[PATCH] NFSD, lost nfserrno() call in nfsd_setattr()

[PATCH] NFSD, lost nfserrno() call in nfsd_setattr()

[J. R. Okajima]

There is a regression in
	208d0ac 2014-01-07 nfsd4: break only delegations when appropriate
which deletes an nfserrno() call in nfsd_setattr() (by accident,
probably), and NFSD becomes ignoring an error from VFS.

Here is a patch to fix it.

J. R. Okajima


diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 017d3cb..6d7be3f 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -449,6 +449,7 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
 	fh_lock(fhp);
 	host_err = notify_change(dentry, iap, NULL);
 	fh_unlock(fhp);
+	err = nfserrno(host_err);
 
 out_put_write_access:
 	if (size_change)

Re: [PATCH] NFSD, lost nfserrno() call in nfsd_setattr()

[J. Bruce Fields]

On Wed, Feb 19, 2014 at 12:27:53AM +0900, J. R. Okajima wrote:
> 
> There is a regression in
> 	208d0ac 2014-01-07 nfsd4: break only delegations when appropriate
> which deletes an nfserrno() call in nfsd_setattr() (by accident,
> probably), and NFSD becomes ignoring an error from VFS.
> 
> Here is a patch to fix it.

Thanks for cathing that!  Queueing up for 3.14.

--b.

> 
> J. R. Okajima
> 
> 
> diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
> index 017d3cb..6d7be3f 100644
> --- a/fs/nfsd/vfs.c
> +++ b/fs/nfsd/vfs.c
> @@ -449,6 +449,7 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap,
>  	fh_lock(fhp);
>  	host_err = notify_change(dentry, iap, NULL);
>  	fh_unlock(fhp);
> +	err = nfserrno(host_err);
>  
>  out_put_write_access:
>  	if (size_change)

Re: [PATCH] NFSD, lost nfserrno() call in nfsd_setattr()

[J. R. Okajima]

"J. Bruce Fields":
> On Wed, Feb 19, 2014 at 12:27:53AM +0900, J. R. Okajima wrote:
> > 
> > There is a regression in
> > 	208d0ac 2014-01-07 nfsd4: break only delegations when appropriate
> > which deletes an nfserrno() call in nfsd_setattr() (by accident,
> > probably), and NFSD becomes ignoring an error from VFS.
> > 
> > Here is a patch to fix it.
>
> Thanks for cathing that!  Queueing up for 3.14.

Now rc8 is released, but this patch is still un-merged.
The bug is actually a show-stopper for my local tests.
Do you really have a plan to do it? Or should I ask Linus Torvalds
directly?


J. R. Okajima

Re: [PATCH] NFSD, lost nfserrno() call in nfsd_setattr()

[J. Bruce Fields]

On Tue, Mar 25, 2014 at 05:19:02PM +0900, J. R. Okajima wrote:
> 
> "J. Bruce Fields":
> > On Wed, Feb 19, 2014 at 12:27:53AM +0900, J. R. Okajima wrote:
> > > 
> > > There is a regression in
> > > 	208d0ac 2014-01-07 nfsd4: break only delegations when appropriate
> > > which deletes an nfserrno() call in nfsd_setattr() (by accident,
> > > probably), and NFSD becomes ignoring an error from VFS.
> > > 
> > > Here is a patch to fix it.
> >
> > Thanks for cathing that!  Queueing up for 3.14.
> 
> Now rc8 is released, but this patch is still un-merged.
> The bug is actually a show-stopper for my local tests.
> Do you really have a plan to do it? Or should I ask Linus Torvalds
> directly?

If you remember, just bug me once a week or so.  Snd then if it gets
this late, yes, sending straight to Linus with a cc: to the unresponsive
maintainer and any relevant mailing list is probably a reasonable thing
to do in this sort of situation.

Anyway, my screwup, apologies for not getting that merged at the start.

--b.