Re: [PATCH 4/5] xen/xenbus: Count pending messages for each watch

[SeongJae Park]

On Thu, 17 Dec 2020 15:50:34 +0100 "Jürgen Groß" <jgross@xxxxxxxx> wrote:

> [-- Attachment #1.1.1: Type: text/plain, Size: 3509 bytes --]
> 
> On 17.12.20 09:17, SeongJae Park wrote:
> > From: SeongJae Park <sjpark@xxxxxxxxx>
> > 
> > This commit adds a counter of pending messages for each watch in the
> > struct.  It is used to skip unnecessary pending messages lookup in
> > 'unregister_xenbus_watch()'.  It could also be used in 'will_handle'
> > callback.
> > 
> > This is part of XSA-349
> > 
> > This is upstream commit 3dc86ca6b4c8cfcba9da7996189d1b5a358a94fc
> > 
> > Cc: stable@xxxxxxxxxxxxxxx
> > Signed-off-by: SeongJae Park <sjpark@xxxxxxxxx>
> > Reported-by: Michael Kurth <mku@xxxxxxxxx>
> > Reported-by: Pawel Wieczorkiewicz <wipawel@xxxxxxxxx>
> > Signed-off-by: Author Redacted <security@xxxxxxx>
> > Reviewed-by: Juergen Gross <jgross@xxxxxxxx>
> > Signed-off-by: Juergen Gross <jgross@xxxxxxxx>
> > ---
> >   drivers/xen/xenbus/xenbus_xs.c | 30 ++++++++++++++++++------------
> >   include/xen/xenbus.h           |  2 ++
> >   2 files changed, 20 insertions(+), 12 deletions(-)
> > 
> > diff --git a/drivers/xen/xenbus/xenbus_xs.c b/drivers/xen/xenbus/xenbus_xs.c
> > index 0ea1c259f2f1..420d478e1708 100644
> > --- a/drivers/xen/xenbus/xenbus_xs.c
> > +++ b/drivers/xen/xenbus/xenbus_xs.c
> > @@ -701,6 +701,8 @@ int register_xenbus_watch(struct xenbus_watch *watch)
> >   
> >   	sprintf(token, "%lX", (long)watch);
> >   
> > +	watch->nr_pending = 0;
> > +
> 
> I'm missing the incrementing of nr_pending, which was present in the
> upstream patch.

Oops, it should be in this patch, but I mistakenly put it in the fifth patch.

  67 --- a/drivers/xen/xenbus/xenbus_xs.c
  68 +++ b/drivers/xen/xenbus/xenbus_xs.c
  69 @@ -917,6 +917,7 @@ static int process_msg(void)
  70                                          msg->u.watch.vec_size))) {
  71                         spin_lock(&watch_events_lock);
  72                         list_add_tail(&msg->list, &watch_events);
  73 +                       msg->u.watch.handle->nr_pending++;
  74                         wake_up(&watch_events_waitq);
  75                         spin_unlock(&watch_events_lock);
  76                 } else {
  77 --

And I just realized I even didn't post the fifth patch.

I will fix this and post new version (v3) soon.

Thank you for catching this, Juergen.


Thanks,
SeongJae Park