From: "Paul E. McKenney" <paulmck@kernel.org>
To: Frederic Weisbecker <frederic@kernel.org>
Cc: "Joel Fernandes (Google)" <joel@joelfernandes.org>,
rcu@vger.kernel.org, linux-kernel@vger.kernel.org,
rushikesh.s.kadam@intel.com, urezki@gmail.com,
neeraj.iitr10@gmail.com, rostedt@goodmis.org,
vineeth@bitbyteword.org
Subject: Re: [PATCH v2 1/8] rcu: Introduce call_rcu_lazy() API implementation
Date: Wed, 29 Jun 2022 10:05:55 -0700 [thread overview]
Message-ID: <20220629170555.GK1790663@paulmck-ThinkPad-P17-Gen-1> (raw)
In-Reply-To: <20220629115349.GA2119930@lothringen>
On Wed, Jun 29, 2022 at 01:53:49PM +0200, Frederic Weisbecker wrote:
> On Wed, Jun 22, 2022 at 10:50:55PM +0000, Joel Fernandes (Google) wrote:
> > @@ -414,30 +427,37 @@ static bool rcu_nocb_try_bypass(struct rcu_data *rdp, struct rcu_head *rhp,
> > }
> > WRITE_ONCE(rdp->nocb_nobypass_count, c);
> >
> > - // If there hasn't yet been all that many ->cblist enqueues
> > - // this jiffy, tell the caller to enqueue onto ->cblist. But flush
> > - // ->nocb_bypass first.
> > - if (rdp->nocb_nobypass_count < nocb_nobypass_lim_per_jiffy) {
> > + // If caller passed a non-lazy CB and there hasn't yet been all that
> > + // many ->cblist enqueues this jiffy, tell the caller to enqueue it
> > + // onto ->cblist. But flush ->nocb_bypass first. Also do so, if total
> > + // number of CBs (lazy + non-lazy) grows too much.
> > + //
> > + // Note that if the bypass list has lazy CBs, and the main list is
> > + // empty, and rhp happens to be non-lazy, then we end up flushing all
> > + // the lazy CBs to the main list as well. That's the right thing to do,
> > + // since we are kick-starting RCU GP processing anyway for the non-lazy
> > + // one, we can just reuse that GP for the already queued-up lazy ones.
> > + if ((rdp->nocb_nobypass_count < nocb_nobypass_lim_per_jiffy && !lazy) ||
> > + (lazy && n_lazy_cbs >= qhimark)) {
> > rcu_nocb_lock(rdp);
> > *was_alldone = !rcu_segcblist_pend_cbs(&rdp->cblist);
> > if (*was_alldone)
> > trace_rcu_nocb_wake(rcu_state.name, rdp->cpu,
> > - TPS("FirstQ"));
> > - WARN_ON_ONCE(!rcu_nocb_flush_bypass(rdp, NULL, j));
> > + lazy ? TPS("FirstLazyQ") : TPS("FirstQ"));
> > + WARN_ON_ONCE(!rcu_nocb_flush_bypass(rdp, NULL, j, false));
>
> That's outside the scope of this patchset but this makes me realize we
> unconditionally try to flush the bypass from call_rcu() fastpath, and
> therefore we unconditionally lock the bypass lock from call_rcu() fastpath.
>
> It shouldn't be contended at this stage since we are holding the nocb_lock
> already, and only the local CPU can hold the nocb_bypass_lock without holding
> the nocb_lock. But still...
>
> It looks safe to locklessly early check if (rcu_cblist_n_cbs(&rdp->nocb_bypass))
> before doing anything. Only the local CPU can enqueue to the bypass list.
>
> Adding that to my TODO list...
That does sound like a potentially very helpful approach! As always,
please analyze and test carefully!
