From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Mikulas Patocka <mpatocka@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Oleg Nesterov <oleg@redhat.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@elte.hu>,
Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
Ananth N Mavinakayanahalli <ananth@in.ibm.com>,
Anton Arapov <anton@redhat.com>,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH RESEND v2 1/1] percpu_rw_semaphore: reimplement to not block the readers unnecessarily
Date: Thu, 8 Nov 2012 19:23:10 -0800 [thread overview]
Message-ID: <20121109032310.GA2438@linux.vnet.ibm.com> (raw)
In-Reply-To: <20121109004136.GH2519@linux.vnet.ibm.com>
On Thu, Nov 08, 2012 at 04:41:36PM -0800, Paul E. McKenney wrote:
> On Thu, Nov 08, 2012 at 06:41:10PM -0500, Mikulas Patocka wrote:
> >
> >
> > On Thu, 8 Nov 2012, Paul E. McKenney wrote:
> >
> > > On Thu, Nov 08, 2012 at 12:07:00PM -0800, Andrew Morton wrote:
> > > > On Thu, 8 Nov 2012 14:48:49 +0100
> > > > Oleg Nesterov <oleg@redhat.com> wrote:
> > > >
> > > > > Currently the writer does msleep() plus synchronize_sched() 3 times
> > > > > to acquire/release the semaphore, and during this time the readers
> > > > > are blocked completely. Even if the "write" section was not actually
> > > > > started or if it was already finished.
> > > > >
> > > > > With this patch down_write/up_write does synchronize_sched() twice
> > > > > and down_read/up_read are still possible during this time, just they
> > > > > use the slow path.
> > > > >
> > > > > percpu_down_write() first forces the readers to use rw_semaphore and
> > > > > increment the "slow" counter to take the lock for reading, then it
> > > > > takes that rw_semaphore for writing and blocks the readers.
> > > > >
> > > > > Also. With this patch the code relies on the documented behaviour of
> > > > > synchronize_sched(), it doesn't try to pair synchronize_sched() with
> > > > > barrier.
> > > > >
> > > > > ...
> > > > >
> > > > > include/linux/percpu-rwsem.h | 83 +++++------------------------
> > > > > lib/Makefile | 2 +-
> > > > > lib/percpu-rwsem.c | 123 ++++++++++++++++++++++++++++++++++++++++++
> > > >
> > > > The patch also uninlines everything.
> > > >
> > > > And it didn't export the resulting symbols to modules, so it isn't an
> > > > equivalent. We can export thing later if needed I guess.
> > > >
> > > > It adds percpu-rwsem.o to lib-y, so the CONFIG_BLOCK=n kernel will
> > > > avoid including the code altogether, methinks?
> > > >
> > > > >
> > > > > ...
> > > > >
> > > > > --- /dev/null
> > > > > +++ b/lib/percpu-rwsem.c
> > > > > @@ -0,0 +1,123 @@
> > > >
> > > > That was nice and terse ;)
> > > >
> > > > > +#include <linux/percpu-rwsem.h>
> > > > > +#include <linux/rcupdate.h>
> > > > > +#include <linux/sched.h>
> > > >
> > > > This list is nowhere near sufficient to support this file's
> > > > requirements. atomic.h, percpu.h, rwsem.h, wait.h, errno.h and plenty
> > > > more. IOW, if it compiles, it was sheer luck.
> > > >
> > > > > +int percpu_init_rwsem(struct percpu_rw_semaphore *brw)
> > > > > +{
> > > > > + brw->fast_read_ctr = alloc_percpu(int);
> > > > > + if (unlikely(!brw->fast_read_ctr))
> > > > > + return -ENOMEM;
> > > > > +
> > > > > + mutex_init(&brw->writer_mutex);
> > > > > + init_rwsem(&brw->rw_sem);
> > > > > + atomic_set(&brw->slow_read_ctr, 0);
> > > > > + init_waitqueue_head(&brw->write_waitq);
> > > > > + return 0;
> > > > > +}
> > > > > +
> > > > > +void percpu_free_rwsem(struct percpu_rw_semaphore *brw)
> > > > > +{
> > > > > + free_percpu(brw->fast_read_ctr);
> > > > > + brw->fast_read_ctr = NULL; /* catch use after free bugs */
> > > > > +}
> > > > > +
> > > > > +static bool update_fast_ctr(struct percpu_rw_semaphore *brw, unsigned int val)
> > > > > +{
> > > > > + bool success = false;
> > > > > +
> > > > > + preempt_disable();
> > > > > + if (likely(!mutex_is_locked(&brw->writer_mutex))) {
> > > > > + __this_cpu_add(*brw->fast_read_ctr, val);
> > > > > + success = true;
> > > > > + }
> > > > > + preempt_enable();
> > > > > +
> > > > > + return success;
> > > > > +}
> > > > > +
> > > > > +/*
> > > > > + * Like the normal down_read() this is not recursive, the writer can
> > > > > + * come after the first percpu_down_read() and create the deadlock.
