From: Josh Triplett <josh@joshtriplett.org>
To: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: linux-kernel@vger.kernel.org, mingo@elte.hu,
laijs@cn.fujitsu.com, dipankar@in.ibm.com,
akpm@linux-foundation.org, mathieu.desnoyers@polymtl.ca,
niv@us.ibm.com, tglx@linutronix.de, peterz@infradead.org,
rostedt@goodmis.org, Valdis.Kletnieks@vt.edu,
dhowells@redhat.com, eric.dumazet@gmail.com, darren@dvhart.com,
fweisbec@gmail.com, patches@linaro.org,
"Paul E. McKenney" <paul.mckenney@linaro.org>
Subject: Re: [PATCH tip/core/rcu 09/15] rcu: Increasing rcu_barrier() concurrency
Date: Fri, 15 Jun 2012 16:31:51 -0700 [thread overview]
Message-ID: <20120615233151.GA7613@leaf> (raw)
In-Reply-To: <1339794370-28119-9-git-send-email-paulmck@linux.vnet.ibm.com>
On Fri, Jun 15, 2012 at 02:06:04PM -0700, Paul E. McKenney wrote:
> From: "Paul E. McKenney" <paul.mckenney@linaro.org>
>
> The traditional rcu_barrier() implementation has serialized all requests,
> regardless of RCU flavor, and also does not coalesce concurrent requests.
> In the past, this has been good and sufficient.
>
> However, systems are getting larger and use of rcu_barrier() has been
> increasing. This commit therefore introduces a counter-based scheme
> that allows _rcu_barrier() calls for the same flavor of RCU to take
> advantage of each others' work.
>
> Signed-off-by: Paul E. McKenney <paul.mckenney@linaro.org>
> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> ---
> kernel/rcutree.c | 27 ++++++++++++++++++++++++++-
> kernel/rcutree.h | 2 ++
> 2 files changed, 28 insertions(+), 1 deletions(-)
>
> diff --git a/kernel/rcutree.c b/kernel/rcutree.c
> index 93358d4..7c299d3 100644
> --- a/kernel/rcutree.c
> +++ b/kernel/rcutree.c
> @@ -2291,13 +2291,32 @@ static void _rcu_barrier(struct rcu_state *rsp)
> unsigned long flags;
> struct rcu_data *rdp;
> struct rcu_data rd;
> + unsigned long snap = ACCESS_ONCE(rsp->n_barrier_done);
> + unsigned long snap_done;
>
> init_rcu_head_on_stack(&rd.barrier_head);
>
> /* Take mutex to serialize concurrent rcu_barrier() requests. */
> mutex_lock(&rsp->barrier_mutex);
>
> - smp_mb(); /* Prevent any prior operations from leaking in. */
> + /*
> + * Ensure tht all prior references, including to ->n_barrier_done,
> + * are ordered before the _rcu_barrier() machinery.
> + */
> + smp_mb(); /* See above block comment. */
If checkpatch complains about the lack of a comment to the right of a
barrier even when the barrier has a comment directly above it, that
seems like a bug in checkpatch that needs fixing, to prevent developers
from having to add noise like "See above block comment.". :)
Also: what type of barriers do mutex_lock and mutex_unlock imply? I
assume they imply some weaker barrier than smp_mb, but I'd still assume
they imply *some* barrier.
> + /* Recheck ->n_barrier_done to see if others did our work for us. */
> + snap_done = ACCESS_ONCE(rsp->n_barrier_done);
> + if (ULONG_CMP_GE(snap_done, ((snap + 1) & ~0x1) + 2)) {
This calculation seems sufficiently clever that it merits an explanatory
comment.
> + smp_mb();
> + mutex_unlock(&rsp->barrier_mutex);
> + return;
> + }
> +
> + /* Increment ->n_barrier_done to avoid duplicate work. */
> + ACCESS_ONCE(rsp->n_barrier_done)++;
Interesting dissonance here: the use of ACCESS_ONCE with ++ implies
exactly two accesses, rather than exactly one. What makes it safe to
not use atomic_inc here, but not safe to drop the ACCESS_ONCE?
Potential use of a cached value read earlier in the function?
