From: Josh Triplett <josh@joshtriplett.org>
To: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: linux-kernel@vger.kernel.org, mingo@kernel.org,
jiangshanlai@gmail.com, dipankar@in.ibm.com,
akpm@linux-foundation.org, mathieu.desnoyers@efficios.com,
tglx@linutronix.de, peterz@infradead.org, rostedt@goodmis.org,
dhowells@redhat.com, edumazet@google.com, dvhart@linux.intel.com,
fweisbec@gmail.com, oleg@redhat.com, bobby.prani@gmail.com,
Boqun Feng <boqun.feng@gmail.com>
Subject: Re: [PATCH tip/core/rcu 04/13] rcu: Don't disable preemption for Tiny and Tree RCU readers
Date: Tue, 6 Oct 2015 10:16:30 -0700 [thread overview]
Message-ID: <20151006171630.GC9600@cloud> (raw)
In-Reply-To: <20151006170101.GG3910@linux.vnet.ibm.com>
On Tue, Oct 06, 2015 at 10:01:01AM -0700, Paul E. McKenney wrote:
> On Tue, Oct 06, 2015 at 09:44:45AM -0700, Josh Triplett wrote:
> > On Tue, Oct 06, 2015 at 09:13:39AM -0700, Paul E. McKenney wrote:
> > > From: Boqun Feng <boqun.feng@gmail.com>
> > >
> > > Because preempt_disable() maps to barrier() for non-debug builds,
> > > it forces the compiler to spill and reload registers. Because Tree
> > > RCU and Tiny RCU now only appear in CONFIG_PREEMPT=n builds, these
> > > barrier() instances generate needless extra code for each instance of
> > > rcu_read_lock() and rcu_read_unlock(). This extra code slows down Tree
> > > RCU and bloats Tiny RCU.
> > >
> > > This commit therefore removes the preempt_disable() and preempt_enable()
> > > from the non-preemptible implementations of __rcu_read_lock() and
> > > __rcu_read_unlock(), respectively. However, for debug purposes,
> > > preempt_disable() and preempt_enable() are still invoked if
> > > CONFIG_PREEMPT_COUNT=y, because this allows detection of sleeping inside
> > > atomic sections in non-preemptible kernels.
> > >
> > > This is based on an earlier patch by Paul E. McKenney, fixing
> > > a bug encountered in kernels built with CONFIG_PREEMPT=n and
> > > CONFIG_PREEMPT_COUNT=y.
> >
> > This also adds explicit barrier() calls to several internal RCU
> > functions, but the commit message doesn't explain those at all.
>
> To compensate for them being removed from rcu_read_lock() and
> rcu_read_unlock(), but yes, I will update.
That much seemed clear from the comments, but that doesn't explain *why*
those functions need barriers of their own even though rcu_read_lock()
and rcu_read_unlock() don't.
> > > Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
> > > Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> > > ---
> > > include/linux/rcupdate.h | 6 ++++--
> > > include/linux/rcutiny.h | 1 +
> > > kernel/rcu/tree.c | 9 +++++++++
> > > 3 files changed, 14 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
> > > index d63bb77dab35..6c3ceceb6148 100644
> > > --- a/include/linux/rcupdate.h
> > > +++ b/include/linux/rcupdate.h
> > > @@ -297,12 +297,14 @@ void synchronize_rcu(void);
> > >
> > > static inline void __rcu_read_lock(void)
> > > {
> > > - preempt_disable();
> > > + if (IS_ENABLED(CONFIG_PREEMPT_COUNT))
> > > + preempt_disable();
> > > }
> > >
> > > static inline void __rcu_read_unlock(void)
> > > {
> > > - preempt_enable();
> > > + if (IS_ENABLED(CONFIG_PREEMPT_COUNT))
> > > + preempt_enable();
> > > }
> > >
> > > static inline void synchronize_rcu(void)
> > > diff --git a/include/linux/rcutiny.h b/include/linux/rcutiny.h
> > > index c8a0722f77ea..4c1aaf9cce7b 100644
> > > --- a/include/linux/rcutiny.h
> > > +++ b/include/linux/rcutiny.h
> > > @@ -216,6 +216,7 @@ static inline bool rcu_is_watching(void)
> > >
> > > static inline void rcu_all_qs(void)
> > > {
> > > + barrier(); /* Avoid RCU read-side critical sections leaking across. */
> > > }
> > >
> > > #endif /* __LINUX_RCUTINY_H */
> > > diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
> > > index b9d9e0249e2f..93c0f23c3e45 100644
> > > --- a/kernel/rcu/tree.c
> > > +++ b/kernel/rcu/tree.c
> > > @@ -337,12 +337,14 @@ static void rcu_momentary_dyntick_idle(void)
> > > */
> > > void rcu_note_context_switch(void)
> > > {
> > > + barrier(); /* Avoid RCU read-side critical sections leaking down. */
> > > trace_rcu_utilization(TPS("Start context switch"));
> > > rcu_sched_qs();
> > > rcu_preempt_note_context_switch();
> > > if (unlikely(raw_cpu_read(rcu_sched_qs_mask)))
> > > rcu_momentary_dyntick_idle();
> > > trace_rcu_utilization(TPS("End context switch"));
> > > + barrier(); /* Avoid RCU read-side critical sections leaking up. */
> > > }
> > > EXPORT_SYMBOL_GPL(rcu_note_context_switch);
> > >
> > > @@ -353,12 +355,19 @@ EXPORT_SYMBOL_GPL(rcu_note_context_switch);
> > > * RCU flavors in desperate need of a quiescent state, which will normally
> > > * be none of them). Either way, do a lightweight quiescent state for
> > > * all RCU flavors.
