All of lore.kernel.org
 help / color / mirror / Atom feed
From: Josef Bacik <jbacik@fusionio.com>
To: Miao Xie <miaox@cn.fujitsu.com>
Cc: Josef Bacik <JBacik@fusionio.com>,
	"linux-btrfs@vger.kernel.org" <linux-btrfs@vger.kernel.org>
Subject: Re: [PATCH] Btrfs: place ordered operations on a per transaction list
Date: Mon, 18 Feb 2013 10:33:13 -0500	[thread overview]
Message-ID: <20130218153034.GA3188@localhost.localdomain> (raw)
In-Reply-To: <51220EE1.6090607@cn.fujitsu.com>

On Mon, Feb 18, 2013 at 04:22:09AM -0700, Miao Xie wrote:
> On wed, 13 Feb 2013 11:13:22 -0500, Josef Bacik wrote:
> > Miao made the ordered operations stuff run async, which introduced a
> > deadlock where we could get somebody (sync) racing in and committing the
> > transaction while a commit was already happening.  The new committer would
> > try and flush ordered operations which would hang waiting for the commit to
> > finish because it is done asynchronously and no longer inherits the callers
> > trans handle.  To fix this we need to make the ordered operations list a per
> > transaction list.  We can get new inodes added to the ordered operation list
> > by truncating them and then having another process writing to them, so this
> > makes it so that anybody trying to add an ordered operation _must_ start a
> > transaction in order to add itself to the list, which will keep new inodes
> > from getting added to the ordered operations list after we start committing.
> > This should fix the deadlock and also keeps us from doing a lot more work
> > than we need to during commit.  Thanks,
> 
> Firstly, thanks to deal with the bug which was introduced by my patch.
> 
> But comparing with this fix method, I prefer the following one because:
> - we won't worry the similar problem if we add more work during commit
>   in the future.
> - it is unnecessary to get a new handle and commit it if the transaction
>   is under the commit.

Mine has the benefit of not making a committing transaction flush more stuff
that it doesn't need to, so I think I'll keep mine as well, but I agree yours is
good for the attach case as well.  So can you send this along properly with a
signed off and such and we can have our cake and eat it too.  Thanks,

Josef

      reply	other threads:[~2013-02-18 15:33 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-02-13 16:13 [PATCH] Btrfs: place ordered operations on a per transaction list Josef Bacik
2013-02-14 12:47 ` David Sterba
2013-02-18 11:22 ` Miao Xie
2013-02-18 15:33   ` Josef Bacik [this message]

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=20130218153034.GA3188@localhost.localdomain \
    --to=jbacik@fusionio.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=miaox@cn.fujitsu.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 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.