- Josh Triplett
next prev parent reply other threads:[~2012-06-15 23:32 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-15 21:05 [PATCH tip/core/rcu 0/15] Improvements to rcu_barrier() and RT response on big systems Paul E. McKenney
2012-06-15 21:05 ` [PATCH tip/core/rcu 01/15] rcu: Control RCU_FANOUT_LEAF from boot-time parameter Paul E. McKenney
2012-06-15 21:05 ` [PATCH tip/core/rcu 02/15] rcu: Size rcu_node tree from nr_cpu_ids rather than NR_CPUS Paul E. McKenney
2012-06-15 21:47 ` Josh Triplett
2012-06-16 0:37 ` Paul E. McKenney
2012-06-16 5:17 ` Josh Triplett
2012-06-16 6:38 ` Paul E. McKenney
2012-06-16 9:17 ` Josh Triplett
2012-06-16 14:44 ` Paul E. McKenney
2012-06-16 14:51 ` Paul E. McKenney
2012-06-16 20:31 ` Josh Triplett
2012-06-15 21:05 ` [PATCH tip/core/rcu 03/15] rcu: Prevent excessive line length in RCU_STATE_INITIALIZER() Paul E. McKenney
2012-06-15 21:48 ` Josh Triplett
2012-06-15 21:05 ` [PATCH tip/core/rcu 04/15] rcu: Place pointer to call_rcu() in rcu_data structure Paul E. McKenney
2012-06-15 22:08 ` Josh Triplett
2012-06-15 21:06 ` [PATCH tip/core/rcu 05/15] rcu: Move _rcu_barrier()'s rcu_head structures to rcu_data structures Paul E. McKenney
2012-06-15 22:19 ` Josh Triplett
2012-06-15 21:06 ` [PATCH tip/core/rcu 06/15] rcu: Move rcu_barrier_cpu_count to rcu_state structure Paul E. McKenney
2012-06-15 22:44 ` Josh Triplett
2012-06-15 21:06 ` [PATCH tip/core/rcu 07/15] rcu: Move rcu_barrier_completion " Paul E. McKenney
2012-06-15 22:51 ` Josh Triplett
2012-06-15 21:06 ` [PATCH tip/core/rcu 08/15] rcu: Move rcu_barrier_mutex " Paul E. McKenney
2012-06-15 22:55 ` Josh Triplett
2012-06-15 21:06 ` [PATCH tip/core/rcu 09/15] rcu: Increasing rcu_barrier() concurrency Paul E. McKenney
2012-06-15 23:31 ` Josh Triplett [this message]
2012-06-16 0:21 ` Steven Rostedt
2012-06-16 0:49 ` Paul E. McKenney
2012-06-16 0:48 ` Paul E. McKenney
2012-06-15 21:06 ` [PATCH tip/core/rcu 10/15] rcu: Add tracing for _rcu_barrier() Paul E. McKenney
2012-06-15 23:35 ` Josh Triplett
2012-06-15 21:06 ` [PATCH tip/core/rcu 11/15] rcu: Add rcu_barrier() statistics to debugfs tracing Paul E. McKenney
2012-06-15 23:38 ` Josh Triplett
2012-06-15 21:06 ` [PATCH tip/core/rcu 12/15] rcu: Remove unneeded __rcu_process_callbacks() argument Paul E. McKenney
2012-06-15 23:37 ` Josh Triplett
2012-06-15 21:06 ` [PATCH tip/core/rcu 13/15] rcu: Introduce for_each_rcu_flavor() and use it Paul E. McKenney
2012-06-15 23:52 ` Josh Triplett
2012-06-16 1:01 ` Paul E. McKenney
2012-06-16 5:35 ` Josh Triplett
2012-06-16 6:36 ` Paul E. McKenney
2012-06-15 21:06 ` [PATCH tip/core/rcu 14/15] rcu: Use for_each_rcu_flavor() in TREE_RCU tracing Paul E. McKenney
2012-06-15 23:59 ` Josh Triplett
2012-06-16 0:56 ` Paul E. McKenney
2012-06-16 5:22 ` Josh Triplett
2012-06-16 6:42 ` Paul E. McKenney
2012-06-15 21:06 ` [PATCH tip/core/rcu 15/15] rcu: RCU_SAVE_DYNTICK code no longer ever dead Paul E. McKenney
2012-06-16 0:02 ` Josh Triplett
2012-06-16 0:04 ` Josh Triplett
2012-06-16 1:04 ` Paul E. McKenney
2012-06-15 21:43 ` [PATCH tip/core/rcu 01/15] rcu: Control RCU_FANOUT_LEAF from boot-time parameter Josh Triplett
2012-06-15 22:10 ` 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=20120615233151.GA7613@leaf \
--to=josh@joshtriplett.org \
--cc=Valdis.Kletnieks@vt.edu \
--cc=akpm@linux-foundation.org \
--cc=darren@dvhart.com \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=eric.dumazet@gmail.com \
--cc=fweisbec@gmail.com \
--cc=laijs@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@polymtl.ca \
--cc=mingo@elte.hu \
--cc=niv@us.ibm.com \
--cc=patches@linaro.org \
--cc=paul.mckenney@linaro.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.