> > > + *
> > > + * The barrier() calls are redundant in the common case when this is
> > > + * called externally, but just in case this is called from within this
> > > + * file.
> > > + *
> > > */
> > > void rcu_all_qs(void)
> > > {
> > > + barrier(); /* Avoid RCU read-side critical sections leaking down. */
> > > if (unlikely(raw_cpu_read(rcu_sched_qs_mask)))
> > > rcu_momentary_dyntick_idle();
> > > this_cpu_inc(rcu_qs_ctr);
> > > + barrier(); /* Avoid RCU read-side critical sections leaking up. */
> > > }
> > > EXPORT_SYMBOL_GPL(rcu_all_qs);
> > >
> > > --
> > > 2.5.2
> > >
> >
>
next prev parent reply other threads:[~2015-10-06 17:16 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-10-06 16:13 [PATCH tip/core/rcu 0/13] Miscellaneous fixes for 4.4 Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 01/13] sched: Export sched_setscheduler_nocheck Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 02/13] rcu: Use rcu_callback_t in call_rcu*() and friends Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 03/13] rcu: Use call_rcu_func_t to replace explicit type equivalents Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 04/13] rcu: Don't disable preemption for Tiny and Tree RCU readers Paul E. McKenney
2015-10-06 16:20 ` [Kernel networking modules.] OSI levels 2 & 3, Assistance - If anyone knows anyone in the US. North West region John D Allen, Leveridge Systems INC
2015-10-06 16:44 ` [PATCH tip/core/rcu 04/13] rcu: Don't disable preemption for Tiny and Tree RCU readers Josh Triplett
2015-10-06 17:01 ` Paul E. McKenney
2015-10-06 17:16 ` Josh Triplett [this message]
2015-10-06 17:42 ` Paul E. McKenney
2015-10-06 17:46 ` Josh Triplett
2015-10-06 20:05 ` Peter Zijlstra
2015-10-06 20:18 ` Paul E. McKenney
2015-10-06 20:52 ` Peter Zijlstra
2015-10-06 21:05 ` Paul E. McKenney
2015-10-07 7:19 ` Peter Zijlstra
2015-10-06 16:13 ` [PATCH tip/core/rcu 05/13] rcu: Eliminate panic when silly boot-time fanout specified Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 06/13] rcu: Add online/offline info to stall warning message Paul E. McKenney
2015-10-06 17:15 ` Josh Triplett
2015-10-06 16:13 ` [PATCH tip/core/rcu 07/13] rcu: Move preemption disabling out of __srcu_read_lock() Paul E. McKenney
2015-10-06 17:18 ` Josh Triplett
2015-10-06 17:36 ` Paul E. McKenney
2015-10-06 17:43 ` Josh Triplett
2015-10-06 18:07 ` Paul E. McKenney
2015-10-06 20:07 ` Peter Zijlstra
2015-10-06 20:19 ` Paul E. McKenney
2015-10-06 20:32 ` Peter Zijlstra
2015-10-06 21:03 ` Paul E. McKenney
2015-10-07 7:20 ` Peter Zijlstra
2015-10-07 14:18 ` Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 08/13] rcu: Finish folding ->fqs_state into ->gp_state Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 09/13] rcu: Correct comment for values of ->gp_state field Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 10/13] rcu: Add rcu_pointer_handoff() Paul E. McKenney
2015-10-06 17:21 ` Josh Triplett
2015-10-06 17:31 ` Paul E. McKenney
2015-10-06 17:36 ` Josh Triplett
2015-10-06 20:27 ` Peter Zijlstra
2015-10-06 21:02 ` Paul E. McKenney
2015-10-07 7:22 ` Peter Zijlstra
2015-10-07 14:20 ` Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 11/13] rculist: Make list_entry_rcu() use lockless_dereference() Paul E. McKenney
2015-10-26 8:45 ` Ingo Molnar
2015-10-26 14:55 ` Paul E. McKenney
2015-10-26 18:02 ` Ingo Molnar
2015-10-27 3:37 ` Linus Torvalds
2015-10-27 5:19 ` Tejun Heo
2015-10-27 5:33 ` Paul E. McKenney
2015-10-28 8:33 ` Ingo Molnar
2015-10-28 20:35 ` Patrick Marlier
2015-10-29 0:00 ` Paul E. McKenney
2015-10-29 2:13 ` Tejun Heo
2015-10-28 20:58 ` [tip:core/rcu] fs/writeback, rcu: Don't use list_entry_rcu() for pointer offsetting in bdi_split_work_to_wbs() tip-bot for Tejun Heo
2015-10-27 5:32 ` [PATCH tip/core/rcu 11/13] rculist: Make list_entry_rcu() use lockless_dereference() Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 12/13] rcu: Remove deprecated rcu_lockdep_assert() Paul E. McKenney
2015-10-06 16:13 ` [PATCH tip/core/rcu 13/13] rculist: Use WRITE_ONCE() when deleting from reader-visible list Paul E. McKenney
2015-10-06 17:23 ` [PATCH tip/core/rcu 0/13] Miscellaneous fixes for 4.4 Josh Triplett
2015-10-06 17:38 ` 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=20151006171630.GC9600@cloud \
--to=josh@joshtriplett.org \
--cc=akpm@linux-foundation.org \
--cc=bobby.prani@gmail.com \
--cc=boqun.feng@gmail.com \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=dvhart@linux.intel.com \
--cc=edumazet@google.com \
--cc=fweisbec@gmail.com \
--cc=jiangshanlai@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--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.