diff mbox

NFSv4.2: handle NFS-specific llseek errors

Message ID 20150723150843.GA12894@fieldses.org (mailing list archive)
State New, archived
Headers show

Commit Message

J. Bruce Fields July 23, 2015, 3:08 p.m. UTC
From: "J. Bruce Fields" <bfields@redhat.com>

Handle NFS-specific llseek errors instead of letting them leak out to
userspace.

Reported-by: Benjamin Coddington <bcodding@redhat.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
---
 fs/nfs/nfs42proc.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

I don't actually have a test case for this, but it looks right....

Comments

Schumaker, Anna July 23, 2015, 3:25 p.m. UTC | #1
Hi Bruce,

On 07/23/2015 11:08 AM, J. Bruce Fields wrote:
> From: "J. Bruce Fields" <bfields@redhat.com>
> 
> Handle NFS-specific llseek errors instead of letting them leak out to
> userspace.
> 
> Reported-by: Benjamin Coddington <bcodding@redhat.com>
> Signed-off-by: J. Bruce Fields <bfields@redhat.com>
> ---
>  fs/nfs/nfs42proc.c | 19 ++++++++++++++++++-
>  1 file changed, 18 insertions(+), 1 deletion(-)
> 
> I don't actually have a test case for this, but it looks right....
> 
> diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
> index f486b80f927a..2fec410bc50f 100644
> --- a/fs/nfs/nfs42proc.c
> +++ b/fs/nfs/nfs42proc.c
> @@ -135,7 +135,7 @@ int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len)
>  	return err;
>  }
>  
> -loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
> +loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
>  {
>  	struct inode *inode = file_inode(filep);
>  	struct nfs42_seek_args args = {
> @@ -171,6 +171,23 @@ loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
>  	return vfs_setpos(filep, res.sr_offset, inode->i_sb->s_maxbytes);
>  }
>  
> +loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
> +{
> +	struct nfs_server *server = NFS_SERVER(file_inode(filep));
> +	struct nfs4_exception exception = { };
> +	int err;

Can you move the nfs_server_capable() check to before the do {} while loop?  I think it's cleaner if we check for support before ever trying to call _nfs42_proc_llseek().

> +
> +	do {
> +		err = _nfs42_proc_llseek(filep, offset, whence);
> +		if (err == -ENOTSUPP)
> +			return -EOPNOTSUPP;

With the change to check support moved to this function, we should also unset NFS_CAP_SEEK if we see -ENOTSUPP here.  I think you'll also need to change nfs4_file_llseek() (in nfs4file.c) to check for -EOPNOTSUPP instead of -ENOTSUPP.

Thanks,
Anna

> +		err = nfs4_handle_exception(server, err, &exception);
> +	} while (exception.retry);
> +
> +	return err;
> +}
> +
> +
>  static void
>  nfs42_layoutstat_prepare(struct rpc_task *task, void *calldata)
>  {
> 

--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
J. Bruce Fields July 23, 2015, 7:21 p.m. UTC | #2
On Thu, Jul 23, 2015 at 11:25:23AM -0400, Anna Schumaker wrote:
> Hi Bruce,
> 
> On 07/23/2015 11:08 AM, J. Bruce Fields wrote:
> > From: "J. Bruce Fields" <bfields@redhat.com>
> > 
> > Handle NFS-specific llseek errors instead of letting them leak out to
> > userspace.
> > 
> > Reported-by: Benjamin Coddington <bcodding@redhat.com>
> > Signed-off-by: J. Bruce Fields <bfields@redhat.com>
> > ---
> >  fs/nfs/nfs42proc.c | 19 ++++++++++++++++++-
> >  1 file changed, 18 insertions(+), 1 deletion(-)
> > 
> > I don't actually have a test case for this, but it looks right....
> > 
> > diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
> > index f486b80f927a..2fec410bc50f 100644
> > --- a/fs/nfs/nfs42proc.c
> > +++ b/fs/nfs/nfs42proc.c
> > @@ -135,7 +135,7 @@ int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len)
> >  	return err;
> >  }
> >  
> > -loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
> > +loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
> >  {
> >  	struct inode *inode = file_inode(filep);
> >  	struct nfs42_seek_args args = {
> > @@ -171,6 +171,23 @@ loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
> >  	return vfs_setpos(filep, res.sr_offset, inode->i_sb->s_maxbytes);
> >  }
> >  
> > +loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
> > +{
> > +	struct nfs_server *server = NFS_SERVER(file_inode(filep));
> > +	struct nfs4_exception exception = { };
> > +	int err;
> 
> Can you move the nfs_server_capable() check to before the do {} while loop?  I think it's cleaner if we check for support before ever trying to call _nfs42_proc_llseek().

I believe the results of that check can change on subsequent iterations,
due to a concurrent SEEK returning NOTSUPP.

The code would be correct either way, because the nfs_server_capable()
check is really just an optimization--it should always be OK just to
send the SEEK and then handle the NOTSUPP reply.

But accidentally sending an unnecessary rpc is much more expensive than
checking a flag, so the code's probably best left as is.

(Also, this is the way it's done elsewhere in nfs{4,42}proc.c, so if
there's some reason to change we'd want to do that everywhere.)

--b.
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
index f486b80f927a..2fec410bc50f 100644
--- a/fs/nfs/nfs42proc.c
+++ b/fs/nfs/nfs42proc.c
@@ -135,7 +135,7 @@  int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len)
 	return err;
 }
 
-loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
+loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
 {
 	struct inode *inode = file_inode(filep);
 	struct nfs42_seek_args args = {
@@ -171,6 +171,23 @@  loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
 	return vfs_setpos(filep, res.sr_offset, inode->i_sb->s_maxbytes);
 }
 
+loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
+{
+	struct nfs_server *server = NFS_SERVER(file_inode(filep));
+	struct nfs4_exception exception = { };
+	int err;
+
+	do {
+		err = _nfs42_proc_llseek(filep, offset, whence);
+		if (err == -ENOTSUPP)
+			return -EOPNOTSUPP;
+		err = nfs4_handle_exception(server, err, &exception);
+	} while (exception.retry);
+
+	return err;
+}
+
+
 static void
 nfs42_layoutstat_prepare(struct rpc_task *task, void *calldata)
 {