lkml.org 
[lkml]   [2017]   [Jan]   [23]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH 8/8] x86/intel_rdt: rmdir,umount and hotcpu updates for MBA


On Mon, 16 Jan 2017, Thomas Gleixner wrote:

> On Tue, 10 Jan 2017, Vikas Shivappa wrote:
>
>> During rmdir only reset the ctrl values for the rdtgroup's closid. This
>> is done so that that next time when the closid is reused they dont
>> reflect old values.
>
> How on earth is that related to MBA?

This is not specific to MBA so i put it as x86/intel_rdt in the subject but will
send this as part of seperate series which you mentioned.

>
>> Remove the closid update during cpuonline in cqm as its
>> already in the CAT code. Since both cqm and CAT want the rmid and closid
>> to be zero when cpu is online, remove the PQR MSR write to zero during
>> cpuonline because the MSRs are at zero after cpu reset and also during
>
> Are you sure that the MSRs are zero when the cpu goes offline and then
> online again? That's not a reset.

Thats right - I was removing this since at the first sched_in the new CLOSID
would be updated anyways. I know this prevents from having a 0 CLOSid at the
time all the hot plug code is run, but we already have run the __cpu_disable and
all the hotplug code when going down on the last used CLOSid. (if the kernel
thread that handles the offline/online should have a 0 CLOS then thats fine
probably..)
So may be the right way is to actually zero the CLOS in PQR_MSR when we are
going down/offline ?

Thanks,
Vikas

>
> Thanks,
>
> tglx
>

\
 
 \ /
  Last update: 2017-01-23 21:18    [W:0.053 / U:0.128 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site