From: Lai Jiangshan <laijs@cn.fujitsu.com>
To: <paulmck@linux.vnet.ibm.com>
Cc: "Andy Lutomirski" <luto@amacapital.net>,
"Borislav Petkov" <bp@alien8.de>, "X86 ML" <x86@kernel.org>,
"Linus Torvalds" <torvalds@linux-foundation.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"Peter Zijlstra" <peterz@infradead.org>,
"Oleg Nesterov" <oleg@redhat.com>,
"Tony Luck" <tony.luck@intel.com>,
"Andi Kleen" <andi@firstfloor.org>,
"Josh Triplett" <josh@joshtriplett.org>,
"Frédéric Weisbecker" <fweisbec@gmail.com>
Subject: Re: [PATCH v4 2/5] x86, traps: Track entry into and exit from IST context
Date: Thu, 27 Nov 2014 15:03:22 +0800 [thread overview]
Message-ID: <5476CCBA.4000003@cn.fujitsu.com> (raw)
In-Reply-To: <20141124235058.GZ5050@linux.vnet.ibm.com>
>
> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
>
Reviewed-by: Lai Jiangshan <laijs@cn.fujitsu.com>
> diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
> index 8749f43f3f05..fc0236992655 100644
> --- a/kernel/rcu/tree.c
> +++ b/kernel/rcu/tree.c
> @@ -759,39 +759,71 @@ void rcu_irq_enter(void)
> /**
> * rcu_nmi_enter - inform RCU of entry to NMI context
> *
> - * If the CPU was idle with dynamic ticks active, and there is no
> - * irq handler running, this updates rdtp->dynticks_nmi to let the
> - * RCU grace-period handling know that the CPU is active.
> + * If the CPU was idle from RCU's viewpoint, update rdtp->dynticks and
> + * rdtp->dynticks_nmi_nesting to let the RCU grace-period handling know
> + * that the CPU is active. This implementation permits nested NMIs, as
> + * long as the nesting level does not overflow an int. (You will probably
> + * run out of stack space first.)
> */
> void rcu_nmi_enter(void)
> {
> struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);
> + int incby = 2;
>
> - if (rdtp->dynticks_nmi_nesting == 0 &&
> - (atomic_read(&rdtp->dynticks) & 0x1))
> - return;
> - rdtp->dynticks_nmi_nesting++;
> - smp_mb__before_atomic(); /* Force delay from prior write. */
> - atomic_inc(&rdtp->dynticks);
> - /* CPUs seeing atomic_inc() must see later RCU read-side crit sects */
> - smp_mb__after_atomic(); /* See above. */
> - WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1));
> + /* Complain about underflow. */
> + WARN_ON_ONCE(rdtp->dynticks_nmi_nesting < 0);
> +
> + /*
> + * If idle from RCU viewpoint, atomically increment ->dynticks
> + * to mark non-idle and increment ->dynticks_nmi_nesting by one.
> + * Otherwise, increment ->dynticks_nmi_nesting by two. This means
> + * if ->dynticks_nmi_nesting is equal to one, we are guaranteed
> + * to be in the outermost NMI handler that interrupted an RCU-idle
> + * period (observation due to Andy Lutomirski).
> + */
> + if (!(atomic_read(&rdtp->dynticks) & 0x1)) {
> + smp_mb__before_atomic(); /* Force delay from prior write. */
> + atomic_inc(&rdtp->dynticks);
> + /* atomic_inc() before later RCU read-side crit sects */
> + smp_mb__after_atomic(); /* See above. */
> + WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1));
> + incby = 1;
> + }
> + rdtp->dynticks_nmi_nesting += incby;
I prefer a "else" branch here.
if (!(atomic_read(&rdtp->dynticks) & 0x1)) {
...
WARN_ON_ONCE(rdtp->dynticks_nmi_nesting); /* paired with "rdtp->dynticks_nmi_nesting = 0" */
rdtp->dynticks_nmi_nesting = 1; /* paired with "if (rdtp->dynticks_nmi_nesting != 1) {" */
} else {
rdtp->dynticks_nmi_nesting += 2;
}
> + barrier();
> }
>
> /**
> * rcu_nmi_exit - inform RCU of exit from NMI context
> *
> - * If the CPU was idle with dynamic ticks active, and there is no
> - * irq handler running, this updates rdtp->dynticks_nmi to let the
> - * RCU grace-period handling know that the CPU is no longer active.
> + * If we are returning from the outermost NMI handler that interrupted an
> + * RCU-idle period, update rdtp->dynticks and rdtp->dynticks_nmi_nesting
> + * to let the RCU grace-period handling know that the CPU is back to
> + * being RCU-idle.
> */
> void rcu_nmi_exit(void)
> {
> struct rcu_dynticks *rdtp = this_cpu_ptr(&rcu_dynticks);
>
> - if (rdtp->dynticks_nmi_nesting == 0 ||
> - --rdtp->dynticks_nmi_nesting != 0)
> + /*
> + * Check for ->dynticks_nmi_nesting underflow and bad ->dynticks.
> + * (We are exiting an NMI handler, so RCU better be paying attention
> + * to us!)
> + */
> + WARN_ON_ONCE(rdtp->dynticks_nmi_nesting <= 0);
> + WARN_ON_ONCE(!(atomic_read(&rdtp->dynticks) & 0x1));
> +
> + /*
> + * If the nesting level is not 1, the CPU wasn't RCU-idle, so
> + * leave it in non-RCU-idle state.
