iocost: account zone append completions in latency stats Tao Cui
2026-09-18 13:22   ` Christoph Hellwig
2026-09-16  8:53 ` [PATCH v2 4/4] blk-iocost: fix stale comment in ioc_rqos_throttle() Tao Cui
2026-09-18 13:23   ` Christoph Hellwig

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=arFiG4sJJGjyoHT3@slm.duckdns.org \
    --to=tj@kernel.org \
    --cc=axboe@kernel.dk \
    --cc=cgroups@vger.kernel.org \
    --cc=cui.tao@linux.dev \
    --cc=cuitao@kylinos.cn \
    --cc=hch@infradead.org \
    --cc=josef@toxicopanda.com \
    --cc=josef@toxicpanda.com \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-kernel@vger.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
all inboxes | Powered by JetHome®