From: "Holger Hoffstaette" <holger.hoffstaette@googlemail.com>
To: linux-fsdevel@vger.kernel.org
Subject: 3.7: massive slowdown of deletion performance with discard
Date: Sun, 16 Dec 2012 16:32:11 +0100 [thread overview]
Message-ID: <pan.2012.12.16.15.32.11.1536@googlemail.com> (raw)
Hi,
I just tested 3.7-rc1 and found that deleting a kernel tree on SSDs now
takes approximately way too long to wait for (as in minutes), wherereas
previously this took <1s. As soon as I remount with nodiscard the
performance becomes tolerable again. It alss happens on two different SSDs
(old Intel G2 & Corsair Force GT), both of which have otherwise excellent
performance and show no noticeable impact of discard with kernels 3.6.x.
Filesystem is ext4, but after reading various kernel changelog summaries
it seems to be caused by multiple changes around discard behaviour in
general. Is anyone else noticing this? I find it hard to believe that such
a massive slowdown would go unnoticed..
thanks
Holger
next reply other threads:[~2012-12-16 16:14 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-12-16 15:32 Holger Hoffstaette [this message]
2012-12-16 16:31 ` 3.7: massive slowdown of deletion performance with discard Holger Hoffstaette
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=pan.2012.12.16.15.32.11.1536@googlemail.com \
--to=holger.hoffstaette@googlemail.com \
--cc=linux-fsdevel@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.