linux-kernel - Re: [RFC PATCH 17/23] thermal: intel: hfi: Enable the Intel Thread Director

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: <20221006015030.GC29251@ranerica-svr.sc.intel.com>
Date:   Wed, 5 Oct 2022 18:50:30 -0700
From:   Ricardo Neri <ricardo.neri-calderon@...ux.intel.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Juri Lelli <juri.lelli@...hat.com>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Ricardo Neri <ricardo.neri@...el.com>,
        "Ravi V. Shankar" <ravi.v.shankar@...el.com>,
        Ben Segall <bsegall@...gle.com>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Len Brown <len.brown@...el.com>, Mel Gorman <mgorman@...e.de>,
        "Rafael J. Wysocki" <rafael.j.wysocki@...el.com>,
        Srinivas Pandruvada <srinivas.pandruvada@...ux.intel.com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Tim Chen <tim.c.chen@...ux.intel.com>,
        Valentin Schneider <vschneid@...hat.com>, x86@...nel.org,
        linux-kernel@...r.kernel.org, "Tim C . Chen" <tim.c.chen@...el.com>
Subject: Re: [RFC PATCH 17/23] thermal: intel: hfi: Enable the Intel Thread
 Director

On Tue, Sep 27, 2022 at 02:00:39PM +0200, Peter Zijlstra wrote:
> On Fri, Sep 09, 2022 at 04:11:59PM -0700, Ricardo Neri wrote:
> 
> > +config INTEL_THREAD_DIRECTOR
> > +	bool "Intel Thread Director"
> > +	depends on INTEL_HFI_THERMAL
> > +	depends on SMP
> > +	select SCHED_TASK_CLASSES
> > +	help
> > +	  Select this option to enable the Intel Thread Director. If selected,
> > +	  hardware classifies tasks based on the type of instructions they
> > +	  execute. It also provides performance capabilities for each class of
> > +	  task. On hybrid processors, the scheduler uses this data to place
> > +	  tasks of classes of higher performance on higher-performnance CPUs.
> 
> Do we really need yet another CONFIG symbol for all this!? AFAICT this
> Thread Director crud simply extends the HFI table and doesn't actually
> carry that much code with it.
> 
> Best to always have it on when HFI is on, no?

I decided to add CONFIG_INTEL_THREAD_DIRECTOR mainly to select CONFIG_IPC_
CLASS and have the needed members of task_struct only when needed. Legacy,
classless, HFI can work with the Thread Director part on some processors
(e.g., Sapphire Rapids or, FWIW, Lakefield).

I could get rid of CONFIG_INTEL_THREAD_DIRECTOR and instead wrap the Thread
Director code in an #ifdef(CONFIG_IPC_CLASS) block instead.

Thanks and BR,
Ricardo