Re: [PATCH 01/11] cifs: track local_nls in volume info
Steve French <[email protected]> Sun, 25 Apr 2010 18:19:21 -0500
| Newsgroups | gmane.linux.file-systems.cifs |
|---|---|
| Message-ID | <[email protected]> |
On Sun, Apr 25, 2010 at 6:14 PM, Jeff Layton <[email protected]> wrote: > On Sun, 25 Apr 2010 14:52:46 -0500 > Steve French <[email protected]> wrote: > >> On Sat, Apr 24, 2010 at 6:57 AM, Jeff Layton <[email protected]> wrote: >> > Add a local_nls field to the smb_vol struct and keep a pointer to the >> > local_nls in it. >> > >> > Signed-off-by: Jeff Layton <[email protected]> >> > --- >> > fs/cifs/connect.c | 11 ++++++----- >> > 1 files changed, 6 insertions(+), 5 deletions(-) >> > >> > diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c >> > index 58a2109..eb85dd8 100644 >> > --- a/fs/cifs/connect.c >> > +++ b/fs/cifs/connect.c >> > @@ -102,6 +102,7 @@ struct smb_vol { >> > bool sockopt_tcp_nodelay:1; >> > unsigned short int port; >> > char *prepath; >> > + struct nls_table *local_nls; >> > }; >> >> If we do this here, should we take it out of the cifs_sb? > > No, the smb_vol doesn't live past the session/tcon establishment. > >> If two volumes mounted to the same cifs_sb - last nls wins? >> > > Not sure I understand this question. The smb_vol is unfortunately > somewhat poorly named. It's mostly a struct for holding the results of > parsing mount options. Sorry about that - I was confusing it with the vfsmnt structure (vs. the subperblock) -- Thanks, Steve