Ian Kent ra...@themaw.net wrote:
(4) Stops pathwalk at the automount point and returns that point in the fs
tree if it decides not to automount rather than reporting ELOOP (see
its
use of EXDEV for this).
Does it make autofs easier if d_op-d_automount() is allowed to return
These are all handled by the userspace mount programs, but older versions
of mount.cifs also handed them off to the kernel. Ignore them.
Signed-off-by: Jeff Layton jlay...@redhat.com
---
fs/cifs/connect.c |6 ++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git
On Mon, Jul 26, 2010 at 9:29 AM, Jeff Layton jlay...@redhat.com wrote:
Seems like a more sensible mapping than -EIO.
I agree. Also helps common case of Samba exporting read only shares by default.
Signed-off-by: Jeff Layton jlay...@redhat.com
---
fs/cifs/netmisc.c | 1 +
Ian Kent ra...@themaw.net wrote:
Does it make autofs easier if d_op-d_automount() is allowed to return
-EXDEV to request this? Then you can return it in Oz mode to allow the
daemon to see/use the underlying mountpoint without recursing back into
d_automount().
Yes, it's really useful.
This mount options is used to clue in init scripts that the filesystem
shouldn't be mounted until networking is available. /bin/mount also passes
that option to the filesystem however, and cifs currently chokes on it.
mount.nfs ignores this option -- have mount.cifs do the same.
Signed-off-by:
Even though all known kernels send the uid= parm to userspace,
cifs.upcall doesn't technically require it. It should though. If one
wasn't sent for some reason, then the setuid wouldn't occur. Error out
if there is no uid= or creduid= parm.
Signed-off-by: Jeff Layton jlay...@samba.org
---
These are filesystem-independent mount options that get passed to
mount.cifs too. Handle them appropriately by enabling and disabling
MS_MANDLOCK and not handing them off to the kernel.
Signed-off-by: Jeff Layton jlay...@samba.org
---
mount.cifs.c | 12
1 files changed, 12
I'm not sure why this was merged with this flag hardcoded on, but it
seems quite dangerous. Turn it off.
Also, mount.cifs hands unrecognized options off to the kernel so there
should be no need for changes there in order to support this.
Signed-off-by: Jeff Layton jlay...@redhat.com
---
agreed - merged
On Mon, Jul 26, 2010 at 1:25 PM, Jeff Layton jlay...@redhat.com wrote:
I'm not sure why this was merged with this flag hardcoded on, but it
seems quite dangerous. Turn it off.
Also, mount.cifs hands unrecognized options off to the kernel so there
should be no need for changes