From: Chris Mason <clm@fb.com>
To: Andreas Dilger <adilger@dilger.ca>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Gregory Farnum <greg@gregs42.com>,
Eric Sandeen <sandeen@redhat.com>,
"Theodore Ts'o" <tytso@mit.edu>,
"Darrick J. Wong" <darrick.wong@oracle.com>,
Dave Chinner <david@fromorbit.com>,
Ric Wheeler <rwheeler@redhat.com>,
Andy Lutomirski <luto@amacapital.net>,
One Thousand Gnomes <gnomes@lxorguk.ukuu.org.uk>,
Martin Petersen <martin.petersen@oracle.com>,
Christoph Hellwig <hch@infradead.org>,
Jens Axboe <axboe@kernel.dk>,
Andrew Morton <akpm@linux-foundation.org>,
Linux API <linux-api@vger.kernel.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
<shane.seymour@hpe.com>, Bruce Fields <bfields@fieldses.org>,
linux-fsdevel <linux-fsdevel@vger.kernel.org>,
Jeff Layton <jlayton@poochiereds.net>
Subject: Re: [PATCH 2/2] block: create ioctl to discard-or-zeroout a range of blocks
Date: Thu, 17 Mar 2016 14:00:18 -0700 [thread overview]
Message-ID: <20160317210018.GA78710@clm-mbp.thefacebook.com> (raw)
In-Reply-To: <819F38A3-51A7-4874-8314-8A6004495716@dilger.ca>
On Thu, Mar 17, 2016 at 02:49:06PM -0600, Andreas Dilger wrote:
> On Mar 17, 2016, at 12:35 PM, Chris Mason <clm@fb.com> wrote:
> >
> > On Thu, Mar 17, 2016 at 10:47:29AM -0700, Linus Torvalds wrote:
> >> On Wed, Mar 16, 2016 at 10:18 PM, Gregory Farnum <greg@gregs42.com> wrote:
> >>>
> >>> So we've not asked for NO_HIDE_STALE on the mailing lists, but I think
> >>> it was one of the problems Sage had using xfs in his BlueStore
> >>> implementation and was a big part of why it moved to pure userspace.
> >>> FileStore might use NO_HIDE_STALE in some places but it would be
> >>> pretty limited. When it came up at Linux FAST we were discussing how
> >>> it and similar things had been problems for us in the past and it
> >>> would've been nice if they were upstream.
> >>
> >> Hmm.
> >>
> >> So to me it really sounds like somebody should cook up a patch, but we
> >> shouldn't put it in the upstream kernel until we get numbers and
> >> actual "yes, we'd use this" from outside of google.
> >
> > We haven't had internal tiers yelling at us for fallocate performance,
> > so I'm unlikely to suggest it, just because its a potential
> > privacy leak we'd have to educate people about. What I'd be more likely
> > to use is code inside the filesystem like this:
> >
> > somefs_fallocate() {
> > if (trim_can_really_zero(my_device)) {
> > trim
> > allocate a regular extent
> > return
> > } else {
> > do normal fallocate
> > }
> > }
>
> We were discussing almost this very same thing in the ext4 concall today.
>
> Ted initially didn't think it was worthwhile to implement, but after looking
> at the whitelist for SATA SSDs it seems that there are enough devices on the
> market that support the ATA_HORKAGE_ZERO_AFTER_TRIM to make this approach
> worthwhile to implement.
We'll end up with people complaining it makes fallocate slower because
of the trims, so it's not a perfect solution. But I much prefer it to
fallocate-stale.
>
> Also, if the ext4 extent size was limited it might even be possible to do
> this efficiently enough with write_same on HDD devices.
>
> > Then the out of tree patch (for google or whoever) becomes a hack to
> > flip trim_can_really_zero on a given block device. The rest of us can
> > use explicit interfaces from the hardware when deciding what we want
> > preallocation to mean.
>
> This might be a bit trickier, since this would affect all zero/trim
> operations, not just ones for uninitialized data extents.
Thinking more, my guess is that google will just keep doing what they
are already doing ;) But there could be a flag in sysfs dedicated to
trim-for-fallocate so admins can see what their devices are reporting.
readonly in mainline, if someone wants to patch it in their large data
center it wouldn't be hard.
