From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Torsten Duwe <duwe@lst.de>
Cc: Scott Wood <scottwood@freescale.com>,
Peter Zijlstra <peterz@infradead.org>,
linux-kernel@vger.kernel.org, Paul Mackerras <paulus@samba.org>,
Anton Blanchard <anton@samba.org>, Tom Musta <tommusta@gmail.com>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
linuxppc-dev@lists.ozlabs.org, Ingo Molnar <mingo@kernel.org>
Subject: Re: [PATCH] Convert powerpc simple spinlocks into ticket locks
Date: Mon, 10 Feb 2014 14:02:31 +1100 [thread overview]
Message-ID: <1392001351.3996.16.camel@pasglop> (raw)
In-Reply-To: <20140207090248.GB26811@lst.de>
On Fri, 2014-02-07 at 10:02 +0100, Torsten Duwe wrote:
> > > > Can you pair lwarx with sthcx ? I couldn't immediately find the answer
> > > > in the PowerISA doc. If so I think you can do better by being able to
> > > > atomically load both tickets but only storing the head without affecting
> > > > the tail.
>
> Can I simply write the half word, without a reservation, or will the HW caches
> mess up the other half? Will it ruin the cache coherency on some (sub)architectures?
Yes, you can, I *think*
> > Plus, sthcx doesn't exist on all PPC chips.
>
> Which ones are lacking it? Do all have at least a simple 16-bit store?
half word atomics (and byte atomics) are new, they've been added in architecture
2.06 I believe so it's fairly recent, but it's still worthwhile to investigate a
way to avoid atomics on unlock on recent processors (we can use instruction patching
if necessary based on CPU features) because there's definitely a significant cost
in doing a larx/stcx. sequence on powerpc, way higher than our current unlock path
of barrier + store.
Cheers,
Ben.
WARNING: multiple messages have this Message-ID (diff)
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Torsten Duwe <duwe@lst.de>
Cc: Tom Musta <tommusta@gmail.com>,
Peter Zijlstra <peterz@infradead.org>,
linux-kernel@vger.kernel.org, Paul Mackerras <paulus@samba.org>,
Anton Blanchard <anton@samba.org>,
Scott Wood <scottwood@freescale.com>,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
linuxppc-dev@lists.ozlabs.org, Ingo Molnar <mingo@kernel.org>
Subject: Re: [PATCH] Convert powerpc simple spinlocks into ticket locks
Date: Mon, 10 Feb 2014 14:02:31 +1100 [thread overview]
Message-ID: <1392001351.3996.16.camel@pasglop> (raw)
In-Reply-To: <20140207090248.GB26811@lst.de>
On Fri, 2014-02-07 at 10:02 +0100, Torsten Duwe wrote:
> > > > Can you pair lwarx with sthcx ? I couldn't immediately find the answer
> > > > in the PowerISA doc. If so I think you can do better by being able to
> > > > atomically load both tickets but only storing the head without affecting
> > > > the tail.
>
> Can I simply write the half word, without a reservation, or will the HW caches
> mess up the other half? Will it ruin the cache coherency on some (sub)architectures?
Yes, you can, I *think*
> > Plus, sthcx doesn't exist on all PPC chips.
>
> Which ones are lacking it? Do all have at least a simple 16-bit store?
half word atomics (and byte atomics) are new, they've been added in architecture
2.06 I believe so it's fairly recent, but it's still worthwhile to investigate a
way to avoid atomics on unlock on recent processors (we can use instruction patching
if necessary based on CPU features) because there's definitely a significant cost
in doing a larx/stcx. sequence on powerpc, way higher than our current unlock path
of barrier + store.
Cheers,
Ben.
next prev parent reply other threads:[~2014-02-10 3:03 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-06 10:37 [PATCH] Convert powerpc simple spinlocks into ticket locks Torsten Duwe
2014-02-06 10:37 ` Torsten Duwe
2014-02-06 15:53 ` Benjamin Herrenschmidt
2014-02-06 15:53 ` Benjamin Herrenschmidt
2014-02-06 16:38 ` Peter Zijlstra
2014-02-06 16:38 ` Peter Zijlstra
2014-02-06 17:37 ` Torsten Duwe
2014-02-06 17:37 ` Torsten Duwe
2014-02-06 18:08 ` Peter Zijlstra
2014-02-06 18:08 ` Peter Zijlstra
2014-02-06 19:28 ` Tom Musta
2014-02-10 2:54 ` Benjamin Herrenschmidt
2014-02-10 2:54 ` Benjamin Herrenschmidt
2014-02-07 8:24 ` Torsten Duwe
2014-02-07 8:24 ` Torsten Duwe
2014-02-06 20:19 ` Scott Wood
2014-02-06 20:19 ` Scott Wood
2014-02-07 9:02 ` Torsten Duwe
2014-02-07 9:02 ` Torsten Duwe
2014-02-07 10:31 ` Peter Zijlstra
2014-02-07 10:31 ` Peter Zijlstra
2014-02-07 10:36 ` Peter Zijlstra
2014-02-07 10:36 ` Peter Zijlstra
2014-02-07 10:45 ` Peter Zijlstra
2014-02-07 10:45 ` Peter Zijlstra
2014-02-07 11:49 ` Torsten Duwe
2014-02-07 11:49 ` Torsten Duwe
2014-02-07 12:28 ` Peter Zijlstra
2014-02-07 12:28 ` Peter Zijlstra
2014-02-07 15:18 ` Peter Zijlstra
2014-02-07 15:18 ` Peter Zijlstra
2014-02-07 15:43 ` Peter Zijlstra
2014-02-07 15:43 ` Peter Zijlstra
2014-02-07 17:08 ` Torsten Duwe
2014-02-07 17:08 ` Torsten Duwe
2014-02-07 17:19 ` Peter Zijlstra
2014-02-07 17:19 ` Peter Zijlstra
2014-02-07 15:51 ` Kumar Gala
2014-02-07 15:51 ` Kumar Gala
2014-02-07 16:10 ` Peter Zijlstra
2014-02-07 16:10 ` Peter Zijlstra
2014-02-10 3:05 ` Benjamin Herrenschmidt
2014-02-10 3:05 ` Benjamin Herrenschmidt
2014-02-10 3:02 ` Benjamin Herrenschmidt [this message]
2014-02-10 3:02 ` Benjamin Herrenschmidt
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=1392001351.3996.16.camel@pasglop \
--to=benh@kernel.crashing.org \
--cc=anton@samba.org \
--cc=duwe@lst.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mingo@kernel.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=paulus@samba.org \
--cc=peterz@infradead.org \
--cc=scottwood@freescale.com \
--cc=tommusta@gmail.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.