mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jeff Layton <jlayton@redhat.com>
To: "J. Bruce Fields" <bfields@fieldses.org>
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	nfs-ganesha-devel@lists.sourceforge.net,
	samba-technical@lists.samba.org
Subject: Re: [RFC PATCH v2 3/5] locks: add new "private" lock type that is owned by the filp
Date: Wed, 20 Nov 2013 15:00:47 -0500	[thread overview]
Message-ID: <20131120150047.06a25ec7@tlielax.poochiereds.net> (raw)
In-Reply-To: <20131120195032.GA7387@fieldses.org>

On Wed, 20 Nov 2013 14:50:32 -0500
"J. Bruce Fields" <bfields@fieldses.org> wrote:

> On Wed, Nov 20, 2013 at 11:45:04AM -0500, Jeff Layton wrote:
> > Due to some unfortunate history, POSIX locks have very strange and
> > unhelpful semantics. The thing that usually catches people by surprise
> > is that they are dropped whenever the process closes any file descriptor
> > associated with the inode.
> > 
> > This is extremely problematic for people developing file servers that
> > need to implement byte-range locks. Developers often need a "lock
> > management" facility to ensure that file descriptors are not closed
> > until all of the locks associated with the inode are finished.
> > 
> > This patchset adds a new type of lock that attempts to address this
> > issue. These locks work just like "normal" POSIX read/write locks, but
> > have semantics that are more like BSD locks with respect to inheritance
> > and behavior on close.
> > 
> > This is implemented primarily by changing how fl_owner field is set for
> > these locks. Instead of having them owned by the files_struct of the
> > process, they are instead owned by the filp on which they were acquired.
> > Thus, they are inherited across fork() and are only released when the
> > last reference to a filp is put.
> > 
> > These new semantics prevent them from being merged with "classic" POSIX
> > locks, even if they are acquired by the same process. These locks will
> > also conflict with "classic" POSIX locks even if they are acquired by
> > the same process or on the same file descriptor.
> > 
> > Signed-off-by: Jeff Layton <jlayton@redhat.com>
> > ---
> >  fs/locks.c                       | 22 +++++++++++++++++++++-
> >  include/uapi/asm-generic/fcntl.h | 15 +++++++++++++++
> >  2 files changed, 36 insertions(+), 1 deletion(-)
> > 
> > diff --git a/fs/locks.c b/fs/locks.c
> > index 86cafc3..3b278a6 100644
> > --- a/fs/locks.c
> > +++ b/fs/locks.c
> > @@ -348,6 +348,26 @@ static int posix_assign_type(struct file_lock *fl, long type)
> >  {
> >  	int err;
> >  
> > +	/*
> > +	 * FL_FILP_PRIVATE locks are "owned" by the filp upon which they were
> > +	 * acquired, regardless of what task is dealing with them. Set the
> > +	 * fl_owner appropriately.
> > +	 */
> > +	switch (type) {
> > +	case F_RDLCKP:
> > +		type = F_RDLCK;
> > +		fl->fl_owner = (fl_owner_t)fl->fl_file;
> > +		break;
> > +	case F_WRLCKP:
> > +		type = F_WRLCK;
> > +		fl->fl_owner = (fl_owner_t)fl->fl_file;
> > +		break;
> > +	case F_UNLCKP:
> > +		type = F_UNLCK;
> > +		fl->fl_owner = (fl_owner_t)fl->fl_file;
> > +		break;
> > +	}
> > +
> 
> After this fl_owner gets set to current->files in
> flock{64}_to_posix_lock and then reset here.  That seems like a trap for
> the unwary reader.
> 
> Could you do something like rename this flock_to_posix_lock_common and
> move all the 32/64-bit-independent initialization here?
> 
> Looks like there's way more duplication than necessary between those two
> cases.
> 

Good point. I'll look at doing some more cleanup in that direction on
the next set.

