From: robherring2@gmail.com (Rob Herring)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH] ARM: implement optimized percpu variable access
Date: Mon, 12 Nov 2012 08:21:27 -0600 [thread overview]
Message-ID: <50A105E7.6000005@gmail.com> (raw)
In-Reply-To: <20121112102354.GA2346@mudshark.cambridge.arm.com>
On 11/12/2012 04:23 AM, Will Deacon wrote:
> Hi Rob,
>
> On Sun, Nov 11, 2012 at 03:20:40AM +0000, Rob Herring wrote:
>> From: Rob Herring <rob.herring@calxeda.com>
>>
>> Use the previously unused TPIDRPRW register to store percpu offsets.
>> TPIDRPRW is only accessible in PL1, so it can only be used in the kernel.
>>
>> This saves 2 loads for each percpu variable access which should yield
>> improved performance, but the improvement has not been quantified.
>
> The patch looks largely fine to me (one minor comment below), but we should
> try and see what the performance difference is like on a few cores before
> merging this. Have you tried something like hackbench to see if the
> difference is measurable there? If not, I guess we'll need something more
> targetted.
Looks like it's about a 1.4% improvement on Cortex-A9 (highbank) with
hackbench.
Average of 30 runs of "hackbench -l 1000":
Before: 6.2190666667
After: 6.1347666667
I'll add this data to the commit msg.
Rob
>
>> diff --git a/arch/arm/include/asm/percpu.h b/arch/arm/include/asm/percpu.h
>> new file mode 100644
>> index 0000000..9eb7372
>> --- /dev/null
>> +++ b/arch/arm/include/asm/percpu.h
>> @@ -0,0 +1,44 @@
>> +/*
>> + * Copyright 2012 Calxeda, Inc.
>> + *
>> + * This program is free software; you can redistribute it and/or modify it
>> + * under the terms and conditions of the GNU General Public License,
>> + * version 2, as published by the Free Software Foundation.
>> + *
>> + * This program is distributed in the hope it will be useful, but WITHOUT
>> + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
>> + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
>> + * more details.
>> + *
>> + * You should have received a copy of the GNU General Public License along with
>> + * this program. If not, see <http://www.gnu.org/licenses/>.
>> + */
>> +#ifndef _ASM_ARM_PERCPU_H_
>> +#define _ASM_ARM_PERCPU_H_
>> +
>> +/*
>> + * Same as asm-generic/percpu.h, except that we store the per cpu offset
>> + * in the TPIDRPRW.
>> + */
>> +#if defined(CONFIG_SMP) && (__LINUX_ARM_ARCH__ >= 6)
>> +
>> +static inline void set_my_cpu_offset(unsigned long off)
>> +{
>> + asm volatile("mcr p15, 0, %0, c13, c0, 4 @ set TPIDRPRW" : : "r" (off) : "cc" );
>> +}
>
> You don't need the "cc" here.
>
> Will
>
>
next prev parent reply other threads:[~2012-11-12 14:21 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-11 3:20 [PATCH] ARM: implement optimized percpu variable access Rob Herring
2012-11-12 10:23 ` Will Deacon
2012-11-12 13:03 ` Rob Herring
2012-11-12 13:28 ` Will Deacon
2012-11-12 14:03 ` Rob Herring
2012-11-27 17:29 ` Nicolas Pitre
2012-11-12 14:21 ` Rob Herring [this message]
2012-11-12 14:41 ` Will Deacon
2012-11-12 16:51 ` Will Deacon
2012-11-12 21:01 ` Rob Herring
2012-11-13 10:40 ` Will Deacon
2012-11-22 11:34 ` Will Deacon
2012-11-22 11:39 ` Russell King - ARM Linux
2012-11-23 17:06 ` Rob Herring
2012-11-23 17:12 ` Russell King - ARM Linux
2012-11-23 17:16 ` Will Deacon
2012-11-23 20:34 ` Tony Lindgren
2012-11-23 20:32 ` Tony Lindgren
2012-11-25 18:46 ` Rob Herring
2012-11-26 11:13 ` Will Deacon
2012-11-26 15:15 ` Will Deacon
2012-11-26 17:30 ` Rob Herring
2012-11-27 13:17 ` Will Deacon
2012-11-27 13:26 ` Russell King - ARM Linux
2012-11-26 21:58 ` Jamie Lokier
2012-11-26 23:50 ` Jamie Lokier
2012-11-27 1:02 ` Jamie Lokier
2012-11-27 22:02 ` Rob Herring
2012-11-28 12:34 ` Will Deacon
2012-11-27 17:35 ` Nicolas Pitre
2012-11-27 19:27 ` Nicolas Pitre
2012-11-27 17:19 ` Nicolas Pitre
2012-11-27 19:37 ` Rob Herring
2012-11-27 20:42 ` Rob Herring
2012-11-27 22:02 ` Nicolas Pitre
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=50A105E7.6000005@gmail.com \
--to=robherring2@gmail.com \
--cc=linux-arm-kernel@lists.infradead.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 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.