From: Mike Snitzer <snitzer@redhat.com>
To: Lukas Czerner <lczerner@redhat.com>
Cc: Christoph Hellwig <hch@infradead.org>,
sandeen@redhat.com, DarkNovaNick@gmail.com, linux-lvm@redhat.com,
linux-ext4@vger.kernel.org, Alasdair G Kergon <agk@redhat.com>,
device-mapper development <dm-devel@redhat.com>
Subject: Re: do not disable ext4 discards on first discard failure? [was: Re: dm snapshot: ignore discards issued to the snapshot-origin target]
Date: Mon, 2 May 2011 08:48:04 -0400 [thread overview]
Message-ID: <20110502124803.GA31034@redhat.com> (raw)
In-Reply-To: <alpine.LFD.2.00.1105021217080.3778@dhcp-27-109.brq.redhat.com>
On Mon, May 02 2011 at 6:24am -0400,
Lukas Czerner <lczerner@redhat.com> wrote:
> On Mon, 2 May 2011, Christoph Hellwig wrote:
>
> > On Mon, May 02, 2011 at 09:13:08AM +0100, Alasdair G Kergon wrote:
> > > On Mon, May 02, 2011 at 09:16:21AM +0200, Lukas Czerner wrote:
> > > > However when we have dm device where part of the device supports
> > > > discard due to underlying hardware capability we just can not return
> > > > EOPNOTSUPP from blkdev_issue_discard, because it is just not true!
> > >
> > > EOPNOTSUPP from dm means the operation was not supported on that *one* bio.
> > > It does *not* tell you anything in general about the device, or whether
> > > you'd get the same error from different bios in future.
> >
> > Exactly. We already have the information in the queue limits to tell
> > the filesystem if discard is supported at all or not.
> >
>
> So I gave it a try. First of all the device composed of SSD and spinning
> disk does export discard_support information properly, however it also
> advertise discard_zeroes_data which is wrong and possibly dangerous and
> should be fixed!
You're effectively advocating that blk_stack_limits() needs to clear the
topmost device's discard_zeroes_data flag if any one bottom device does
not have discard_zeroes_data.
> And second of all, strictly speaking if EOPNOTSUPP from dm means that
> operation was not supported on that *one* bio, blkdev_issue_discard()
> should handle that and do not return EOPNOTSUPP further if queue limits
> tells that device has discard support. Is this acceptable solution for
> you guys ? I can make that change since I am changing blkdev_issue_discard()
> anyway. Or, we can make that change in filesystem where we disable
> discard on mount time, when we notice that discard mount option was
> specified, but the device does not support it (we should probably do
> this regardless on blkdev_issue_discard() change).
The blkdev_issue_discard() change you propose could be fine (mask
EOPNOTSUPP return if device advertises support for discards) -- though
Eric said we shouldn't ever say we did something when we didn't.
But that blkdev_issue_discard() change is really only safe if the
discard_zeroes_data flag is cleared by blk_stack_limits() if finds
inconsistent discard_zeroes_data support.
Mike
next prev parent reply other threads:[~2011-05-02 12:48 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20110412234706.GA11244@redhat.com>
[not found] ` <20cf301cbef67d323104a0c2ff52@google.com>
[not found] ` <20110413224025.GA18589@redhat.com>
[not found] ` <20110413234854.GA19793@redhat.com>
[not found] ` <20110426173213.GA19604@redhat.com>
[not found] ` <20110428001912.GA14659@redhat.com>
[not found] ` <20110428075355.GA2190@infradead.org>
2011-04-28 20:59 ` do not disable ext4 discards on first discard failure? [was: Re: dm snapshot: ignore discards issued to the snapshot-origin target] Mike Snitzer
2011-04-28 21:28 ` Eric Sandeen
2011-04-28 22:59 ` [linux-lvm] " Alasdair G Kergon
2011-04-28 23:01 ` Eric Sandeen
2011-04-28 23:11 ` Alasdair G Kergon
2011-04-29 1:12 ` Andreas Dilger
2011-04-29 9:30 ` Lukas Czerner
2011-04-29 12:24 ` [dm-devel] " Alasdair G Kergon
2011-04-29 12:29 ` Christoph Hellwig
2011-04-29 14:28 ` Eric Sandeen
2011-04-29 15:13 ` [linux-lvm] " Ray Morris
2011-05-04 16:33 ` Ted Ts'o
2011-05-04 16:51 ` Eric Sandeen
2011-05-04 16:57 ` Lukas Czerner
2011-05-04 17:02 ` Lukas Czerner
2011-05-02 7:16 ` Lukas Czerner
2011-05-02 8:13 ` Alasdair G Kergon
2011-05-02 8:19 ` Christoph Hellwig
2011-05-02 10:24 ` Lukas Czerner
2011-05-02 12:48 ` Mike Snitzer [this message]
2011-05-02 13:05 ` Lukas Czerner
2011-05-02 14:47 ` Eric Sandeen
2011-05-02 14:48 ` Christoph Hellwig
2011-05-02 14:58 ` Lukas Czerner
2011-05-02 13:48 ` [dm-devel] " Martin K. Petersen
2011-05-02 14:20 ` Martin K. Petersen
2011-05-02 14:39 ` [dm-devel] " Lukas Czerner
2011-05-02 14:50 ` Martin K. Petersen
2011-05-02 14:58 ` Mike Snitzer
2011-05-02 16:58 ` [dm-devel] " Martin K. Petersen
2011-05-03 8:57 ` Lukas Czerner
2011-05-04 15:10 ` Martin K. Petersen
2011-05-04 16:02 ` Mike Snitzer
2011-05-04 16:50 ` Martin K. Petersen
2011-05-04 18:03 ` Mike Snitzer
2011-05-04 17:10 ` [dm-devel] " Lukas Czerner
2011-05-04 17:32 ` Martin K. Petersen
2011-05-04 17:35 ` Lukas Czerner
2011-05-18 12:16 ` Mike Snitzer
2011-05-18 12:52 ` Mike Snitzer
2011-05-04 15:16 ` [dm-devel] " Martin K. Petersen
2011-05-04 16:12 ` Lukas Czerner
2011-05-05 8:33 ` Karel Zak
2011-05-05 10:48 ` Lukas Czerner
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=20110502124803.GA31034@redhat.com \
--to=snitzer@redhat.com \
--cc=DarkNovaNick@gmail.com \
--cc=agk@redhat.com \
--cc=dm-devel@redhat.com \
--cc=hch@infradead.org \
--cc=lczerner@redhat.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-lvm@redhat.com \
--cc=sandeen@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).