From: Joel Fernandes <joel@joelfernandes.org>
To: Scott Wood <swood@redhat.com>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
linux-rt-users@vger.kernel.org, linux-kernel@vger.kernel.org,
"Paul E . McKenney" <paulmck@linux.ibm.com>,
Thomas Gleixner <tglx@linutronix.de>,
Steven Rostedt <rostedt@goodmis.org>,
Peter Zijlstra <peterz@infradead.org>,
Juri Lelli <juri.lelli@redhat.com>,
Clark Williams <williams@redhat.com>
Subject: Re: [PATCH RT v3 1/5] rcu: Acquire RCU lock when disabling BHs
Date: Thu, 12 Sep 2019 18:09:40 -0400 [thread overview]
Message-ID: <20190912220940.GB150506@google.com> (raw)
In-Reply-To: <20190911165729.11178-2-swood@redhat.com>
On Wed, Sep 11, 2019 at 05:57:25PM +0100, Scott Wood wrote:
> A plain local_bh_disable() is documented as creating an RCU critical
> section, and (at least) rcutorture expects this to be the case. However,
> in_softirq() doesn't block a grace period on PREEMPT_RT, since RCU checks
> preempt_count() directly. Even if RCU were changed to check
> in_softirq(), that wouldn't allow blocked BH disablers to be boosted.
>
> Fix this by calling rcu_read_lock() from local_bh_disable(), and update
> rcu_read_lock_bh_held() accordingly.
>
> Signed-off-by: Scott Wood <swood@redhat.com>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
thanks,
- Joel
> ---
> v3: Remove change to rcu_read_lock_bh_held(), and move debug portions
> of rcu_read_[un]lock_bh() to separate functions
> ---
> include/linux/rcupdate.h | 40 ++++++++++++++++++++++++++++++++--------
> kernel/softirq.c | 12 +++++++++---
> 2 files changed, 41 insertions(+), 11 deletions(-)
>
> diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
> index 388ace315f32..9ce7c5006a5e 100644
> --- a/include/linux/rcupdate.h
> +++ b/include/linux/rcupdate.h
> @@ -600,6 +600,36 @@ static inline void rcu_read_unlock(void)
> rcu_lock_release(&rcu_lock_map); /* Keep acq info for rls diags. */
> }
>
> +#ifdef CONFIG_PREEMPT_RT_FULL
> +/*
> + * On RT, local_bh_disable() calls rcu_read_lock() -- no need to
> + * track it twice.
> + */
> +static inline void rcu_bh_lock_acquire(void)
> +{
> +}
> +
> +static inline void rcu_bh_lock_release(void)
> +{
> +}
> +#else
> +static inline void rcu_bh_lock_acquire(void)
> +{
> + __acquire(RCU_BH);
> + rcu_lock_acquire(&rcu_bh_lock_map);
> + RCU_LOCKDEP_WARN(!rcu_is_watching(),
> + "rcu_read_lock_bh() used illegally while idle");
> +}
> +
> +static inline void rcu_bh_lock_release(void)
> +{
> + RCU_LOCKDEP_WARN(!rcu_is_watching(),
> + "rcu_read_unlock_bh() used illegally while idle");
> + rcu_lock_release(&rcu_bh_lock_map);
> + __release(RCU_BH);
> +}
> +#endif
> +
> /**
> * rcu_read_lock_bh() - mark the beginning of an RCU-bh critical section
> *
> @@ -615,10 +645,7 @@ static inline void rcu_read_unlock(void)
> static inline void rcu_read_lock_bh(void)
> {
> local_bh_disable();
> - __acquire(RCU_BH);
> - rcu_lock_acquire(&rcu_bh_lock_map);
> - RCU_LOCKDEP_WARN(!rcu_is_watching(),
> - "rcu_read_lock_bh() used illegally while idle");
> + rcu_bh_lock_acquire();
> }
>
> /*
> @@ -628,10 +655,7 @@ static inline void rcu_read_lock_bh(void)
> */
> static inline void rcu_read_unlock_bh(void)
> {
> - RCU_LOCKDEP_WARN(!rcu_is_watching(),
> - "rcu_read_unlock_bh() used illegally while idle");
> - rcu_lock_release(&rcu_bh_lock_map);
> - __release(RCU_BH);