> (Also, why do we have an fl_owner_t instead of using a void?)
> 

Another good point. As far as I can tell nothing uses fl_owner for
anything but a generic ownership "cookie". The pointer is never
actually dereferenced to get back to a files_struct, so a void pointer
(or even an unsigned long) would make more sense there.

I'll look at that before sending out the next version of this.

Thanks,

> 
> >  	err = assign_type(fl, type);
> >  	if (err)
> >  		return err;
> > @@ -2225,7 +2245,7 @@ void locks_remove_filp(struct file *filp)
> >  
> >  	while ((fl = *before) != NULL) {
> >  		if (fl->fl_file == filp) {
> > -			if (IS_FLOCK(fl)) {
> > +			if (IS_FLOCK(fl) || IS_POSIX(fl)) {
> >  				locks_delete_lock(before);
> >  				continue;
> >  			}
> > diff --git a/include/uapi/asm-generic/fcntl.h b/include/uapi/asm-generic/fcntl.h
> > index 95e46c8..6b7b68a 100644
> > --- a/include/uapi/asm-generic/fcntl.h
> > +++ b/include/uapi/asm-generic/fcntl.h
> > @@ -151,6 +151,21 @@ struct f_owner_ex {
> >  #define F_UNLCK		2
> >  #endif
> >  
> > +/*
> > + * fd "private" POSIX locks.
> > + *
> > + * Usually POSIX locks held by a process are released on *any* close and are
> > + * not inherited across a fork().
> > + *
> > + * These lock types will conflict with normal POSIX locks, but are "owned"
> > + * by the fd, not the process. This means that they are inherited across
> > + * fork() like BSD (flock) locks, and they are only closed when the last
> > + * reference to the the filp against which were acquired is closed.
> > + */
> > +#define F_RDLCKP	5
> > +#define F_WRLCKP	6
> > +#define F_UNLCKP	7
> > +
> >  /* for old implementation of bsd flock () */
> >  #ifndef F_EXLCK
> >  #define F_EXLCK		4	/* or 3 */
> > -- 
> > 1.8.3.1
> > 
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html


-- 
Jeff Layton <jlayton@redhat.com>

  reply	other threads:[~2013-11-20 20:01 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-20 16:45 [RFC PATCH v2 0/5] locks: implement "filp-private" (aka UNPOSIX) locks Jeff Layton
2013-11-20 16:45 ` [RFC PATCH v2 1/5] locks: consolidate checks for compatible filp->f_mode values in setlk handlers Jeff Layton
2013-11-21 17:35   ` Christoph Hellwig
2013-11-20 16:45 ` [RFC PATCH v2 2/5] locks: rename locks_remove_flock to locks_remove_filp Jeff Layton
2013-11-21 17:35   ` Christoph Hellwig
2013-11-20 16:45 ` [RFC PATCH v2 3/5] locks: add new "private" lock type that is owned by the filp Jeff Layton
2013-11-20 19:50   ` J. Bruce Fields
2013-11-20 20:00     ` Jeff Layton [this message]
2013-12-09 16:46     ` [Nfs-ganesha-devel] " Jeff Layton
2013-12-10 20:35       ` J. Bruce Fields
2013-11-21 17:36   ` Christoph Hellwig
2013-11-20 16:45 ` [RFC PATCH v2 4/5] locks: show private lock types in /proc/locks Jeff Layton
2013-11-20 20:00   ` J. Bruce Fields
2013-11-20 20:03     ` Jeff Layton
2013-11-20 20:04     ` J. Bruce Fields
2013-11-20 16:45 ` [RFC PATCH v2 5/5] locks: report l_pid as -1 for FL_FILP_PRIVATE locks Jeff Layton
2013-11-20 17:55   ` Frank Filz
2013-11-20 18:57     ` Jeff Layton

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20131120150047.06a25ec7@tlielax.poochiereds.net \
    --to=jlayton@redhat.com \
    --cc=bfields@fieldses.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nfs-ganesha-devel@lists.sourceforge.net \
    --cc=samba-technical@lists.samba.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®