From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: Andy Lutomirski <luto@kernel.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
Stephen Rothwell <sfr@canb.auug.org.au>,
Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@elte.hu>,
"H. Peter Anvin" <hpa@zytor.com>,
Peter Zijlstra <peterz@infradead.org>,
Linux-Next Mailing List <linux-next@vger.kernel.org>,
linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: linux-next: manual merge of the rcu tree with the tip tree
Date: Tue, 1 Aug 2017 15:40:32 +0000 (UTC) [thread overview]
Message-ID: <1205506069.1647.1501602032052.JavaMail.zimbra@efficios.com> (raw)
In-Reply-To: <CALCETrWWx8h=74td8bxbR6KZJN2GLAZ_jz9G-x9Q+Vn2AHSijQ@mail.gmail.com>
----- On Aug 1, 2017, at 10:15 AM, Andy Lutomirski luto@kernel.org wrote:
> On Tue, Aug 1, 2017 at 7:02 AM, Mathieu Desnoyers
> <mathieu.desnoyers@efficios.com> wrote:
>> /*
>> * The full memory barrier implied by mm_cpumask update operations
>> * is required by the membarrier system call.
>> */
>>
>> What we want to order here is:
>>
>> prev userspace memory accesses
>> schedule
>> <full mb> (it's already there) [A]
>> update to rq->curr changing the rq->curr->mm value
>> <full mb> (provided by mm_cpumask updates in switch_mm on x86) [B]
>
> If I understand this right, the issue with relying on CR3 writes is
> that the target CPU could switch to a kernel thread and back to the
> same user mm white the membarrier caller is reading its mm, right?
The current implementation of context_switch() does:
mm = next->mm;
oldmm = prev->active_mm;
if (!mm)
next->active_mm = oldmm;
if (!prev->mm) {
prev->active_mm = NULL;
rq->prev_mm = oldmm;
}
so basically the only way to have a non-null rq->prev_mm when we
reach finish_task_switch() is to have a non-null prev->active_mm
in context_switch (kernel thread).
finish_task_switch() has:
struct mm_struct *mm = rq->prev_mm;
[...]
if (mm)
mmdrop(mm);
which issues a full memory barrier through atomic_dec_and_test(). This
happens to take care of this kthread->uthread scenario. I think it would
be important to document though.
Thanks,
Mathieu
--
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
next prev parent reply other threads:[~2017-08-01 15:40 UTC|newest]
Thread overview: 71+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-07-31 3:50 linux-next: manual merge of the rcu tree with the tip tree Stephen Rothwell
2017-07-31 16:13 ` Paul E. McKenney
2017-08-01 0:04 ` Mathieu Desnoyers
2017-08-01 4:03 ` Paul E. McKenney
2017-08-01 4:25 ` Mathieu Desnoyers
2017-08-01 16:31 ` Paul E. McKenney
2017-08-01 13:43 ` Andy Lutomirski
2017-08-01 13:58 ` Peter Zijlstra
2017-08-01 14:15 ` Peter Zijlstra
2017-08-01 14:17 ` Andy Lutomirski
2017-08-01 14:02 ` Mathieu Desnoyers
2017-08-01 14:15 ` Andy Lutomirski
2017-08-01 15:40 ` Mathieu Desnoyers [this message]
2017-08-01 21:36 ` Paul E. McKenney
-- strict thread matches above, loose matches on Subject: below --
2024-02-27 1:55 Stephen Rothwell
2022-04-06 2:45 Stephen Rothwell
2022-04-06 16:28 ` Paul E. McKenney
2022-02-21 18:17 broonie
2021-10-12 4:48 Stephen Rothwell
2021-10-13 16:31 ` Paul E. McKenney
2021-08-17 7:09 Stephen Rothwell
2021-06-23 5:33 Stephen Rothwell
2021-06-22 4:51 Stephen Rothwell
2021-06-22 4:47 Stephen Rothwell
2021-06-22 5:04 ` Stephen Rothwell
2021-06-22 17:10 ` Paul E. McKenney
2020-10-09 4:59 Stephen Rothwell
2020-07-29 6:23 Stephen Rothwell
2020-06-26 3:14 Stephen Rothwell
2020-06-25 2:44 Stephen Rothwell
2020-06-25 3:44 ` Paul E. McKenney
2020-06-24 3:04 Stephen Rothwell
2020-06-24 4:06 ` Paul E. McKenney
2020-05-29 6:22 Stephen Rothwell
2020-05-29 6:41 ` Stephen Rothwell
2020-05-29 14:15 ` Paul E. McKenney
2020-05-29 23:38 ` Stephen Rothwell
2020-03-25 3:08 Stephen Rothwell
2020-03-25 3:18 ` Paul E. McKenney
2020-03-25 21:31 ` Paul E. McKenney
2019-12-19 0:50 Stephen Rothwell
2019-12-19 1:27 ` Paul E. McKenney
2019-12-19 1:31 ` Paul E. McKenney
2019-12-19 8:41 ` Peter Zijlstra
2019-12-19 13:38 ` Paul E. McKenney
2019-12-16 23:37 Stephen Rothwell
2018-06-22 2:27 Stephen Rothwell
2018-06-26 19:33 ` Paul E. McKenney
2017-11-10 2:14 Stephen Rothwell
2017-08-22 4:13 Stephen Rothwell
2016-07-18 5:26 Stephen Rothwell
2016-07-19 3:00 ` Paul E. McKenney
2016-06-09 5:14 Stephen Rothwell
2016-06-09 15:59 ` Paul E. McKenney
2016-03-04 4:13 Stephen Rothwell
2016-03-04 15:04 ` Paul E. McKenney
2015-07-16 2:57 Stephen Rothwell
2015-05-07 3:56 Stephen Rothwell
2014-02-24 4:18 Stephen Rothwell
2014-02-24 4:42 ` Paul E. McKenney
2012-09-05 3:59 Stephen Rothwell
2012-09-05 16:39 ` Paul E. McKenney
2012-09-05 17:11 ` Peter Zijlstra
2012-08-23 3:01 Stephen Rothwell
2012-08-23 3:22 ` Paul E. McKenney
2012-08-22 4:27 Stephen Rothwell
2012-08-22 5:05 ` Paul E. McKenney
2012-08-22 4:27 Stephen Rothwell
2012-08-22 5:03 ` Paul E. McKenney
2011-06-20 4:47 Stephen Rothwell
2011-06-20 15:17 ` 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=1205506069.1647.1501602032052.JavaMail.zimbra@efficios.com \
--to=mathieu.desnoyers@efficios.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-next@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@elte.hu \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=sfr@canb.auug.org.au \
--cc=tglx@linutronix.de \
/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).