linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@fb.com>
To: <linux-kernel@vger.kernel.org>
Cc: <snitzer@redhat.com>, <hch@lst.de>
Subject: [PATCH 0/3 v2] Configurable max discard size
Date: Tue, 14 Jul 2015 12:48:20 -0600	[thread overview]
Message-ID: <1436899703-31966-1-git-send-email-axboe@fb.com> (raw)

Hi,

Most drivers use UINT_MAX (or some variant thereof) for max discard
size, since they don't have a real limit for a non-data transferring
command. This is fine from a throughput point of view, but for a lot
of devices (all?), it truly sucks on latency. We've seen cases of
hundreds of msec in latencies for reads/writes when deleting files
on an fs with discard enabled. For the problematic devices that we
have tested, artificially limiting the size of the discards issued
brings it down to a more manageable 1-2ms max latencies.

Changes since v1:

- Keep separate sysfs files for hw and sw max discard values.

- Add patch 3, defaulting to 64MB discard sizes.

-- 
Jens Axboe



             reply	other threads:[~2015-07-14 18:48 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-14 18:48 Jens Axboe [this message]
2015-07-14 18:48 ` [PATCH 1/3] block: have drivers use blk_queue_max_discard_sectors() Jens Axboe
2015-07-16 14:46   ` Jeff Moyer
2015-07-14 18:48 ` [PATCH 2/3] block: make /sys/block/<dev>/queue/discard_max_bytes writeable Jens Axboe
2015-07-16 15:07   ` Jeff Moyer
2015-07-16 15:11     ` Jens Axboe
2015-07-14 18:48 ` [PATCH 3/3] block: by default, limit maximum discard size to 64MB Jens Axboe
2015-07-14 20:44   ` Mike Snitzer
2015-07-14 20:45     ` Jens Axboe
2015-07-14 21:48       ` Jens Axboe
2015-07-15 11:46         ` Austin S Hemmelgarn
2015-07-15 15:30           ` Jens Axboe
2015-07-15 16:29             ` Mike Snitzer
2015-07-15 22:14               ` Jens Axboe
2015-08-05 13:34                 ` Mike Snitzer

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=1436899703-31966-1-git-send-email-axboe@fb.com \
    --to=axboe@fb.com \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=snitzer@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).