Re: [PATCH] locks: ensure that fl_owner is always initialized properly in flock and lease codepaths

2014-05-06 Thread Gregory Farnum
The Ceph bit is fine. Acked-by: Greg Farnum On Mon, Apr 28, 2014 at 10:50 AM, Jeff Layton wrote: > Currently, the fl_owner isn't set for flock locks. Some filesystems use > byte-range locks to simulate flock locks and there is a common idiom in > those that does: > > fl->fl_owner =

Re: [PATCH] locks: ensure that fl_owner is always initialized properly in flock and lease codepaths

2014-05-06 Thread Gregory Farnum
The Ceph bit is fine. Acked-by: Greg Farnum g...@inktank.com On Mon, Apr 28, 2014 at 10:50 AM, Jeff Layton jlay...@poochiereds.net wrote: Currently, the fl_owner isn't set for flock locks. Some filesystems use byte-range locks to simulate flock locks and there is a common idiom in those that

Re: [PATCH] locks: ensure that fl_owner is always initialized properly in flock and lease codepaths

2014-05-05 Thread J. Bruce Fields
Looks fine to me, ACK.--b. On Mon, Apr 28, 2014 at 01:50:13PM -0400, Jeff Layton wrote: > Currently, the fl_owner isn't set for flock locks. Some filesystems use > byte-range locks to simulate flock locks and there is a common idiom in > those that does: > > fl->fl_owner = (fl_owner_t)filp;

Re: [PATCH] locks: ensure that fl_owner is always initialized properly in flock and lease codepaths

2014-05-05 Thread J. Bruce Fields
Looks fine to me, ACK.--b. On Mon, Apr 28, 2014 at 01:50:13PM -0400, Jeff Layton wrote: Currently, the fl_owner isn't set for flock locks. Some filesystems use byte-range locks to simulate flock locks and there is a common idiom in those that does: fl-fl_owner = (fl_owner_t)filp;

[PATCH] locks: ensure that fl_owner is always initialized properly in flock and lease codepaths

2014-04-28 Thread Jeff Layton
Currently, the fl_owner isn't set for flock locks. Some filesystems use byte-range locks to simulate flock locks and there is a common idiom in those that does: fl->fl_owner = (fl_owner_t)filp; fl->fl_start = 0; fl->fl_end = OFFSET_MAX; Since flock locks are generally "owned" by the

Re: [PATCH] locks: ensure that fl_owner is always initialized properly in flock and lease codepaths

2014-04-28 Thread Greg Kroah-Hartman
On Mon, Apr 28, 2014 at 01:50:13PM -0400, Jeff Layton wrote: > Currently, the fl_owner isn't set for flock locks. Some filesystems use > byte-range locks to simulate flock locks and there is a common idiom in > those that does: > > fl->fl_owner = (fl_owner_t)filp; > fl->fl_start = 0; >

Re: [PATCH] locks: ensure that fl_owner is always initialized properly in flock and lease codepaths

2014-04-28 Thread Greg Kroah-Hartman
On Mon, Apr 28, 2014 at 01:50:13PM -0400, Jeff Layton wrote: Currently, the fl_owner isn't set for flock locks. Some filesystems use byte-range locks to simulate flock locks and there is a common idiom in those that does: fl-fl_owner = (fl_owner_t)filp; fl-fl_start = 0;

[PATCH] locks: ensure that fl_owner is always initialized properly in flock and lease codepaths

2014-04-28 Thread Jeff Layton
Currently, the fl_owner isn't set for flock locks. Some filesystems use byte-range locks to simulate flock locks and there is a common idiom in those that does: fl-fl_owner = (fl_owner_t)filp; fl-fl_start = 0; fl-fl_end = OFFSET_MAX; Since flock locks are generally owned by the open