> + rcu_bh_lock_release();
> local_bh_enable();
> }
>
> diff --git a/kernel/softirq.c b/kernel/softirq.c
> index d16d080a74f7..6080c9328df1 100644
> --- a/kernel/softirq.c
> +++ b/kernel/softirq.c
> @@ -115,8 +115,10 @@ void __local_bh_disable_ip(unsigned long ip, unsigned int cnt)
> long soft_cnt;
>
> WARN_ON_ONCE(in_irq());
> - if (!in_atomic())
> + if (!in_atomic()) {
> local_lock(bh_lock);
> + rcu_read_lock();
> + }
> soft_cnt = this_cpu_inc_return(softirq_counter);
> WARN_ON_ONCE(soft_cnt == 0);
> current->softirq_count += SOFTIRQ_DISABLE_OFFSET;
> @@ -151,8 +153,10 @@ void _local_bh_enable(void)
> #endif
>
> current->softirq_count -= SOFTIRQ_DISABLE_OFFSET;
> - if (!in_atomic())
> + if (!in_atomic()) {
> + rcu_read_unlock();
> local_unlock(bh_lock);
> + }
> }
>
> void _local_bh_enable_rt(void)
> @@ -185,8 +189,10 @@ void __local_bh_enable_ip(unsigned long ip, unsigned int cnt)
> WARN_ON_ONCE(count < 0);
> local_irq_enable();
>
> - if (!in_atomic())
> + if (!in_atomic()) {
> + rcu_read_unlock();
> local_unlock(bh_lock);
> + }
>
> current->softirq_count -= SOFTIRQ_DISABLE_OFFSET;
> preempt_check_resched();
> --
> 1.8.3.1
>
next prev parent reply other threads:[~2019-09-12 22:09 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-11 16:57 [PATCH RT v3 0/5] RCU fixes Scott Wood
2019-09-11 16:57 ` [PATCH RT v3 1/5] rcu: Acquire RCU lock when disabling BHs Scott Wood
2019-09-12 22:09 ` Joel Fernandes [this message]
2019-09-17 7:44 ` Sebastian Andrzej Siewior
2019-09-17 14:06 ` Scott Wood
2019-09-17 14:42 ` Sebastian Andrzej Siewior
2019-09-17 16:12 ` Scott Wood
2019-09-23 16:41 ` Sebastian Andrzej Siewior
2019-09-11 16:57 ` [PATCH RT v3 2/5] sched: Rename sleeping_lock to rt_invol_sleep Scott Wood
2019-09-17 7:52 ` Sebastian Andrzej Siewior
2019-09-11 16:57 ` [PATCH RT v3 3/5] sched: migrate_dis/enable: Use rt_invol_sleep Scott Wood
2019-09-17 7:59 ` Sebastian Andrzej Siewior
2019-09-17 14:06 ` Scott Wood
2019-09-23 16:59 ` Scott Wood
2019-09-23 17:52 ` Sebastian Andrzej Siewior
2019-09-24 11:21 ` Sebastian Andrzej Siewior
2019-09-24 13:53 ` Scott Wood
2019-09-24 15:25 ` Sebastian Andrzej Siewior
2019-09-24 15:47 ` Scott Wood
2019-09-24 16:05 ` Sebastian Andrzej Siewior
2019-09-24 16:35 ` Scott Wood
2019-10-04 16:45 ` Sebastian Andrzej Siewior
2019-09-11 16:57 ` [PATCH RT v3 4/5] rcu: Disable use_softirq on PREEMPT_RT Scott Wood
2019-09-12 21:38 ` Joel Fernandes
2019-09-12 22:19 ` Joel Fernandes
2019-09-17 9:31 ` Sebastian Andrzej Siewior
2019-09-17 14:08 ` Scott Wood
2019-09-11 16:57 ` [PATCH RT v3 5/5] rcutorture: Avoid problematic critical section nesting on RT Scott Wood
2019-09-12 22:17 ` Joel Fernandes
2019-09-16 16:55 ` Scott Wood
2019-09-17 10:07 ` Sebastian Andrzej Siewior
2019-09-17 14:36 ` Scott Wood
2019-09-17 14:50 ` Sebastian Andrzej Siewior
2019-09-17 16:32 ` Scott Wood
2019-09-23 16:25 ` Sebastian Andrzej Siewior
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=20190912220940.GB150506@google.com \
--to=joel@joelfernandes.org \
--cc=bigeasy@linutronix.de \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rt-users@vger.kernel.org \
--cc=paulmck@linux.ibm.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=swood@redhat.com \
--cc=tglx@linutronix.de \
--cc=williams@redhat.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).