From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: linux-kernel@vger.kernel.org, mingo@elte.hu,
laijs@cn.fujitsu.com, dipankar@in.ibm.com,
akpm@linux-foundation.org, mathieu.desnoyers@efficios.com,
josh@joshtriplett.org, niv@us.ibm.com, tglx@linutronix.de,
peterz@infradead.org, Valdis.Kletnieks@vt.edu,
dhowells@redhat.com, edumazet@google.com, darren@dvhart.com,
fweisbec@gmail.com, sbw@mit.edu, torvalds@linux-foundation.org
Subject: Re: [PATCH RFC ticketlock] Auto-queued ticketlock
Date: Mon, 10 Jun 2013 14:30:24 -0700 [thread overview]
Message-ID: <20130610213023.GS5146@linux.vnet.ibm.com> (raw)
In-Reply-To: <1370898505.9844.123.camel@gandalf.local.home>
On Mon, Jun 10, 2013 at 05:08:25PM -0400, Steven Rostedt wrote:
> On Sun, 2013-06-09 at 12:36 -0700, Paul E. McKenney wrote:
> >
> > +#else /* #ifndef CONFIG_TICKET_LOCK_QUEUED */
> > +
> > +bool tkt_spin_pass(arch_spinlock_t *ap, struct __raw_tickets inc);
> > +
> > +static __always_inline void __ticket_spin_lock(arch_spinlock_t *lock)
> > +{
> > + register struct __raw_tickets inc = { .tail = 2 };
> > +
> > + inc = xadd(&lock->tickets, inc);
> > + for (;;) {
> > + if (inc.head == inc.tail || tkt_spin_pass(lock, inc))
> > + break;
> > + inc.head = ACCESS_ONCE(lock->tickets.head);
> > + }
> > + barrier(); /* smp_mb() on Power or ARM. */
> > +}
> > +
> > +#endif /* #else #ifndef CONFIG_TICKET_LOCK_QUEUED */
> > +
>
> To avoid the above code duplication, I would have this instead:
Nice! I have updated accordingly.
Thanx, Paul
> #ifdef CONFIG_TICKET_LOCK_QUEUED
>
> bool tkt_spin_pass(arch_spinlock_t *ap, struct __raw_tickets inc);
> #define __TKT_SPIN_INC 2
>
> #else
>
> static inline bool tkt_spin_pass(arch_spinlock_t *ap, struct
> __raw_tickets inc)
> {
> return false;
> }
>
> #define __TKT_SPIN_INC 1
>
> #endif
>
> static __always_inline void __ticket_spin_lock(arch_spinlock_t *lock)
> {
> register struct __raw_tickets inc = { .tail = __TKT_SPIN_INC };
>
> inc = xadd(&lock->tickets, inc);
>
> for (;;) {
> if (inc.head == inc.tail || tkt_spin_pass(lock, inc))
> break;
> cpu_relax();
> inc.head = ACCESS_ONCE(lock->tickets.head);
> }
> barrier; /* make sure nothing creeps before the lock is taken */
> }
>
> -- Steve
>
>
next prev parent reply other threads:[~2013-06-10 21:30 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-06-09 19:36 [PATCH RFC ticketlock] Auto-queued ticketlock Paul E. McKenney
2013-06-10 20:47 ` Steven Rostedt
2013-06-10 20:57 ` Paul E. McKenney
2013-06-10 21:01 ` Thomas Gleixner
2013-06-10 21:15 ` Paul E. McKenney
2013-06-10 21:08 ` Steven Rostedt
2013-06-10 21:30 ` Paul E. McKenney [this message]
2013-06-10 21:35 ` Eric Dumazet
2013-06-10 21:54 ` Paul E. McKenney
2013-06-10 23:02 ` Steven Rostedt
2013-06-11 0:22 ` Paul E. McKenney
2013-06-11 0:44 ` Steven Rostedt
2013-06-11 0:51 ` Linus Torvalds
2013-06-11 7:53 ` Lai Jiangshan
2013-06-11 10:14 ` Paul E. McKenney
2013-06-11 15:22 ` Steven Rostedt
2013-06-11 16:45 ` Paul E. McKenney
2013-06-11 10:06 ` Paul E. McKenney
2013-06-11 17:53 ` Davidlohr Bueso
2013-06-11 18:05 ` Paul E. McKenney
2013-06-11 18:10 ` Steven Rostedt
2013-06-11 18:14 ` Davidlohr Bueso
2013-06-11 18:46 ` Paul E. McKenney
2013-06-12 17:50 ` Davidlohr Bueso
