From: Jeremy Fitzhardinge <jeremy@goop.org>
To: Peter Zijlstra <peterz@infradead.org>
Cc: "H. Peter Anvin" <hpa@zytor.com>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Jan Beulich <JBeulich@novell.com>, Avi Kivity <avi@redhat.com>,
Xen-devel <xen-devel@lists.xensource.com>,
Linux Virtualization <virtualization@lists.linux-foundation.org>,
Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>,
Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Subject: Re: [PATCH 00/20] x86: ticket lock rewrite and paravirtualization
Date: Mon, 15 Nov 2010 13:02:09 -0800 [thread overview]
Message-ID: <4CE19FD1.2000804@goop.org> (raw)
In-Reply-To: <1289852088.2109.553.camel@laptop>
On 11/15/2010 12:14 PM, Peter Zijlstra wrote:
> On Mon, 2010-11-15 at 12:03 -0800, H. Peter Anvin wrote:
>> On 11/15/2010 12:00 PM, Jeremy Fitzhardinge wrote:
>>> Another approach I discussed with PeterZ and Mathieu is to steal the LSB
>>> of the ticket counters (halving the max CPU count) to use as a "there is
>>> someone in slowpath waiting on this lock". But I haven't spent the time
>>> to work out an algorithm to maintain that flag (or flags, since there
>>> are bits available) in a correct and efficient way.
>>>
>> Definitely worth pondering.
> Right, so the idea was to make the ticket increment 2, which would leave
> the LSB of both the head and tail available. I think that if one were to
> set both (using a cmpxchg), the ticket fast-path wouldn't need any
> changes since head==tail is still the correct condition for acquisition.
>
> Then the unlock needs an added conditional:
> if (tail & 1)
> unlock_slowpath()
The tricky part is knowing how to clear the bit(s) on the last person
dropping out of the slow path, and making that race-free with respect to
new lockers entering the slow path. I guess you could leave it in
slowpath state until you're the last unlocker (ie, you're unlocking into
uncontended state), whereupon you also clear the bits; I guess that
would probably need a cmpxchg to make it safe WRT new lockers entering
slowpath.
As a heuristic, it shouldn't be too bad performancewise, since
(handwaving) if ticketholder N has entered the slowpath, then its likely
that N+1 will as well.
J
next prev parent reply other threads:[~2010-11-15 21:02 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-11-03 14:59 [PATCH 00/20] x86: ticket lock rewrite and paravirtualization Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 01/20] x86/ticketlock: clean up types and accessors Jeremy Fitzhardinge
2010-11-13 9:57 ` Américo Wang
2010-11-15 19:36 ` Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 02/20] x86/ticketlock: convert spin loop to C Jeremy Fitzhardinge
2010-11-03 15:11 ` Eric Dumazet
2010-11-03 15:38 ` Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 03/20] x86/ticketlock: Use C for __ticket_spin_unlock Jeremy Fitzhardinge
2010-11-03 15:13 ` Eric Dumazet
2010-11-03 18:00 ` Jeremy Fitzhardinge
2010-11-13 10:05 ` Américo Wang
2010-11-13 22:34 ` Paolo Bonzini
2010-11-15 19:38 ` Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 04/20] x86/ticketlock: make large and small ticket versions of spin_lock the same Jeremy Fitzhardinge
2010-11-12 12:19 ` Srivatsa Vaddagiri
2010-11-12 16:27 ` Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 05/20] x86/ticketlock: make __ticket_spin_lock common Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 06/20] x86/ticketlock: make __ticket_spin_trylock common Jeremy Fitzhardinge
2010-11-13 10:17 ` Américo Wang
2010-11-13 10:48 ` Eric Dumazet
2010-11-15 19:39 ` Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 07/20] x86/spinlocks: replace pv spinlocks with pv ticketlocks Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 08/20] x86/ticketlock: collapse a layer of functions Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 09/20] xen/pvticketlock: Xen implementation for PV ticket locks Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 10/20] x86/pvticketlock: keep count of blocked cpus Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 11/20] x86/pvticketlock: use callee-save for lock_spinning Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 12/20] x86/pvticketlock: use callee-save for unlock_kick as well Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 13/20] x86/pvticketlock: make sure unlock is seen by everyone before checking waiters Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 14/20] x86/ticketlock: loosen ordering restraints on unlock Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 15/20] x86/ticketlock: prevent compiler reordering into locked region Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 16/20] x86/ticketlock: don't inline _spin_unlock when using paravirt spinlocks Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 17/20] x86/ticketlock: clarify barrier in arch_spin_lock Jeremy Fitzhardinge
2010-11-03 14:59 ` [PATCH 18/20] x86/ticketlock: remove .slock Jeremy Fitzhardinge
2010-11-03 15:00 ` [PATCH 19/20] x86/ticketlocks: use overlapping read to eliminate mb() Jeremy Fitzhardinge
2010-11-03 15:00 ` [PATCH 20/20] x86/ticketlock: rename ticketpair to head_tail Jeremy Fitzhardinge
2010-11-12 22:12 ` [PATCH 00/20] x86: ticket lock rewrite and paravirtualization H. Peter Anvin
2010-11-12 22:17 ` Jeremy Fitzhardinge
2010-11-12 22:20 ` H. Peter Anvin
2010-11-15 20:00 ` Jeremy Fitzhardinge
2010-11-15 20:03 ` H. Peter Anvin
2010-11-15 20:14 ` Peter Zijlstra
2010-11-15 21:02 ` Jeremy Fitzhardinge [this message]
2010-11-15 21:08 ` Peter Zijlstra
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=4CE19FD1.2000804@goop.org \
--to=jeremy@goop.org \
--cc=JBeulich@novell.com \
--cc=avi@redhat.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@polymtl.ca \
--cc=peterz@infradead.org \
--cc=vatsa@linux.vnet.ibm.com \
--cc=virtualization@lists.linux-foundation.org \
--cc=xen-devel@lists.xensource.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).