From: Uladzislau Rezki <urezki@gmail.com>
To: "Zhang, Qiang" <Qiang.Zhang@windriver.com>
Cc: "Uladzislau Rezki (Sony)" <urezki@gmail.com>,
LKML <linux-kernel@vger.kernel.org>, RCU <rcu@vger.kernel.org>,
"Paul E . McKenney" <paulmck@kernel.org>,
Michael Ellerman <mpe@ellerman.id.au>,
Andrew Morton <akpm@linux-foundation.org>,
Daniel Axtens <dja@axtens.net>,
Frederic Weisbecker <frederic@kernel.org>,
Neeraj Upadhyay <neeraju@codeaurora.org>,
Joel Fernandes <joel@joelfernandes.org>,
Peter Zijlstra <peterz@infradead.org>,
Michal Hocko <mhocko@suse.com>,
Thomas Gleixner <tglx@linutronix.de>,
"Theodore Y . Ts'o" <tytso@mit.edu>,
Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
Oleksiy Avramchenko <oleksiy.avramchenko@sonymobile.com>
Subject: Re: 回复: [PATCH 3/3] kvfree_rcu: use migrate_disable/enable()
Date: Sun, 24 Jan 2021 22:57:00 +0100 [thread overview]
Message-ID: <20210124215700.GB1076@pc636> (raw)
In-Reply-To: <BYAPR11MB263252B1BD73A38DD8C0AF4EFFBF0@BYAPR11MB2632.namprd11.prod.outlook.com>
Hello, Zhang.
> >________________________________________
> >发件人: Uladzislau Rezki (Sony) <urezki@gmail.com>
> >发送时间: 2021年1月21日 0:21
> >收件人: LKML; RCU; Paul E . McKenney; Michael Ellerman
> >抄送: Andrew Morton; Daniel Axtens; Frederic Weisbecker; Neeraj >Upadhyay; Joel Fernandes; Peter Zijlstra; Michal Hocko; Thomas >Gleixner; Theodore Y . Ts'o; Sebastian Andrzej Siewior; Uladzislau >Rezki; Oleksiy Avramchenko
> >主题: [PATCH 3/3] kvfree_rcu: use migrate_disable/enable()
> >
> >Since the page is obtained in a fully preemptible context, dropping
> >the lock can lead to migration onto another CPU. As a result a prev.
> >bnode of that CPU may be underutilised, because a decision has been
> >made for a CPU that was run out of free slots to store a pointer.
> >
> >migrate_disable/enable() are now independent of RT, use it in order
> >to prevent any migration during a page request for a specific CPU it
> >is requested for.
>
>
> Hello Rezki
>
> The critical migrate_disable/enable() area is not allowed to block, under RT and non RT.
> There is such a description in preempt.h
>
>
> * Notes on the implementation.
> *
> * The implementation is particularly tricky since existing code patterns
> * dictate neither migrate_disable() nor migrate_enable() is allowed to block.
> * This means that it cannot use cpus_read_lock() to serialize against hotplug,
> * nor can it easily migrate itself into a pending affinity mask change on
> * migrate_enable().
>
How i interpret it is migrate_enable()/migrate_disable() are not allowed to
use any blocking primitives, such as rwsem/mutexes/etc. in order to mark a
current context as non-migratable.
void migrate_disable(void)
{
struct task_struct *p = current;
if (p->migration_disabled) {
p->migration_disabled++;
return;
}
preempt_disable();
this_rq()->nr_pinned++;
p->migration_disabled = 1;
preempt_enable();
}
It does nothing that prevents you from doing schedule() or even wait for any
event(mutex slow path behaviour), when the process is removed from the run-queue.
I mean after the migrate_disable() is invoked. Or i miss something?
>
> How about the following changes:
>
> diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
> index e7a226abff0d..2aa19537ac7c 100644
> --- a/kernel/rcu/tree.c
> +++ b/kernel/rcu/tree.c
> @@ -3488,12 +3488,10 @@ add_ptr_to_bulk_krc_lock(struct kfree_rcu_cpu **krcp,
> (*krcp)->bkvhead[idx]->nr_records == KVFREE_BULK_MAX_ENTR) {
> bnode = get_cached_bnode(*krcp);
> if (!bnode && can_alloc) {
> - migrate_disable();
> krc_this_cpu_unlock(*krcp, *flags);
> bnode = (struct kvfree_rcu_bulk_data *)
> __get_free_page(GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_NOMEMALLOC | __GFP_NOWARN);
> - *krcp = krc_this_cpu_lock(flags);
> - migrate_enable();
> + raw_spin_lock_irqsave(&(*krcp)->lock, *flags);
>
Hm.. Taking the former lock can lead to a pointer leaking, i mean a CPU associated
with "krcp" might go offline during a page request process, so a queuing occurs on
off-lined CPU. Apat of that, acquiring a former lock still does not solve:
- CPU1 in process of page allocation;
- CPU1 gets migrated to CPU2;
- another task running on CPU1 also allocate a page;
- both bnodes are added to krcp associated with CPU1.