> > > > > + */
> > > > > +void percpu_down_read(struct percpu_rw_semaphore *brw)
> > > > > +{
> > > > > + if (likely(update_fast_ctr(brw, +1)))
> > > > > + return;
> > > > > +
> > > > > + down_read(&brw->rw_sem);
> > > > > + atomic_inc(&brw->slow_read_ctr);
> > > > > + up_read(&brw->rw_sem);
> > > > > +}
> > > > > +
> > > > > +void percpu_up_read(struct percpu_rw_semaphore *brw)
> > > > > +{
> > > > > + if (likely(update_fast_ctr(brw, -1)))
> > > > > + return;
> > > > > +
> > > > > + /* false-positive is possible but harmless */
> > > > > + if (atomic_dec_and_test(&brw->slow_read_ctr))
> > > > > + wake_up_all(&brw->write_waitq);
> > > > > +}
> > > > > +
> > > > > +static int clear_fast_ctr(struct percpu_rw_semaphore *brw)
> > > > > +{
> > > > > + unsigned int sum = 0;
> > > > > + int cpu;
> > > > > +
> > > > > + for_each_possible_cpu(cpu) {
> > > > > + sum += per_cpu(*brw->fast_read_ctr, cpu);
> > > > > + per_cpu(*brw->fast_read_ctr, cpu) = 0;
> > > > > + }
> > > > > +
> > > > > + return sum;
> > > > > +}
> > > > > +
> > > > > +/*
> > > > > + * A writer takes ->writer_mutex to exclude other writers and to force the
> > > > > + * readers to switch to the slow mode, note the mutex_is_locked() check in
> > > > > + * update_fast_ctr().
> > > > > + *
> > > > > + * After that the readers can only inc/dec the slow ->slow_read_ctr counter,
> > > > > + * ->fast_read_ctr is stable. Once the writer moves its sum into the slow
> > > > > + * counter it represents the number of active readers.
> > > > > + *
> > > > > + * Finally the writer takes ->rw_sem for writing and blocks the new readers,
> > > > > + * then waits until the slow counter becomes zero.
> > > > > + */
> > > >
> > > > Some overview of how fast/slow_read_ctr are supposed to work would be
> > > > useful. This comment seems to assume that the reader already knew
> > > > that.
> > > >
> > > > > +void percpu_down_write(struct percpu_rw_semaphore *brw)
> > > > > +{
> > > > > + /* also blocks update_fast_ctr() which checks mutex_is_locked() */
> > > > > + mutex_lock(&brw->writer_mutex);
> > > > > +
> > > > > + /*
> > > > > + * 1. Ensures mutex_is_locked() is visible to any down_read/up_read
> > > > > + * so that update_fast_ctr() can't succeed.
> > > > > + *
> > > > > + * 2. Ensures we see the result of every previous this_cpu_add() in
> > > > > + * update_fast_ctr().
> > > > > + *
> > > > > + * 3. Ensures that if any reader has exited its critical section via
> > > > > + * fast-path, it executes a full memory barrier before we return.
> > > > > + */
> > > > > + synchronize_sched();
> > > >
> > > > Here's where I get horridly confused. Your patch completely deRCUifies
> > > > this code, yes? Yet here we're using an RCU primitive. And we seem to
> > > > be using it not as an RCU primitive but as a handy thing which happens
> > > > to have desirable side-effects. But the implementation of
> > > > synchronize_sched() differs considerably according to which rcu
> > > > flavor-of-the-minute you're using.
> > >
> > > The trick is that the preempt_disable() call in update_fast_ctr()
> > > acts as an RCU read-side critical section WRT synchronize_sched().
> > >
> > > The algorithm would work given rcu_read_lock()/rcu_read_unlock() and
> > > synchronize_rcu() in place of preempt_disable()/preempt_enable() and
> > > synchronize_sched(). The real-time guys would prefer the change
> > > to rcu_read_lock()/rcu_read_unlock() and synchronize_rcu(), now that
> > > you mention it.
