linux-kernel - Re: [RFC PATCH] arch_topology: Use llc_id instead of package

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]

Date:   Mon, 16 May 2022 11:35:24 +0100
From:   Sudeep Holla <sudeep.holla@....com>
To:     Dietmar Eggemann <dietmar.eggemann@....com>
Cc:     Qing Wang <wangqing@...o.com>,
        Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
        "Rafael J . Wysocki" <rafael@...nel.org>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Sudeep Holla <sudeep.holla@....com>,
        Morten Rasmussen <morten.rasmussen@....com>,
        linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH] arch_topology: Use llc_id instead of package_id

On Fri, May 13, 2022 at 02:04:29PM +0200, Dietmar Eggemann wrote:
> On 13/05/2022 13:03, Sudeep Holla wrote:
> > On Fri, May 13, 2022 at 12:42:00PM +0200, Dietmar Eggemann wrote:
> >> On 13/05/2022 11:03, Sudeep Holla wrote:
> >>> On Fri, May 13, 2022 at 10:34:00AM +0200, Dietmar Eggemann wrote:
> >>
> >> [...]
> >>
> >>>> @@ -527,7 +528,8 @@ static int __init parse_core(struct device_node *core, int package_id,
> >>>>  			return -EINVAL;
> >>>>  		}
> >>>>
> >>>> -		cpu_topology[cpu].package_id = package_id;
> >>>> +		cpu_topology[cpu].package_id = 0;
> >>>
> >>> While the above looks good and matches with what I am attempting to do
> >>> as well ...
> >>>
> >>>> +		cpu_topology[cpu].llc_id = llc_id;
> >>>
> >>> This looks wrong for simple reason that this is derived incorrectly from
> >>> the cpu-map while there is no guarantee that it matches the last level
> >>> cache ID on the system as we didn't parse the cache topology for this.
> >>> So I disagree with this change as it might conflict with the actual and
> >>> correct llc_id.
> >>
> >> It might not match the LLC, that's true. Something we have already today
> >> in Android for DynamIQ clusters with big/Little. People using 1. level
> >> clusters to group CPUs according to uArch.
> >
> > Not sure if that is the correct representation of h/w cluster on those
> > platforms, but if they want to misguide OS with the f/w(DT in this case)
> > well that's their choice.
> >
> > The main point is we need to get the exact h/w topology information and
> > then we can decide how to present the below masks as required by the
> > scheduler for its sched domains.
> >
> >> My point is we manage to get:
> >>
> >> SMT - cpu_smt_mask()
> >> CLS - cpu_clustergroup_mask()
> >> MC  - cpu_coregroup_mask()
> >> DIE - cpu_cpu_mask()
> >>
> >> covered in ACPI with the cpu_topology[] structure and if we want CLS on
> >> DT we have to save cluster_id for the 2. level (DT) cluster.
> >>
> >
> > I am not sure on the above point. Even with ACPI PPTT we are just setting
> > cluster_id based on first or leaf level of the clusters ignoring the
>
> Not sure about this. cluster_id was introduced last year into ACPI PPTT
> commit c5e22feffdd7 ("topology: Represent clusters of CPUs within a
> die") to cover L3-tag (4 CPUs) within L3 (24 CPUs) on Kunpeng920 for
> instance.
>
> cat /sys/kernel/debug/sched/domains/cpu0/domain*/name
> CLS
> MC
> ... I skip the NUMA levels
>
> # cat /proc/schedstat | awk '{print $1 " " $2 }' | grep ^[cd] | head -5
> cpu0 0
> domain0 00000000,00000000,0000000f <--  4 CPUs <-- cluster_id
> domain1 00000000,00000000,00ffffff <-- 24 CPUs
>
> If you use cluster_id for 1. level cluster then you cover MC's 24 CPUs

OK, the way I am looking at this from CPU topology perspective seem to be
different from what you are expecting here w.r.t sched_domains.

IMO, these cpumasks on its own must represent real CPU topology and how it
is used via cpu_*_mask by the scheduler domains is different.

> > nesting ATM. And that's exactly what I am trying to get with this series[1]
> >
> >
> >> And that's why I proposed to (ab)use llc_id to form the MC mask.
> >>
> >
> > Sure, it is already supported IIUC by cpu_coregroup_mask in arch_topology.c
> > We just need to make sure llc_id is set correctly in case of DT. Now if
> > you are saying llc_sibling is not what you need but something else, then
> > we may need to add that new mask and update cpu_coregroup_mask to choose
> > that based on certain condition which I believe is bit complicated.
> >
> >> I'm not currently aware of another solution to get CLS somehow elegantly
> >> into a DT system.
> >
> > Will grouping of CPUs into cluster they belong not good enough for CLS ?
>
> No, IMHO then you'll cover MC and it's cpu_coregroup_mask() and you get
> MC.                                        ^^^^
>

I see on Juno with SCHED_CLUSTER and cluster masks set, I see CLS and MC
domains.

> > I thought that should suffice based on what we have in cpu_clustergroup_mask
> > (i.e. cluster sibling mask)
>
> For one level (MC) yes, but not for 2 (MC and CLS). And cluster_id was
> introduces for the 2. level.
>

That sounds wrong and not what ACPI PPTT code says. My series just aligns
with what is done with ACPI PPTT IIUC. I need to check that again if my
understand differs from what is being done. But the example of Kunpeng920
aligns with my understanding.

--
Regards,
Sudeep