All of lore.kernel.org
 help / color / mirror / Atom feed
From: "J. Bruce Fields" <bfields@fieldses.org>
To: Christoph Hellwig <hch@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
	linux-fsdevel@vger.kernel.org,
	Peter Staubach <staubach@redhat.com>,
	Trond Myklebust <Trond.Myklebust@netapp.com>
Subject: Re: [PATCH 6/6] nfs: disable leases over NFS
Date: Thu, 5 Jul 2007 11:41:00 -0400	[thread overview]
Message-ID: <20070705154100.GC29867@fieldses.org> (raw)
In-Reply-To: <20070630092516.GD22050@infradead.org>

On Sat, Jun 30, 2007 at 10:25:16AM +0100, Christoph Hellwig wrote:
> On Fri, Jun 29, 2007 at 03:21:30PM -0400, J. Bruce Fields wrote:
> > Define an nfs setlease method that just returns -EOPNOTSUPP.
> > 
> > If someone can demonstrate a real need, perhaps we could reenable
> > them in the presence of the "nolock" mount option.
> 
> I'm not a big fan of default methods that do the wrong thing instead
> of just missing functionality.  Would you mind just returning
> -EOPNOTSUPP if ->setlease is not implemented and add it to all
> the local filesystems while all the network/distributed filesystems
> should not have it, not just nfs.

OK, after looking at this a little more, I'm less happy about the idea
of erroring out by default:

	- There are a ton of filesystems that probably should allow
	  leases, and only a few (network filesystems) that shouldn't,
	  so leaving leases on by default seems simpler.
	- We already fall back on the local method by default in the case
	  of locks, and I don't see a reason to treat leases differently.
	- The patch to add
		.setlease = setlease,
	  to all the file_operations is going to be a big patch that
	  changes behavior in a way that might be easy to miss (because
	  it changes behavior exactly on those filesystems it *doesn't*
	  touch.)  I think it'll be easier to get better review on the
	  patch that adds a method just to those filesystems that we're
	  disabling leases for.

I agree about dealing with the other network filesystems, though, and
not just NFS and GFS2; I'll look into that.

--b.

  parent reply	other threads:[~2007-07-05 15:41 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-06-29 19:21 vfs lease api J. Bruce Fields
2007-06-29 19:21 ` [PATCH 1/6] locks: share more common lease code J. Bruce Fields
2007-06-29 19:21   ` J. Bruce Fields
2007-06-29 19:21   ` [PATCH 2/6] locks: provide a file lease method enabling cluster-coherent leases J. Bruce Fields
2007-06-29 19:21     ` J. Bruce Fields
2007-06-30  9:26     ` Christoph Hellwig
2007-06-29 19:21   ` [PATCH 3/6] locks: rename lease functions to reflect locks.c conventions J. Bruce Fields
2007-06-29 19:21     ` J. Bruce Fields
2007-06-30  9:22     ` Christoph Hellwig
2007-07-04 21:42       ` J. Bruce Fields
2007-06-29 19:21   ` [PATCH 4/6] locks: fix locks.c lease symbol exports J. Bruce Fields
2007-06-29 19:21     ` J. Bruce Fields
2007-06-30  9:23     ` Christoph Hellwig
2007-07-04 21:42       ` J. Bruce Fields
2007-06-29 19:21   ` [PATCH 5/6] gfs2: stop giving out non-cluster-coherent leases J. Bruce Fields
2007-06-29 19:21     ` J. Bruce Fields
2007-06-30  9:27     ` Christoph Hellwig
2007-07-01 15:00       ` J. Bruce Fields
2007-07-04 21:48       ` J. Bruce Fields
2007-06-29 19:21   ` [PATCH 6/6] nfs: disable leases over NFS J. Bruce Fields
2007-06-29 19:21     ` J. Bruce Fields
2007-06-29 21:16       ` Peter Staubach
2007-06-29 21:39         ` J. Bruce Fields
2007-06-29 22:30           ` Steven Whitehouse
2007-06-29 22:21             ` J. Bruce Fields
2007-06-30  9:25     ` Christoph Hellwig
2007-07-04 23:22       ` J. Bruce Fields
2007-07-05 15:41       ` J. Bruce Fields [this message]
2007-07-11 10:20         ` Christoph Hellwig
2007-07-11 23:10           ` J. Bruce Fields
2007-06-30  9:20   ` [PATCH 1/6] locks: share more common lease code Christoph Hellwig
2007-07-03 22:17     ` J. Bruce Fields
2007-07-04  8:26       ` Christoph Hellwig

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=20070705154100.GC29867@fieldses.org \
    --to=bfields@fieldses.org \
    --cc=Trond.Myklebust@netapp.com \
    --cc=akpm@linux-foundation.org \
    --cc=hch@infradead.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=staubach@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 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.