From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Alan Stern <stern@rowland.harvard.edu>
Cc: Daniel Lustig <dlustig@nvidia.com>,
Will Deacon <will.deacon@arm.com>,
Andrea Parri <andrea.parri@amarulasolutions.com>,
Andrea Parri <parri.andrea@gmail.com>,
Kernel development list <linux-kernel@vger.kernel.org>,
linux-arch@vger.kernel.org, mingo@kernel.org,
peterz@infradead.org, boqun.feng@gmail.com, npiggin@gmail.com,
dhowells@redhat.com, Jade Alglave <j.alglave@ucl.ac.uk>,
Luc Maranget <luc.maranget@inria.fr>,
akiyks@gmail.com, Palmer Dabbelt <palmer@sifive.com>
Subject: Re: [PATCH RFC LKMM 1/7] tools/memory-model: Add extra ordering for locks and remove it for ordinary release/acquire
Date: Tue, 11 Sep 2018 13:03:28 -0700 [thread overview]
Message-ID: <20180911200328.GA4225@linux.vnet.ibm.com> (raw)
Message-ID: <20180911200328.n5p0g-kjeu3LUMg3BZYnJbGGu3f7r7ZCDmOBY8C3Yt0@z> (raw)
In-Reply-To: <Pine.LNX.4.44L0.1809111455420.1461-100000@iolanthe.rowland.org>
On Tue, Sep 11, 2018 at 03:31:53PM -0400, Alan Stern wrote:
> On Thu, 12 Jul 2018, Paul E. McKenney wrote:
>
> > > > Take for instance the pattern where RCU relies on RCsc locks, this is an
> > > > entirely simple and straight forward use of locks, yet completely fails
> > > > on this subtle point.
> > >
> > > Do you happen to remember exactly where in the kernel source this
> > > occurs?
> >
> > Look for the uses of raw_spin_lock_irq_rcu_node() and friends in
> > kernel/rcu and include/linux/*rcu*, along with the explanation in
> > Documentation/RCU/Design/Memory-Ordering/Tree-RCU-Memory-Ordering.html
>
> I just now started looking at this for the first time, and I was struck
> by the sloppy thinking displayed in the very first paragraph of the
> HTML document! For example, consider the third sentence:
>
> Similarly, any code that happens before the beginning of a
> given RCU grace period is guaranteed to see the effects of all
> accesses following the end of that grace period that are within
> RCU read-side critical sections.
>
> Is RCU now a time machine? :-)
Why not? ;-)
> I think what you meant to write in the second and third sentences was
> something more like this:
>
> Any code in an RCU critical section that extends beyond the
> end of a given RCU grace period is guaranteed to see the
> effects of all accesses which were visible to the grace
> period's CPU before the start of the grace period. Similarly,
> any code that follows an RCU grace period (on the grace
> period's CPU) is guaranteed to see the effects of all accesses
> which were visible to an RCU critical section that began
> before the start of the grace period.
That looks to me to be an improvement, other than that the "(on the
grace period's CPU)" seems a bit restrictive -- you could for example
have a release-acquire chain starting after the grace period, right?
> Also, the document doesn't seem to explain how Tree RCU relies on the
> lock-ordering guarantees of raw_spin_lock_rcu_node() and friends. It
> _says_ that these guarantees are used, but not how or where. (Unless I
> missed something; I didn't read the document all that carefully.)
The closest is this sentence: "But the only part of rcu_prepare_for_idle()
that really matters for this discussion are lines 37–39", which
refers to this code:
37 raw_spin_lock_rcu_node(rnp);
38 needwake = rcu_accelerate_cbs(rsp, rnp, rdp);
39 raw_spin_unlock_rcu_node(rnp);
I could add a sentence explaining the importance of the
smp_mb__after_unlock_lock() -- is that what you are getting at?
> In any case, you should bear in mind that the lock ordering provided by
> Peter's raw_spin_lock_rcu_node() and friends is not the same as what we
> have been discussing for the LKMM:
>
> Peter's routines are meant for the case where you release
> one lock and then acquire another (for example, locks in
> two different levels of the RCU tree).
>
> The LKMM patch applies only to cases where one CPU releases
> a lock and then that CPU or another acquires the _same_ lock
> again.
