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 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).