> + */
> + if (rdtp->dynticks_nmi_nesting != 1) {
> + rdtp->dynticks_nmi_nesting -= 2;
> return;
> + }
> +
> + /* This NMI interrupted an RCU-idle CPU, restore RCU-idleness. */
> + rdtp->dynticks_nmi_nesting = 0;
> /* CPUs seeing atomic_inc() must see prior RCU read-side crit sects */
> smp_mb__before_atomic(); /* See above. */
> atomic_inc(&rdtp->dynticks);
>
> --
> 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:[~2014-11-27 6:59 UTC|newest]
Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-21 21:26 [PATCH v4 0/5] x86: Rework IST interrupts Andy Lutomirski
2014-11-21 21:26 ` [PATCH v4 1/5] uprobes, x86: Fix _TIF_UPROBE vs _TIF_NOTIFY_RESUME Andy Lutomirski
2014-11-22 16:55 ` Borislav Petkov
2014-11-24 17:58 ` Andy Lutomirski
2014-11-21 21:26 ` [PATCH v4 2/5] x86, traps: Track entry into and exit from IST context Andy Lutomirski
2014-11-21 21:32 ` Andy Lutomirski
2014-11-21 22:07 ` Paul E. McKenney
2014-11-21 22:19 ` Andy Lutomirski
2014-11-21 22:55 ` Paul E. McKenney
2014-11-21 23:06 ` Andy Lutomirski
2014-11-21 23:38 ` Paul E. McKenney
2014-11-22 2:00 ` Andy Lutomirski
2014-11-22 4:20 ` Paul E. McKenney
2014-11-22 5:53 ` Andy Lutomirski
2014-11-22 23:41 ` Paul E. McKenney
2014-11-24 20:22 ` Andy Lutomirski
2014-11-24 20:54 ` Paul E. McKenney
2014-11-24 21:02 ` Andy Lutomirski
2014-11-24 21:35 ` Paul E. McKenney
2014-11-24 22:34 ` Paul E. McKenney
2014-11-24 22:36 ` Andy Lutomirski
2014-11-24 22:57 ` Paul E. McKenney
2014-11-24 23:31 ` Paul E. McKenney
2014-11-24 23:35 ` Andy Lutomirski
2014-11-24 23:50 ` Paul E. McKenney
2014-11-24 23:52 ` Andy Lutomirski
2014-11-25 18:58 ` Borislav Petkov
2014-11-25 19:16 ` Paul E. McKenney
2014-12-11 0:22 ` Tony Luck
2014-12-11 0:24 ` Andy Lutomirski
2015-01-05 21:46 ` Tony Luck
2015-01-05 21:54 ` Andy Lutomirski
2015-01-06 0:44 ` [PATCH] x86, mce: Get rid of TIF_MCE_NOTIFY and associated mce tricks Luck, Tony
2015-01-06 1:01 ` Andy Lutomirski
2015-01-06 18:00 ` Luck, Tony
2015-01-07 12:13 ` Borislav Petkov
2015-01-07 15:51 ` Andy Lutomirski
2015-01-07 15:58 ` Borislav Petkov
2015-01-07 16:12 ` Paul E. McKenney
2014-11-25 17:13 ` [PATCH v4 2/5] x86, traps: Track entry into and exit from IST context Paul E. McKenney
2014-11-27 7:03 ` Lai Jiangshan [this message]
2014-11-27 16:46 ` Paul E. McKenney
2014-11-24 21:27 ` Paul E. McKenney
2014-11-21 22:20 ` Frederic Weisbecker
2014-11-21 22:00 ` Paul E. McKenney
2014-11-22 17:20 ` Borislav Petkov
2014-11-24 19:48 ` Andy Lutomirski
2015-01-22 21:52 ` Sasha Levin
2015-01-23 17:58 ` Andy Lutomirski
2015-01-23 18:04 ` Borislav Petkov
2015-01-23 18:34 ` Andy Lutomirski
2015-01-23 20:48 ` Sasha Levin
2015-01-24 1:25 ` Andy Lutomirski
2015-01-28 16:33 ` Andy Lutomirski
2015-01-28 17:48 ` Paul E. McKenney
2015-01-28 21:02 ` Andy Lutomirski
2015-01-30 19:57 ` Sasha Levin
2015-01-31 1:28 ` Sasha Levin
2015-01-31 3:12 ` Andy Lutomirski
2015-01-31 12:50 ` Andy Lutomirski
2015-01-31 13:01 ` [PATCH] x86, traps: Fix ist_enter from userspace Andy Lutomirski
2015-01-31 15:09 ` Sasha Levin
2015-01-31 16:18 ` Paul E. McKenney
2015-02-01 2:17 ` Andy Lutomirski
2015-02-04 6:01 ` [tip:x86/asm] " tip-bot for Andy Lutomirski
2014-11-21 21:26 ` [PATCH v4 3/5] x86, entry: Switch stacks on a paranoid entry " Andy Lutomirski
2014-11-24 15:55 ` Borislav Petkov
2014-11-21 21:26 ` [PATCH v4 4/5] x86: Clean up current_stack_pointer Andy Lutomirski
2014-11-24 11:39 ` Borislav Petkov
2014-11-21 21:26 ` [PATCH v4 5/5] x86, traps: Add ist_begin_non_atomic and ist_end_non_atomic Andy Lutomirski
2014-11-24 15:54 ` Borislav Petkov
2014-11-24 19:52 ` Andy Lutomirski
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=5476CCBA.4000003@cn.fujitsu.com \
--to=laijs@cn.fujitsu.com \
--cc=andi@firstfloor.org \
--cc=bp@alien8.de \
--cc=fweisbec@gmail.com \
--cc=josh@joshtriplett.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@amacapital.net \
--cc=oleg@redhat.com \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
--cc=tony.luck@intel.com \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.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).