From: "Martin K. Petersen" <martin.petersen@oracle.com>
To: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Christoph Hellwig <hch@lst.de>,
Jens Axboe <jens.axboe@oracle.com>,
Matthew Wilcox <matthew@wil.cx>,
linux-mm@kvack.org, linux-scsi@vger.kernel.org
Subject: Re: [PATCH] mm: don't discard unused swap slots by default
Date: Mon, 30 Nov 2009 13:58:39 -0500 [thread overview]
Message-ID: <yq1aay3261s.fsf@sermon.lab.mkp.net> (raw)
In-Reply-To: <Pine.LNX.4.64.0911301752070.10043@sister.anvils> (Hugh Dickins's message of "Mon, 30 Nov 2009 18:28:41 +0000 (GMT)")
>>>>> "Hugh" == Hugh Dickins <hugh.dickins@tiscali.co.uk> writes:
Hugh> You mentioned an "-o discard" mount option before: so I think what
Hugh> we ought to be doing is an option to swapon. But you can imagine
Hugh> that I'd prefer to avoid that too, if we can work this out without
Hugh> it.
The main problem we have is that the devices currently supporting TRIM
are doing a piss poor job at it.
We have pretty good vendor guarantees that discards are going to be
essentially free on SCSI-class hardware. But in the ATA space things
are currently being driven by early adopters / tweakers that care more
about benchmarketing and feature checklists. Whether things actually
work as intended is mostly irrelevant.
I think we'll need to give things a little bit of time for decent ATA
TRIM implementations to materialize. And then we can switch to an
"assume it works, blacklist bad eggs" approach. Until then I think we
need to make discard an explicit opt-in feature.
--
Martin K. Petersen Oracle Linux Engineering
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2009-11-30 18:58 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-30 6:51 unconditional discard calls in the swap code Christoph Hellwig
2009-10-30 17:26 ` Hugh Dickins
2009-10-30 17:26 ` Hugh Dickins
2009-11-18 17:12 ` Christoph Hellwig
2009-11-30 17:22 ` [PATCH] mm: don't discard unused swap slots by default Christoph Hellwig
2009-11-30 18:28 ` Hugh Dickins
2009-11-30 18:58 ` Martin K. Petersen [this message]
2009-12-31 0:33 ` Hugh Dickins
2009-12-31 0:33 ` Hugh Dickins
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=yq1aay3261s.fsf@sermon.lab.mkp.net \
--to=martin.petersen@oracle.com \
--cc=hch@lst.de \
--cc=hugh.dickins@tiscali.co.uk \
--cc=jens.axboe@oracle.com \
--cc=linux-mm@kvack.org \
--cc=linux-scsi@vger.kernel.org \
--cc=matthew@wil.cx \
/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.