[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Y5uOnzBA7D30VUQY@pc636>
Date: Thu, 15 Dec 2022 22:16:15 +0100
From: Uladzislau Rezki <urezki@...il.com>
To: "Paul E. McKenney" <paulmck@...nel.org>
Cc: "Uladzislau Rezki (Sony)" <urezki@...il.com>,
LKML <linux-kernel@...r.kernel.org>, RCU <rcu@...r.kernel.org>,
Frederic Weisbecker <frederic@...nel.org>,
Neeraj Upadhyay <neeraj.iitr10@...il.com>,
Joel Fernandes <joel@...lfernandes.org>,
Oleksiy Avramchenko <oleksiy.avramchenko@...y.com>
Subject: Re: [PATCH 1/2] rcu/kvfree: Carefully reset number of objects in krcp
On Thu, Dec 15, 2022 at 12:40:22PM -0800, Paul E. McKenney wrote:
> On Wed, Dec 14, 2022 at 01:06:29PM +0100, Uladzislau Rezki (Sony) wrote:
> > Problem. A schedule_delayed_monitor_work() relays on the
> > number of pointers queued into krcp. Based on that number
> > and threshold the work is rearmed with different delayed
> > intervals, i.e. sooner or later.
> >
> > There are three pipes where pointers can be placed. When
> > any pipe is offloaded the krcp->count counter is set to
> > zero - what is wrong. Because another pipes might not be
> > empty.
> >
> > Fix it by maintaining a counter individually per a pipe.
> > In order to get a number of objects per a krcp introduce
> > a krc_count() helper.
> >
> > Signed-off-by: Uladzislau Rezki (Sony) <urezki@...il.com>
>
> Queued and pushed for further review and testing, and with the usual
> wordsmithing. Much better use of workqueue handlers, thank you!!!
>
<snip>
This commit therefore maintains per-pipe counters, and introduces a a
krc_count() helper to access the aggregate value of those
counters.
<snip>
a a ? :)
--
Uladzislau Rezki
Powered by blists - more mailing lists