>
> As another difference, the litmus test given near the start of the
> "Tree RCU Grace Period Memory Ordering Building Blocks" section would
> not be forbidden by the LKMM, even with RCtso locks, if it didn't use
> raw_spin_lock_rcu_node(). This is because the litmus test is forbidden
> only when locks are RCsc, which is what raw_spin_lock_rcu_node()
> provides.
Agreed.
> So I don't see how the RCU code can be held up as an example either for
> or against requiring locks to be RCtso.
Agreed again. The use of smp_mb__after_unlock_lock() instead
provides RCsc. But this use case is deemed sufficiently rare that
smp_mb__after_unlock_lock() is defined within RCU.
Thanx, Paul
next prev parent reply other threads:[~2018-09-12 1:04 UTC|newest]
Thread overview: 112+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-08-29 21:10 [PATCH RFC memory-model 0/7] Memory-model changes Paul E. McKenney
2018-08-29 21:10 ` Paul E. McKenney
2018-08-29 21:10 ` [PATCH RFC LKMM 1/7] tools/memory-model: Add extra ordering for locks and remove it for ordinary release/acquire Paul E. McKenney
2018-08-29 21:10 ` Paul E. McKenney
2018-08-30 12:50 ` Andrea Parri
2018-08-30 12:50 ` Andrea Parri
2018-08-30 21:31 ` Alan Stern
2018-08-30 21:31 ` Alan Stern
2018-08-31 9:17 ` Andrea Parri
2018-08-31 9:17 ` Andrea Parri
2018-08-31 14:52 ` Alan Stern
2018-08-31 14:52 ` Alan Stern
2018-08-31 16:06 ` Will Deacon
2018-08-31 16:06 ` Will Deacon
2018-08-31 18:28 ` Andrea Parri
2018-08-31 18:28 ` Andrea Parri
2018-09-03 9:01 ` Andrea Parri
2018-09-03 9:01 ` Andrea Parri
2018-09-03 17:04 ` Will Deacon
2018-09-03 17:04 ` Will Deacon
2018-09-04 8:11 ` Andrea Parri
2018-09-04 8:11 ` Andrea Parri
2018-09-04 19:09 ` Alan Stern
2018-09-04 19:09 ` Alan Stern
2018-09-05 7:21 ` Andrea Parri
2018-09-05 7:21 ` Andrea Parri
2018-09-05 14:33 ` Akira Yokosawa
2018-09-05 14:33 ` Akira Yokosawa
2018-09-05 14:53 ` Paul E. McKenney
2018-09-05 14:53 ` Paul E. McKenney
2018-09-05 15:00 ` Andrea Parri
2018-09-05 15:00 ` Andrea Parri
2018-09-05 15:04 ` Akira Yokosawa
2018-09-05 15:04 ` Akira Yokosawa
2018-09-05 15:24 ` Andrea Parri
2018-09-05 15:24 ` Andrea Parri
2018-09-03 17:52 ` Alan Stern
2018-09-03 17:52 ` Alan Stern
2018-09-03 18:28 ` Andrea Parri
2018-09-03 18:28 ` Andrea Parri
2018-09-06 1:25 ` Alan Stern
2018-09-06 1:25 ` Alan Stern
2018-09-06 9:36 ` Andrea Parri
2018-09-06 9:36 ` Andrea Parri
2018-09-07 16:00 ` Alan Stern
2018-09-07 16:00 ` Alan Stern
2018-09-07 16:09 ` Will Deacon
2018-09-07 16:09 ` Will Deacon
2018-09-07 16:39 ` Daniel Lustig
2018-09-07 16:39 ` Daniel Lustig
2018-09-07 17:38 ` Alan Stern
2018-09-07 17:38 ` Alan Stern
2018-09-08 0:04 ` Daniel Lustig
2018-09-08 0:04 ` Daniel Lustig
2018-09-08 9:58 ` Andrea Parri
2018-09-08 9:58 ` Andrea Parri
2018-09-11 19:31 ` Alan Stern
2018-09-11 19:31 ` Alan Stern
2018-09-11 20:03 ` Paul E. McKenney [this message]
2018-09-11 20:03 ` Paul E. McKenney