Thanx, Paul
next prev parent reply other threads:[~2022-06-29 17:06 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-22 22:50 [PATCH v2 0/8] Implement call_rcu_lazy() and miscellaneous fixes Joel Fernandes (Google)
2022-06-22 22:50 ` [PATCH v2 1/1] context_tracking: Use arch_atomic_read() in __ct_state for KASAN Joel Fernandes (Google)
2022-06-22 22:58 ` Joel Fernandes
2022-06-22 22:50 ` [PATCH v2 1/8] rcu: Introduce call_rcu_lazy() API implementation Joel Fernandes (Google)
2022-06-22 23:18 ` Joel Fernandes
2022-06-26 4:00 ` Paul E. McKenney
2022-06-23 1:38 ` kernel test robot
2022-06-26 4:00 ` Paul E. McKenney
2022-07-08 18:43 ` Joel Fernandes
2022-07-08 23:10 ` Paul E. McKenney
2022-07-10 2:26 ` Joel Fernandes
2022-07-10 16:03 ` Paul E. McKenney
2022-07-12 20:53 ` Joel Fernandes
2022-07-12 21:04 ` Paul E. McKenney
2022-07-12 21:10 ` Joel Fernandes
2022-07-12 22:41 ` Paul E. McKenney
2022-06-29 11:53 ` Frederic Weisbecker
2022-06-29 17:05 ` Paul E. McKenney [this message]
2022-06-29 20:29 ` Joel Fernandes
2022-06-29 22:01 ` Frederic Weisbecker
2022-06-30 14:08 ` Joel Fernandes
2022-06-22 22:50 ` [PATCH v2 2/8] rcu: shrinker for lazy rcu Joel Fernandes (Google)
2022-06-22 22:50 ` [PATCH v2 3/8] fs: Move call_rcu() to call_rcu_lazy() in some paths Joel Fernandes (Google)
2022-06-22 22:50 ` [PATCH v2 4/8] rcu/nocb: Add option to force all call_rcu() to lazy Joel Fernandes (Google)
2022-06-22 22:50 ` [PATCH v2 5/8] rcu/nocb: Wake up gp thread when flushing Joel Fernandes (Google)
2022-06-26 4:06 ` Paul E. McKenney
2022-06-26 13:45 ` Joel Fernandes
2022-06-26 13:52 ` Paul E. McKenney
2022-06-26 14:37 ` Joel Fernandes
2022-06-22 22:51 ` [PATCH v2 6/8] rcuscale: Add test for using call_rcu_lazy() to emulate kfree_rcu() Joel Fernandes (Google)
2022-06-23 2:09 ` kernel test robot
2022-06-23 3:00 ` kernel test robot
2022-06-23 8:10 ` kernel test robot
2022-06-26 4:13 ` Paul E. McKenney
2022-07-08 4:25 ` Joel Fernandes
2022-07-08 23:06 ` Paul E. McKenney
2022-07-12 20:27 ` Joel Fernandes
2022-07-12 20:58 ` Paul E. McKenney
2022-07-12 21:15 ` Joel Fernandes
2022-07-12 22:41 ` Paul E. McKenney
2022-06-22 22:51 ` [PATCH v2 7/8] rcu/nocb: Rewrite deferred wake up logic to be more clean Joel Fernandes (Google)
2022-06-22 22:51 ` [PATCH v2 8/8] rcu/kfree: Fix kfree_rcu_shrink_count() return value Joel Fernandes (Google)
2022-06-26 4:17 ` Paul E. McKenney
2022-06-27 18:56 ` Uladzislau Rezki
2022-06-27 20:59 ` Paul E. McKenney
2022-06-27 21:18 ` Joel Fernandes
2022-06-27 21:43 ` Paul E. McKenney
2022-06-28 16:56 ` Joel Fernandes
2022-06-28 21:13 ` Joel Fernandes
2022-06-29 16:56 ` Paul E. McKenney
2022-06-29 19:47 ` Joel Fernandes
2022-06-29 21:07 ` Paul E. McKenney
2022-06-30 14:25 ` Joel Fernandes
2022-06-30 15:29 ` Paul E. McKenney
2022-06-29 16:52 ` Paul E. McKenney
2022-06-26 3:12 ` [PATCH v2 0/8] Implement call_rcu_lazy() and miscellaneous fixes Paul E. McKenney
2022-07-08 4:17 ` Joel Fernandes
2022-07-08 22:45 ` Paul E. McKenney
2022-07-10 1:38 ` Joel Fernandes
2022-07-10 15:47 ` Paul E. McKenney
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=20220629170555.GK1790663@paulmck-ThinkPad-P17-Gen-1 \
--to=paulmck@kernel.org \
--cc=frederic@kernel.org \
--cc=joel@joelfernandes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=neeraj.iitr10@gmail.com \
--cc=rcu@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=rushikesh.s.kadam@intel.com \
--cc=urezki@gmail.com \
--cc=vineeth@bitbyteword.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;
as well as URLs for NNTP newsgroup(s).