> > >
> > > Oleg, Mikulas, any reason not to move to rcu_read_lock()/rcu_read_unlock()
> > > and synchronize_rcu()?
> >
> > preempt_disable/preempt_enable is faster than
> > rcu_read_lock/rcu_read_unlock for preemptive kernels.
>
> Significantly faster in this case? Can you measure the difference
> from a user-mode test?
>
> Hmmm. I have been avoiding moving the preemptible-RCU state from
> task_struct to thread_info, but if the difference really matters,
> perhaps that needs to be done.
Actually, the fact that __this_cpu_add() will malfunction on some
architectures is preemption is not disabled seems a more compelling
reason to keep preempt_enable() than any performance improvement. ;-)
Thanx, Paul
> > Regarding real-time response - the region blocked with
> > preempt_disable/preempt_enable contains a few instructions (one test for
> > mutex_is_locked and one increment of percpu variable), so it isn't any
> > threat to real time response. There are plenty of longer regions in the
> > kernel that are executed with interrupts or preemption disabled.
>
> Careful. The real-time guys might take the same every-little-bit approach
> to latency that you seem to be taking for CPU cycles. ;-)
>
> Thanx, Paul
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
next prev parent reply other threads:[~2012-11-09 3:23 UTC|newest]
Thread overview: 103+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-15 19:09 [RFC PATCH 0/2] uprobes: register/unregister can race with fork Oleg Nesterov
2012-10-15 19:10 ` [PATCH 1/2] brw_mutex: big read-write mutex Oleg Nesterov
2012-10-15 23:28 ` Paul E. McKenney
2012-10-16 15:56 ` Oleg Nesterov
2012-10-16 18:58 ` Paul E. McKenney
2012-10-17 16:37 ` Oleg Nesterov
2012-10-17 22:28 ` Paul E. McKenney
2012-10-16 19:56 ` Linus Torvalds
2012-10-17 16:59 ` Oleg Nesterov
2012-10-17 22:44 ` Paul E. McKenney
2012-10-18 16:24 ` Oleg Nesterov
2012-10-18 16:38 ` Paul E. McKenney
2012-10-18 17:57 ` Oleg Nesterov
2012-10-18 19:28 ` Mikulas Patocka
2012-10-19 12:38 ` Peter Zijlstra
2012-10-19 15:32 ` Mikulas Patocka
2012-10-19 17:40 ` Peter Zijlstra
2012-10-19 17:57 ` Oleg Nesterov
2012-10-19 22:54 ` Mikulas Patocka
2012-10-24 3:08 ` Dave Chinner
2012-10-25 14:09 ` Mikulas Patocka
2012-10-25 23:40 ` Dave Chinner
2012-10-26 12:06 ` Oleg Nesterov
2012-10-26 13:22 ` Mikulas Patocka
2012-10-26 14:12 ` Oleg Nesterov
2012-10-26 15:23 ` mark_files_ro && sb_end_write Oleg Nesterov
2012-10-26 16:09 ` [PATCH 1/2] brw_mutex: big read-write mutex Mikulas Patocka
2012-10-19 17:49 ` Oleg Nesterov
2012-10-22 23:09 ` Mikulas Patocka
2012-10-23 15:12 ` Oleg Nesterov
2012-10-19 19:28 ` Paul E. McKenney
2012-10-22 23:36 ` [PATCH 0/2] fix and improvements for percpu-rw-semaphores (was: brw_mutex: big read-write mutex) Mikulas Patocka
2012-10-22 23:37 ` [PATCH 1/2] percpu-rw-semaphores: use light/heavy barriers Mikulas Patocka
2012-10-22 23:39 ` [PATCH 2/2] percpu-rw-semaphores: use rcu_read_lock_sched Mikulas Patocka
2012-10-24 16:16 ` Paul E. McKenney
2012-10-24 17:18 ` Oleg Nesterov
2012-10-24 18:20 ` Paul E. McKenney
2012-10-24 18:43 ` Oleg Nesterov
2012-10-24 19:43 ` Paul E. McKenney
2012-10-25 14:54 ` Mikulas Patocka
