[Cluster-devel] [PATCH] filelock: move file locking definitions to separate header file

[Jeff Layton <jlayton@kernel.org>]

On Fri, 2022-11-25 at 03:48 +0000, Al Viro wrote:
> On Sun, Nov 20, 2022 at 03:59:57PM -0500, Jeff Layton wrote:
> 
> > --- /dev/null
> > +++ b/include/linux/filelock.h
> > @@ -0,0 +1,428 @@
> > +/* SPDX-License-Identifier: GPL-2.0 */
> > +#ifndef _LINUX_FILELOCK_H
> > +#define _LINUX_FILELOCK_H
> > +
> > +#include <linux/list.h>
> > +#include <linux/nfs_fs_i.h>
> 
> Umm... I'd add a comment along the lines of "struct file_lock has
> a BS union by fs type; NFS side of things needs nfs_fs_i.h"
> 

Ok.

> > +struct lock_manager_operations {
> > +	void *lm_mod_owner;
> > +	fl_owner_t (*lm_get_owner)(fl_owner_t);
> 
> Probably take fl_owner_t to some more neutral header...
> 

I left it in fs.h for now. Some of the file_operations prototypes need
that typedef, and I figure that anyone who is including filelock.h will
almost certainly need to include fs.h anyway. We could move it into a
separate header too, but it's probably not worth it.

HCH mentioned years ago though that we should just get rid of fl_owner_t
altogether and just use 'void *'. I didn't do it at the time because I
was focused on other changes, but this might be a good time to change
it.

> > +#define locks_inode(f) file_inode(f)
> 
> Why do we still have that one, anyway?  Separate patch, obviously,
> but I would take Occam's Razor to that entity...
> 

I can spin up a patch to nuke that too. I count only 30 callsites
remaining anyway.

> > +struct files_struct;
> > +extern void show_fd_locks(struct seq_file *f,
> > +			 struct file *filp, struct files_struct *files);
> 
> If anything, that would be better off as fl_owner_t...  Again, a separate
> patch.

I'm not sure what you mean here. This prototype hasn't changed, and is
only called from procfs.

-- 
Jeff Layton <jlayton@kernel.org>