public inbox for linux-arm-kernel@lists.infradead.org
 help / color / mirror / Atom feed
From: Will Deacon <will@kernel.org>
To: Shaokun Zhang <zhangshaokun@hisilicon.com>
Cc: Yuqi Jin <jinyuqi@huawei.com>,
	Andrew Murray <amurray@thegoodpenguin.co.uk>,
	linux-arm-kernel@lists.infradead.org,
	Catalin Marinas <catalin.marinas@arm.com>
Subject: Re: [PATCH] arm64: atomics: Fix the issue on xchg when switch to atomic instruction
Date: Tue, 26 May 2020 20:55:37 +0100	[thread overview]
Message-ID: <20200526195536.GD2206@willie-the-truck> (raw)
In-Reply-To: <d1e62e64-9cda-eb70-42f8-f65e43632add@hisilicon.com>

On Mon, May 25, 2020 at 05:27:30PM +0800, Shaokun Zhang wrote:
> On 2020/5/7 15:54, Shaokun Zhang wrote:
> > On 2020/5/6 19:30, Shaokun Zhang wrote:
> >> On 2020/5/6 18:44, Will Deacon wrote:
> >>> Good to hear there's not a bug, but if you see a performance benefit from
> >>> using the static-key for xchg() then I'd obviously be open to changing it
> >>
> >> Thanks your reply, if I follow the two methods correctly, static-key will
> >> not consume '__nops(3)', others are the same.
> >>
> >> I will run some tests to check the performance  ;-)
> >>
> > 
> > We compare the two methods on Huawei Kunpeng920 and the throughput per second
> > as follows:
> > 
> > one core  |without delay| 200ns delay|
> > --------------------------------------
> > static-key| 55294942    | 3937156    |
> > --------------------------------------
> > runtime   | 54706282    | 3918188    |
> > --------------------------------------
> > 
> 
> Are you happy to pick up this patch since it has some benefits for single core?  ;-)

Is it really worth it? I don't think so.

Will

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

      reply	other threads:[~2020-05-26 19:55 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-05  9:02 [PATCH] arm64: atomics: Fix the issue on xchg when switch to atomic instruction Shaokun Zhang
2020-05-05  9:15 ` Will Deacon
2020-05-06  7:00   ` Shaokun Zhang
2020-05-06  7:53     ` Will Deacon
2020-05-06 10:39       ` Shaokun Zhang
2020-05-06 10:44         ` Will Deacon
2020-05-06 11:30           ` Shaokun Zhang
2020-05-07  7:54             ` Shaokun Zhang
2020-05-25  9:27               ` Shaokun Zhang
2020-05-26 19:55                 ` Will Deacon [this message]

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=20200526195536.GD2206@willie-the-truck \
    --to=will@kernel.org \
    --cc=amurray@thegoodpenguin.co.uk \
    --cc=catalin.marinas@arm.com \
    --cc=jinyuqi@huawei.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=zhangshaokun@hisilicon.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