I agree that such scenario probably will never happen or i would say, can be
considered as a corner case. We can drop the:
[PATCH 3/3] kvfree_rcu: use migrate_disable/enable()
and live with: an allocated bnode can be queued to another CPU, so its prev.
"bnode" can be underutilized. What is also can be considered as a corner case.
According to my tests, it is hard to achieve:
Running kvfree_rcu() simultaneously in a tight loop, 1 000 000 allocations/freeing:
- 64 CPUs and 64 threads showed 1 migration;
- 64 CPUs and 128 threads showed 0 migrations;
- 64 CPUs and 32 threads showed 0 migration.
Thoughts?
Thank you for your comments!
--
Vlad Rezki
next prev parent reply other threads:[~2021-01-24 21:58 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-20 16:21 [PATCH 1/3] kvfree_rcu: Allocate a page for a single argument Uladzislau Rezki (Sony)
2021-01-20 16:21 ` [PATCH 2/3] kvfree_rcu: Use __GFP_NOMEMALLOC for single-argument kvfree_rcu() Uladzislau Rezki (Sony)
2021-01-28 18:06 ` Uladzislau Rezki
2021-01-20 16:21 ` [PATCH 3/3] kvfree_rcu: use migrate_disable/enable() Uladzislau Rezki (Sony)
2021-01-20 19:45 ` Sebastian Andrzej Siewior
2021-01-20 21:42 ` Paul E. McKenney
2021-01-23 9:31 ` 回复: " Zhang, Qiang
2021-01-24 21:57 ` Uladzislau Rezki [this message]
2021-01-25 1:50 ` 回复: " Zhang, Qiang
2021-01-25 2:18 ` Zhang, Qiang
2021-01-25 13:49 ` Uladzislau Rezki
2021-01-26 9:33 ` 回复: " Zhang, Qiang
2021-01-26 13:43 ` Uladzislau Rezki
2021-01-20 18:40 ` [PATCH 1/3] kvfree_rcu: Allocate a page for a single argument Paul E. McKenney
2021-01-20 19:57 ` Sebastian Andrzej Siewior
2021-01-20 21:54 ` Paul E. McKenney
2021-01-21 13:35 ` Uladzislau Rezki
2021-01-21 15:07 ` Paul E. McKenney
2021-01-21 19:17 ` Uladzislau Rezki
2021-01-22 11:17 ` Sebastian Andrzej Siewior
2021-01-22 15:28 ` Paul E. McKenney
2021-01-21 12:38 ` Uladzislau Rezki
2021-01-22 11:34 ` Sebastian Andrzej Siewior
2021-01-22 14:21 ` Uladzislau Rezki
2021-01-25 13:22 ` Michal Hocko
2021-01-25 14:31 ` Uladzislau Rezki
2021-01-25 15:39 ` Michal Hocko
2021-01-25 16:25 ` Uladzislau Rezki
2021-01-28 15:11 ` Uladzislau Rezki
2021-01-28 15:17 ` Michal Hocko
2021-01-28 15:30 ` Uladzislau Rezki
2021-01-28 18:02 ` Uladzislau Rezki
[not found] ` <YBPNvbJLg56XU8co@dhcp22.suse.cz>
2021-01-29 16:35 ` Uladzislau Rezki
2021-02-01 11:47 ` Michal Hocko
2021-02-01 14:44 ` Uladzislau Rezki
2021-02-03 19:37 ` 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=20210124215700.GB1076@pc636 \
--to=urezki@gmail.com \
--cc=Qiang.Zhang@windriver.com \
--cc=akpm@linux-foundation.org \
--cc=bigeasy@linutronix.de \
--cc=dja@axtens.net \
--cc=frederic@kernel.org \
--cc=joel@joelfernandes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mhocko@suse.com \
--cc=mpe@ellerman.id.au \
--cc=neeraju@codeaurora.org \
--cc=oleksiy.avramchenko@sonymobile.com \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=rcu@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=tytso@mit.edu \
/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).