David Disseldorp ddiss at suse.de
Wed Feb 15 06:39:42 MST 2012


On Wed, 15 Feb 2012 07:43:57 +0100
Volker Lendecke <Volker.Lendecke at SerNet.DE> wrote:

> On Tue, Feb 14, 2012 at 03:56:38PM -0800, Jeremy Allison wrote:
> > I'd still like to get the COPYCHUNK VFS changes into master
> > in order to have them for a Samba4.0 release.

As would I :), so long as everyone is satisfied with the changes.

> > Metze, are you happy with the :
> > 
> > +static NTSTATUS vfswrap_copy_chunk(struct vfs_handle_struct *handle,
> > +                                  struct files_struct *src_fsp,
> > +                                  SMB_OFF_T src_off,
> > +                                  struct files_struct *dest_fsp,
> > +                                  SMB_OFF_T dest_off,
> > +                                  SMB_OFF_T num,
> > +                                  SMB_OFF_T *copied)
> > 
> > interface for the VFS ?

Metze's last round of feedback indicated his preference for using the
existing SMB_VFS_FSCTL hook. I don't mind going down that path, but I
haven't yet found the time.
It's currently quite a messy interface IMO, given the differences
between the smb1 and smb2 ioctl code paths. Furthermore, some fsctls
already have operation specific VFS hooks (e.g. GET_SHADOW_COPY_DATA).

> Can we get a _send and _recv pair of those into the VFS, for
> obvious reasons?

Should copy_chunk asynchronicity be handled at a higher layer, or is it
the intention that VFS presents send/receive pairs for all operations
that are potentially IO intensive?

Cheers, David

More information about the samba-technical mailing list