2012-10-25 15:07 ` Paul E. McKenney
2012-10-25 16:15 ` Mikulas Patocka
2012-10-23 16:59 ` [PATCH 1/2] percpu-rw-semaphores: use light/heavy barriers Oleg Nesterov
2012-10-23 18:05 ` Paul E. McKenney
2012-10-23 18:27 ` Oleg Nesterov
2012-10-23 18:41 ` Oleg Nesterov
2012-10-23 20:29 ` Paul E. McKenney
2012-10-23 20:32 ` Paul E. McKenney
2012-10-23 21:39 ` Mikulas Patocka
2012-10-24 16:23 ` Paul E. McKenney
2012-10-24 20:22 ` Mikulas Patocka
2012-10-24 20:36 ` Paul E. McKenney
2012-10-24 20:44 ` Mikulas Patocka
2012-10-24 23:57 ` Paul E. McKenney
2012-10-25 12:39 ` Paul E. McKenney
2012-10-25 13:48 ` Mikulas Patocka
2012-10-23 19:23 ` Oleg Nesterov
2012-10-23 20:45 ` Peter Zijlstra
2012-10-23 20:57 ` Peter Zijlstra
2012-10-24 15:11 ` Oleg Nesterov
2012-10-23 21:26 ` Mikulas Patocka
2012-10-23 20:32 ` Peter Zijlstra
2012-10-30 18:48 ` [PATCH 0/2] fix and improvements for percpu-rw-semaphores (was: brw_mutex: big read-write mutex) Oleg Nesterov
2012-10-31 19:41 ` [PATCH 0/1] percpu_rw_semaphore: reimplement to not block the readers unnecessarily Oleg Nesterov
2012-10-31 19:41 ` [PATCH 1/1] " Oleg Nesterov
2012-11-01 15:10 ` Linus Torvalds
2012-11-01 15:34 ` Oleg Nesterov
2012-11-02 18:06 ` [PATCH v2 0/1] " Oleg Nesterov
2012-11-02 18:06 ` [PATCH v2 1/1] " Oleg Nesterov
2012-11-07 17:04 ` [PATCH v3 " Mikulas Patocka
2012-11-07 17:47 ` Oleg Nesterov
2012-11-07 19:17 ` Mikulas Patocka
2012-11-08 13:42 ` Oleg Nesterov
2012-11-08 1:23 ` Paul E. McKenney
2012-11-08 1:16 ` [PATCH v2 " Paul E. McKenney
2012-11-08 13:33 ` Oleg Nesterov
2012-11-08 16:27 ` Paul E. McKenney
2012-11-08 13:48 ` [PATCH RESEND v2 0/1] " Oleg Nesterov
2012-11-08 13:48 ` [PATCH RESEND v2 1/1] " Oleg Nesterov
2012-11-08 20:07 ` Andrew Morton
2012-11-08 21:08 ` Paul E. McKenney
2012-11-08 23:41 ` Mikulas Patocka
2012-11-09 0:41 ` Paul E. McKenney
2012-11-09 3:23 ` Paul E. McKenney [this message]
2012-11-09 16:35 ` Oleg Nesterov
2012-11-09 16:59 ` Paul E. McKenney
2012-11-09 12:47 ` Mikulas Patocka
2012-11-09 15:46 ` Oleg Nesterov
2012-11-09 17:01 ` Paul E. McKenney
2012-11-09 18:10 ` Oleg Nesterov
2012-11-09 18:19 ` Oleg Nesterov
2012-11-10 0:55 ` Paul E. McKenney
2012-11-11 15:45 ` Oleg Nesterov
2012-11-12 18:38 ` Paul E. McKenney
2012-11-11 18:27 ` [PATCH -mm] percpu_rw_semaphore-reimplement-to-not-block-the-readers-unnecessari ly.fix Oleg Nesterov
2012-11-12 18:31 ` Paul E. McKenney
2012-11-16 23:22 ` Andrew Morton
2012-11-18 19:32 ` Oleg Nesterov
2012-11-01 15:43 ` [PATCH 1/1] percpu_rw_semaphore: reimplement to not block the readers unnecessarily Paul E. McKenney
2012-11-01 18:33 ` Oleg Nesterov
2012-11-02 16:18 ` Oleg Nesterov
2012-10-15 19:10 ` [PATCH 2/2] uprobes: Use brw_mutex to fix register/unregister vs dup_mmap() race Oleg Nesterov
2012-10-18 7:03 ` Srikar Dronamraju
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=20121109032310.GA2438@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=ananth@in.ibm.com \
--cc=anton@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=mpatocka@redhat.com \
--cc=oleg@redhat.com \
--cc=peterz@infradead.org \
--cc=srikar@linux.vnet.ibm.com \
--cc=torvalds@linux-foundation.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).