From: Frederic Weisbecker <frederic@kernel.org>
To: nicolas saenz julienne <nsaenz@kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Phil Auld <pauld@redhat.com>, Alex Belits <abelits@marvell.com>,
Xiongfeng Wang <wangxiongfeng2@huawei.com>,
Neeraj Upadhyay <quic_neeraju@quicinc.com>,
Thomas Gleixner <tglx@linutronix.de>,
Yu Liao <liaoyu15@huawei.com>, Boqun Feng <boqun.feng@gmail.com>,
"Paul E . McKenney" <paulmck@kernel.org>,
Marcelo Tosatti <mtosatti@redhat.com>,
Paul Gortmaker <paul.gortmaker@windriver.com>,
Uladzislau Rezki <uladzislau.rezki@sony.com>,
Joel Fernandes <joel@joelfernandes.org>,
Mark Rutland <mark.rutland@arm.com>
Subject: Re: [PATCH 20/21] rcu/context_tracking: Merge dynticks counter and context tracking states
Date: Wed, 8 Jun 2022 16:29:27 +0200 [thread overview]
Message-ID: <20220608142927.GA1735086@lothringen> (raw)
In-Reply-To: <bcc4c9fa41ace6f9d98d88d24d1bd67a469cbfeb.camel@kernel.org>
On Tue, May 31, 2022 at 06:15:36PM +0200, nicolas saenz julienne wrote:
> On Tue, 2022-05-31 at 16:23 +0200, Frederic Weisbecker wrote:
> > But idle at least is an exception and CONTEXT_IDLE will remain during the
> > interrupt handling. It's not that trivial to handle the idle case because
> > ct_irq_exit() needs to know that it is called between ct_idle_enter() and
> > ct_idle_exit().
>
> Just for the record, this behaviour was already here regardless of this series,
> so it's not something it needs to fix.
Right.
>
> Something like this should work, right?
>
> ct_idle_enter()
> //IRQ or NMI
> if (__ct_state() == CONTEXT_IDLE)
> ct_idle_exit()
Right but that's one more costly operation (atomic_add_return())
> ct_irq_enter()
Ideally this should increment by RCU_DYNTICKS_IDX - CONTEXT_IDLE
> ...
> ct_irq_exit()
And this should increment by RCU_DYNTICKS_IDX + CONTEXT_IDLE
I guess the CONTEXT_IDLE state should be remembered on some per cpu
variable somewhere.
BTW one interesting optimization to do when an idle interrupt leads to
setting need_resched() would be to have:
idle_loop() {
while (!need_resched) {
rcu_idle_enter();
mwait();
//IRQ {
rcu_irq_enter();
do_irq()... //set need_resched()
rcu_irq_exit() // but no need to do the atomic_add_return() here
// since we want to keep RCU watching as we'll
// escape from idle
}
rcu_idle_exit() // and no need to do the atomic_add_return() here either
That's two expensive operations spared for a pretty common event.
> if (needs_update_state()) //using irqentry_state_t for ex.
> ct_idle_entry()
> ct_idle_exit()
>
> Note that it's not a big issue as we can work around this behaviour by checking
> through dynticks whether a CPU is really idle.
>
> Do you think it's worth fixing nonetheless?
Nothing urgent for sure.
>
> Regards,
> Nicolas
next prev parent reply other threads:[~2022-06-08 14:29 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-05-19 14:58 [PATCH 00/21] rcu/context-tracking: Merge RCU eqs-dynticks counter to context tracking v3 Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 01/21] context_tracking: Remove unused context_tracking_in_user() Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 02/21] rcu: Tag rcu_irq_*_irqson() as noinstr Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 03/21] context_tracking: Add a note about noinstr VS unsafe context tracking functions Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 04/21] context_tracking: Rename __context_tracking_enter/exit() to __ct_user_enter/exit() Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 05/21] context_tracking: Rename context_tracking_user_enter/exit() to user_enter/exit_callable() Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 06/21] context_tracking: Rename context_tracking_enter/exit() to ct_user_enter/exit() Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 07/21] context_tracking: Rename context_tracking_cpu_set() to ct_cpu_track_user() Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 08/21] context_tracking: Split user tracking Kconfig Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 09/21] context_tracking: Take idle eqs entrypoints over RCU Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 10/21] context_tracking: Take IRQ " Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 11/21] context_tracking: Take NMI " Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 12/21] rcu/context-tracking: Remove rcu_irq_enter/exit() Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 13/21] rcu/context_tracking: Move dynticks counter to context tracking Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 14/21] rcu/context_tracking: Move dynticks_nesting " Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 15/21] rcu/context_tracking: Move dynticks_nmi_nesting " Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 16/21] rcu/context-tracking: Move deferred nocb resched " Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 17/21] rcu/context-tracking: Move RCU-dynticks internal functions to context_tracking Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 18/21] rcu/context-tracking: Remove unused and/or unecessary middle functions Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 19/21] context_tracking: Convert state to atomic_t Frederic Weisbecker
2022-05-19 14:58 ` [PATCH 20/21] rcu/context_tracking: Merge dynticks counter and context tracking states Frederic Weisbecker
2022-05-30 18:02 ` nicolas saenz julienne
2022-05-31 14:23 ` Frederic Weisbecker
2022-05-31 16:15 ` nicolas saenz julienne
2022-06-08 14:29 ` Frederic Weisbecker [this message]
2022-06-08 17:43 ` nicolas saenz julienne
2022-05-19 14:58 ` [PATCH 21/21] MAINTAINERS: Add Paul as context tracking maintainer Frederic Weisbecker
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=20220608142927.GA1735086@lothringen \
--to=frederic@kernel.org \
--cc=abelits@marvell.com \
--cc=boqun.feng@gmail.com \
--cc=joel@joelfernandes.org \
--cc=liaoyu15@huawei.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mtosatti@redhat.com \
--cc=nsaenz@kernel.org \
--cc=paul.gortmaker@windriver.com \
--cc=pauld@redhat.com \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=quic_neeraju@quicinc.com \
--cc=tglx@linutronix.de \
--cc=uladzislau.rezki@sony.com \
--cc=wangxiongfeng2@huawei.com \
/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).