[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Wed, 8 Aug 2018 10:27:00 -0400
From: Steven Rostedt <rostedt@...dmis.org>
To: "Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>
Cc: Joel Fernandes <joelaf@...gle.com>,
Joel Fernandes <joel@...lfernandes.org>,
LKML <linux-kernel@...r.kernel.org>,
"Cc: Android Kernel" <kernel-team@...roid.com>,
Boqun Feng <boqun.feng@...il.com>,
Byungchul Park <byungchul.park@....com>,
Ingo Molnar <mingo@...hat.com>,
Masami Hiramatsu <mhiramat@...nel.org>,
Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
Namhyung Kim <namhyung@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
Thomas Glexiner <tglx@...utronix.de>,
Tom Zanussi <tom.zanussi@...ux.intel.com>
Subject: Re: [PATCH v12 3/3] tracing: Centralize preemptirq tracepoints and
unify their usage
On Wed, 8 Aug 2018 06:00:41 -0700
"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com> wrote:
>
> I suppose that an srcu_read_lock_nmi() and srcu_read_unlock_nmi() could
> be added, which would do atomic ops on sp->sda->srcu_lock_count. Not sure
> whether this would be fast enough to be useful, but easy to provide:
>
> int __srcu_read_lock_nmi(struct srcu_struct *sp) /* UNTESTED. */
> {
> int idx;
>
> idx = READ_ONCE(sp->srcu_idx) & 0x1;
> atomic_inc(&sp->sda->srcu_lock_count[idx]);
> smp_mb__after_atomic(); /* B */ /* Avoid leaking critical section. */
> return idx;
> }
>
> void __srcu_read_unlock_nmi(struct srcu_struct *sp, int idx)
> {
> smp_mb__before_atomic(); /* C */ /* Avoid leaking critical section. */
> atomic_inc(&sp->sda->srcu_unlock_count[idx]);
> }
>
> With appropriate adjustments to also allow Tiny RCU to also work.
>
> Note that you have to use _nmi() everywhere, not just in NMI handlers.
> In fact, the NMI handlers are the one place you -don't- need to use
> _nmi(), strangely enough.
>
> Might be worth a try -- smp_mb__{before,after}_atomic() is a no-op on
> some architectures, for example.
Note this would kill the performance that srcu gives that Joel wants.
Switching from a this_cpu_inc() to a atomic_inc() would be a huge
impact.
There's also a local_inc() if you are using per cpu pointers, that is
suppose to guarantee atomicity for single cpu operations. That's what
the ftrace ring buffer uses.
-- Steve
Powered by blists - more mailing lists