2018-09-12 14:24 ` Alan Stern
2018-09-12 14:24 ` Alan Stern
2018-09-13 17:07 ` Alan Stern
2018-09-13 17:07 ` Alan Stern
2018-09-14 14:37 ` Andrea Parri
2018-09-14 14:37 ` Andrea Parri
2018-09-14 16:29 ` Alan Stern
2018-09-14 16:29 ` Alan Stern
2018-09-14 19:44 ` Andrea Parri
2018-09-14 19:44 ` Andrea Parri
2018-09-14 21:08 ` [PATCH v5] " Alan Stern
2018-09-14 21:08 ` Alan Stern
2018-09-15 3:56 ` Paul E. McKenney
2018-09-15 3:56 ` Paul E. McKenney
2018-09-03 17:05 ` [PATCH RFC LKMM 1/7] " Will Deacon
2018-09-03 17:05 ` Will Deacon
2018-08-31 17:55 ` Andrea Parri
2018-08-31 17:55 ` Andrea Parri
2018-08-29 21:10 ` [PATCH RFC LKMM 2/7] doc: Replace smp_cond_acquire() with smp_cond_load_acquire() Paul E. McKenney
2018-08-29 21:10 ` Paul E. McKenney
2018-09-14 16:59 ` Will Deacon
2018-09-14 16:59 ` Will Deacon
2018-09-14 18:20 ` Paul E. McKenney
2018-09-14 18:20 ` Paul E. McKenney
2018-08-29 21:10 ` [PATCH RFC LKMM 3/7] EXP tools/memory-model: Add more LKMM limitations Paul E. McKenney
2018-08-29 21:10 ` Paul E. McKenney
2018-08-30 9:17 ` Andrea Parri
2018-08-30 9:17 ` Andrea Parri
2018-08-30 22:18 ` Paul E. McKenney
2018-08-30 22:18 ` Paul E. McKenney
2018-08-31 9:43 ` Andrea Parri
2018-08-31 9:43 ` Andrea Parri
2018-09-06 18:34 ` Paul E. McKenney
2018-09-06 18:34 ` Paul E. McKenney
2018-08-29 21:10 ` [PATCH RFC LKMM 4/7] tools/memory-model: Fix a README typo Paul E. McKenney
2018-08-29 21:10 ` Paul E. McKenney
2018-08-29 21:10 ` [PATCH RFC LKMM 5/7] EXP tools/memory-model: Add scripts to check github litmus tests Paul E. McKenney
2018-08-29 21:10 ` Paul E. McKenney
2018-08-29 21:10 ` [PATCH RFC LKMM 6/7] EXP tools/memory-model: Make scripts take "-j" abbreviation for "--jobs" Paul E. McKenney
2018-08-29 21:10 ` Paul E. McKenney
2018-08-29 21:10 ` [PATCH RFC LKMM 7/7] EXP tools/memory-model: Add .cfg and .cat files for s390 Paul E. McKenney
2018-08-29 21:10 ` Paul E. McKenney
2018-08-31 16:06 ` Will Deacon
2018-08-31 16:06 ` Will Deacon
2018-09-01 17:08 ` Paul E. McKenney
2018-09-01 17:08 ` Paul E. McKenney
2018-09-14 16:36 ` [PATCH RFC memory-model 0/7] Memory-model changes Paul E. McKenney
2018-09-14 16:36 ` Paul E. McKenney
2018-09-14 17:19 ` Alan Stern
2018-09-14 17:19 ` Alan Stern
2018-09-14 18:29 ` Paul E. McKenney
2018-09-14 18:29 ` 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=20180911200328.GA4225@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=akiyks@gmail.com \
--cc=andrea.parri@amarulasolutions.com \
--cc=boqun.feng@gmail.com \
--cc=dhowells@redhat.com \
--cc=dlustig@nvidia.com \
--cc=j.alglave@ucl.ac.uk \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luc.maranget@inria.fr \
--cc=mingo@kernel.org \
--cc=npiggin@gmail.com \
--cc=palmer@sifive.com \
--cc=parri.andrea@gmail.com \
--cc=peterz@infradead.org \
--cc=stern@rowland.harvard.edu \
--cc=will.deacon@arm.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).