-chris
next prev parent reply other threads:[~2016-03-17 21:01 UTC|newest]
Thread overview: 82+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-02 4:09 [PATCH v5.1 0/2] create BLKZEROOUT ioctl that invalidates page cache Darrick J. Wong
2016-03-02 4:09 ` [PATCH 1/2] block: invalidate the page cache when issuing BLKZEROOUT Darrick J. Wong
2016-03-02 9:19 ` Christoph Hellwig
2016-03-02 4:09 ` [PATCH 2/2] block: create ioctl to discard-or-zeroout a range of blocks Darrick J. Wong
2016-03-02 9:20 ` Christoph Hellwig
2016-03-02 18:52 ` Linus Torvalds
2016-03-02 22:56 ` Darrick J. Wong
2016-03-02 23:49 ` Linus Torvalds
2016-03-03 17:02 ` Theodore Ts'o
2016-03-03 17:55 ` Linus Torvalds
2016-03-03 18:00 ` Christoph Hellwig
2016-03-03 18:14 ` Martin K. Petersen
2016-03-03 18:21 ` Theodore Ts'o
2016-03-03 18:01 ` Martin K. Petersen
2016-03-03 18:09 ` Christoph Hellwig
2016-03-03 18:12 ` Darrick J. Wong
2016-03-03 18:54 ` Martin K. Petersen
2016-03-03 22:39 ` Theodore Ts'o
2016-03-03 23:10 ` Dave Chinner
2016-03-04 0:20 ` Theodore Ts'o
2016-03-09 22:20 ` Gregory Farnum
2016-03-09 23:08 ` Theodore Ts'o
2016-03-10 14:58 ` Ric Wheeler
2016-03-10 18:33 ` Linus Torvalds
2016-03-10 21:47 ` Theodore Ts'o
2016-03-11 4:42 ` Ric Wheeler
2016-03-11 13:59 ` One Thousand Gnomes
2016-03-11 15:27 ` Theodore Ts'o
2016-03-11 17:23 ` Linus Torvalds
2016-03-11 17:30 ` Andy Lutomirski
2016-03-11 18:25 ` Linus Torvalds
2016-03-11 22:30 ` Dave Chinner
2016-03-12 0:33 ` Linus Torvalds
2016-03-12 0:35 ` Theodore Ts'o
2016-03-12 0:44 ` Linus Torvalds
2016-03-12 7:19 ` Theodore Ts'o
2016-03-12 10:11 ` Thomas Schoebel-Theuer
2016-03-13 23:30 ` Dave Chinner
2016-03-14 10:34 ` Ric Wheeler
2016-03-14 14:46 ` Theodore Ts'o
2016-03-15 20:14 ` Dave Chinner
2016-03-15 20:43 ` Linus Torvalds
2016-03-15 21:29 ` Theodore Ts'o
2016-03-15 22:33 ` Dave Chinner
2016-03-15 22:52 ` Theodore Ts'o
2016-03-16 1:51 ` Darrick J. Wong
2016-03-16 21:45 ` Andreas Dilger
2016-03-17 0:15 ` Theodore Ts'o
2016-03-17 0:33 ` Eric Sandeen
2016-03-17 0:59 ` Theodore Ts'o
2016-03-17 5:18 ` Gregory Farnum
2016-03-17 12:36 ` Theodore Ts'o
2016-03-17 17:47 ` Linus Torvalds
2016-03-17 17:50 ` Ric Wheeler
2016-03-17 17:59 ` Linus Torvalds
2016-03-17 18:35 ` Chris Mason
2016-03-17 20:49 ` Andreas Dilger
2016-03-17 21:00 ` Chris Mason [this message]
2016-03-18 3:20 ` Theodore Ts'o
2016-03-18 15:15 ` Jeff Moyer
2016-03-18 20:05 ` Martin K. Petersen
2016-03-18 6:52 ` Gregory Farnum
2016-03-18 7:19 ` Linus Torvalds
2016-03-17 1:01 ` Dave Chinner
2016-03-17 2:38 ` Darrick J. Wong
2016-03-18 22:55 ` NeilBrown
2016-03-15 23:06 ` Linus Torvalds
2016-03-15 23:14 ` Linus Torvalds
2016-03-16 0:08 ` Dave Chinner
2016-03-15 23:52 ` Dave Chinner
2016-03-16 0:06 ` Linus Torvalds
2016-03-16 0:30 ` Eric Sandeen
2016-03-16 0:51 ` Chris Mason
2016-03-16 22:23 ` Chris Mason
2016-03-17 13:49 ` Ric Wheeler
2016-03-15 22:38 ` Eric Sandeen
2016-03-03 22:56 ` Dave Chinner
2016-03-04 2:30 ` Thomas Schoebel-Theuer
2016-03-03 18:14 ` Linus Torvalds
2016-03-02 9:15 ` [PATCH v5.1 0/2] create BLKZEROOUT ioctl that invalidates page cache Arnd Bergmann
2016-03-02 9:44 ` Christoph Hellwig
2016-03-02 10:55 ` Arnd Bergmann
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=20160317210018.GA78710@clm-mbp.thefacebook.com \
--to=clm@fb.com \
--cc=adilger@dilger.ca \
--cc=akpm@linux-foundation.org \
--cc=axboe@kernel.dk \
--cc=bfields@fieldses.org \
--cc=darrick.wong@oracle.com \
--cc=david@fromorbit.com \
--cc=gnomes@lxorguk.ukuu.org.uk \
--cc=greg@gregs42.com \
--cc=hch@infradead.org \
--cc=jlayton@poochiereds.net \
--cc=linux-api@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@amacapital.net \
--cc=martin.petersen@oracle.com \
--cc=rwheeler@redhat.com \
--cc=sandeen@redhat.com \
--cc=shane.seymour@hpe.com \
--cc=torvalds@linux-foundation.org \
--cc=tytso@mit.edu \
/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).