linux-kernel - RE: hotplug support for arch/arc/plat-eznps platform

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <VI1PR0501MB211071DC936129CBC5C4A397B2B50@VI1PR0501MB2110.eurprd05.prod.outlook.com>
Date:   Mon, 7 Aug 2017 13:41:38 +0000
From:   "Ofer Levi(SW)" <oferle@...lanox.com>
To:     Peter Zijlstra <peterz@...radead.org>,
        "rusty@...tcorp.com.au" <rusty@...tcorp.com.au>,
        "vatsa@...ibm.com" <vatsa@...ibm.com>,
        "mingo@...hat.com" <mingo@...hat.com>
CC:     "Vineet.Gupta1@...opsys.com" <Vineet.Gupta1@...opsys.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: RE: hotplug support for arch/arc/plat-eznps platform

> You've failed to explain why you think hotplug should be a performance
> critical path.
1. hotplug bring up of 4K cpus takes 40 minutes.  Way too much for any user.
2. plat-eznps is a network processor, where bring up time is sensitive.

> I'm also not seeing how it would be different from boot; you'd be
> looking at a similar cost for SMP bringup.
bring up time of 4k cpus during kernel boot takes 4.5 minutes. 
The function in question is performed only when smp init was performed. 
If I understand correctly, whatever this function is doing is performed after all cpus 
were brought up during kernel boot.


Thanks
-Ofer


 On Monday, August 7, 2017 11:34 AM +0000, Peter Zijlstra wrote:
> On Sun, Aug 06, 2017 at 05:53:37AM +0000, Ofer Levi(SW) wrote:
> > I'm adding hot-plug support for the above arch and ran into performance
> issue with execution of
> > partition_sched_domains ()  - About 0.5 sec per cpu, which is unacceptable
> with the arch supported 4k cpus.
> > To my limited understanding, on the plat-eznps arch, where each cpu is
> always running a single threaded process in
> > isolation mode, the above call is not necessary,
> > Can you please confirm (or reject) my understanding? If I'm wrong short
> explanation will be appreciated.
> > If I'm correct, I will be grateful if you can point me to how to skip this call in
> a way acceptable by community.
> 
> You've failed to explain why you think hotplug should be a performance
> critical path.
> 
> I'm also not seeing how it would be different from boot; you'd be
> looking at a similar cost for SMP bringup.