2013-06-12 18:15 ` Linus Torvalds
2013-06-12 20:03 ` Davidlohr Bueso
2013-06-12 20:26 ` Linus Torvalds
2013-06-12 20:40 ` Davidlohr Bueso
2013-06-12 21:06 ` Raymond Jennings
2013-06-12 23:32 ` Al Viro
2013-06-13 0:01 ` Linus Torvalds
2013-06-13 0:20 ` Al Viro
2013-06-13 0:38 ` Linus Torvalds
2013-06-13 0:49 ` Al Viro
2013-06-13 0:59 ` Linus Torvalds
2013-06-14 15:00 ` Waiman Long
2013-06-14 15:37 ` Linus Torvalds
2013-06-14 18:17 ` Waiman Long
2013-06-15 1:26 ` Benjamin Herrenschmidt
2013-06-15 3:36 ` Waiman Long
2013-06-12 20:37 ` Linus Torvalds
2013-06-12 18:18 ` Steven Rostedt
2013-06-11 9:56 ` Paul E. McKenney
2013-06-11 15:00 ` Paul E. McKenney
2013-06-11 1:04 ` Steven Rostedt
2013-06-11 9:52 ` Paul E. McKenney
2013-06-11 14:48 ` Lai Jiangshan
2013-06-11 15:10 ` Lai Jiangshan
2013-06-11 16:48 ` Paul E. McKenney
2013-06-11 17:17 ` Linus Torvalds
2013-06-11 17:30 ` Paul E. McKenney
2013-06-11 16:21 ` Paul E. McKenney
2013-06-11 15:57 ` Waiman Long
2013-06-11 16:20 ` Steven Rostedt
2013-06-11 16:43 ` Paul E. McKenney
2013-06-11 17:13 ` Steven Rostedt
2013-06-11 17:43 ` Paul E. McKenney
2013-06-11 17:35 ` Waiman Long
2013-06-11 16:36 ` Paul E. McKenney
2013-06-11 17:01 ` Steven Rostedt
2013-06-11 17:16 ` Paul E. McKenney
2013-06-11 18:41 ` Waiman Long
2013-06-11 18:54 ` Davidlohr Bueso
2013-06-11 19:49 ` Paul E. McKenney
2013-06-11 20:09 ` Steven Rostedt
2013-06-11 20:32 ` Paul E. McKenney
2013-06-11 20:53 ` Steven Rostedt
2013-06-11 20:25 ` Jason Low
2013-06-11 20:36 ` Paul E. McKenney
2013-06-11 20:56 ` Steven Rostedt
2013-06-11 21:09 ` Paul E. McKenney
2013-06-12 1:19 ` Lai Jiangshan
2013-06-12 1:58 ` Steven Rostedt
2013-06-12 10:12 ` Paul E. McKenney
2013-06-12 11:06 ` Lai Jiangshan
2013-06-12 14:21 ` Paul E. McKenney
2013-06-12 14:15 ` Lai Jiangshan
2013-06-12 14:44 ` Paul E. McKenney
2013-06-11 17:02 ` [PATCH RFC ticketlock] v2 " Paul E. McKenney
2013-06-11 17:35 ` Linus Torvalds
2013-06-11 17:49 ` Paul E. McKenney
2013-06-11 17:36 ` Steven Rostedt
2013-06-11 17:52 ` Paul E. McKenney
2013-06-12 15:40 ` [PATCH RFC ticketlock] v3 " Paul E. McKenney
2013-06-12 16:13 ` Lai Jiangshan
2013-06-12 16:59 ` Paul E. McKenney
2013-06-13 2:55 ` Lai Jiangshan
2013-06-13 15:22 ` Paul E. McKenney
2013-06-13 23:25 ` Lai Jiangshan
2013-06-13 23:57 ` Paul E. McKenney
2013-06-14 1:28 ` Lai Jiangshan
2013-06-14 23:49 ` Paul E. McKenney
2013-06-14 7:12 ` Lai Jiangshan
2013-06-14 23:46 ` Paul E. McKenney
[not found] ` <CAC4Lta3dpTDc19rXLVQkZrxbu8AJL+Foc6ocAktUAozCpk2-Mg@mail.gmail.com>
2013-07-01 9:19 ` Raghavendra KT
2013-07-02 5:56 ` 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=20130610213023.GS5146@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=Valdis.Kletnieks@vt.edu \
--cc=akpm@linux-foundation.org \
--cc=darren@dvhart.com \
--cc=dhowells@redhat.com \
--cc=dipankar@in.ibm.com \
--cc=edumazet@google.com \
--cc=fweisbec@gmail.com \
--cc=josh@joshtriplett.org \
--cc=laijs@cn.fujitsu.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mingo@elte.hu \
--cc=niv@us.ibm.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=sbw@mit.edu \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.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).