All of lore.kernel.org
 help / color / mirror / Atom feed
From: Trond Myklebust <trondmy@hammerspace.com>
To: "neilb@suse.de" <neilb@suse.de>
Cc: "linux-nfs@vger.kernel.org" <linux-nfs@vger.kernel.org>,
	"chuck.lever@oracle.com" <chuck.lever@oracle.com>
Subject: Re: [PATCH v2 1/2] nfsd: Fix a regression in nfsd_setattr()
Date: Mon, 19 Feb 2024 01:33:52 +0000	[thread overview]
Message-ID: <740318c3e5a063db6555455a434c4a86c0e9db48.camel@hammerspace.com> (raw)
In-Reply-To: <170829343514.1530.1077342787397990579@noble.neil.brown.name>

On Mon, 2024-02-19 at 08:57 +1100, NeilBrown wrote:
> On Fri, 16 Feb 2024, trondmy@kernel.org wrote:
> > From: Trond Myklebust <trond.myklebust@hammerspace.com>
> > 
> > Commit bb4d53d66e4b broke the NFSv3 pre/post op attributes
> > behaviour
> > when doing a SETATTR rpc call by stripping out the calls to
> > fh_fill_pre_attrs() and fh_fill_post_attrs().
> > 
> > Fixes: bb4d53d66e4b ("NFSD: use (un)lock_inode instead of
> > fh_(un)lock for file operations")
> > Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
> > ---
> >  fs/nfsd/nfs4proc.c | 4 ++++
> >  fs/nfsd/vfs.c      | 9 +++++++--
> >  2 files changed, 11 insertions(+), 2 deletions(-)
> > 
> > diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
> > index 14712fa08f76..e6d8624efc83 100644
> > --- a/fs/nfsd/nfs4proc.c
> > +++ b/fs/nfsd/nfs4proc.c
> > @@ -1143,6 +1143,7 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct
> > nfsd4_compound_state *cstate,
> >  	};
> >  	struct inode *inode;
> >  	__be32 status = nfs_ok;
> > +	bool save_no_wcc;
> >  	int err;
> >  
> >  	if (setattr->sa_iattr.ia_valid & ATTR_SIZE) {
> > @@ -1168,8 +1169,11 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct
> > nfsd4_compound_state *cstate,
> >  
> >  	if (status)
> >  		goto out;
> > +	save_no_wcc = cstate->current_fh.fh_no_wcc;
> > +	cstate->current_fh.fh_no_wcc = true;
> >  	status = nfsd_setattr(rqstp, &cstate->current_fh, &attrs,
> >  				0, (time64_t)0);
> > +	cstate->current_fh.fh_no_wcc = save_no_wcc;
> 
> This looks clumsy.
> I think the background is that NFSv3 needs atomic wcc attributes for
> file operations, but NFSv4 doesn't - it only has them for directory
> ops.
> So NFSv4, like NFSv2, doesn't want fh_fill_pre_attrs() to be called
> by
> nfsd_setattr().
> 
> NFSv2 avoids it by always setting ->fh_no_wcc.  Here you temporarily
> set
> fh_no_wcc to true for the same effect.  So the code is correct.
> But it is not obvious to the casual reader why this is happening.
> 
> I would rather a "wcc_wanted" flag or similar, but that can be done
> in a
> separate clean-up patch later.

That is in theory what the fh_no_wcc flag is for, however the issue is
that it got overloaded to also mean 'change_info4 wanted' when we added
support for NFSv4 to knfsd.
NFSv4 does not have a concept of weak cache consistency, but it does
try to track updates to the change attribute atomically (ideally) for
most operations that change the directory contents.

IOW: I think a better clean up would be to separate out 'wcc' and
'change_info4' as representing different functionality.

-- 
Trond Myklebust
Linux NFS client maintainer, Hammerspace
trond.myklebust@hammerspace.com



  reply	other threads:[~2024-02-19  1:33 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-02-16  1:24 [PATCH v2 0/2] Fix NFSv3 SETATTR behaviours trondmy
2024-02-16  1:24 ` [PATCH v2 1/2] nfsd: Fix a regression in nfsd_setattr() trondmy
2024-02-16  1:24   ` [PATCH v2 2/2] nfsd: Fix NFSv3 atomicity bugs " trondmy
2024-02-18 14:24     ` Jeff Layton
2024-02-18 22:01     ` NeilBrown
2024-02-16 13:33   ` [PATCH v2 1/2] nfsd: Fix a regression " Chuck Lever
2024-02-16 18:18     ` Trond Myklebust
2024-02-16 18:19       ` Chuck Lever III
2024-02-16 18:25         ` Chuck Lever III
2024-02-16 18:57           ` Trond Myklebust
2024-02-17 17:05             ` Chuck Lever
2024-02-17 18:16               ` Trond Myklebust
2024-02-18 14:21                 ` Jeff Layton
2024-02-18 21:57   ` NeilBrown
2024-02-19  1:33     ` Trond Myklebust [this message]
2024-02-20 15:28 ` [PATCH v2 0/2] Fix NFSv3 SETATTR behaviours Chuck Lever

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=740318c3e5a063db6555455a434c4a86c0e9db48.camel@hammerspace.com \
    --to=trondmy@hammerspace.com \
    --cc=chuck.lever@oracle.com \
    --cc=linux-nfs@vger.kernel.org \
    --cc=neilb@suse.de \
    /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.