On Tue, Jun 27, 2017 at 08:07:10PM +0700, Suravee Suthikulpanit wrote:
> As I have described in the cover letter, this patch series changes how
Which you didn't send to me... thanks for that.
> kernel derives cpu "package" from package-as-socket to package-as-die in
> order to fix following issu
+ Matt and Mel.
On Wed, Jun 28, 2017 at 03:26:10AM +0700, Suravee Suthikulpanit wrote:
> So, from the definition above, we would like all those 16 threads to be in
> the same sched-domain, where threads from C0,1,2,3 are in the same
> sched-group, and threads in C4,5,6,7 are in another sched-group
On 6/28/17 00:44, Borislav Petkov wrote:
So let's try to discuss this without using DIE sched-domain, CCX, etc,
and let's start simple.
So in that die graphic:
C0 | T0 T1 |||| T0 T1 | C4
||||
On Tue, Jun 27, 2017 at 06:32:34PM +, Ghannam, Yazen wrote:
> > you want all those threads to belong to a single scheduling group.
> > Correct?
> >
> > Now that thing has a memory controller attached to it, correct?
> >
> > If so, why is this thing not a logical NUMA node, as described in SRA
n, Leo
> ; Ghannam, Yazen ;
> Peter Zijlstra
> Subject: Re: [PATCH 1/2] x86/CPU/AMD: Present package as die instead of
> socket
>
> On Tue, Jun 27, 2017 at 11:54:12PM +0700, Suravee Suthikulpanit wrote:
> > The 8 threads sharing each L3 are already in the same sched-domain1
&
On Tue, Jun 27, 2017 at 11:54:12PM +0700, Suravee Suthikulpanit wrote:
> The 8 threads sharing each L3 are already in the same sched-domain1 (MC
> CCX). So, cpu0 is in the same sched-domain1 as cpu1,2,3,64,65,66,67. Here,
> we need the DIE sched-domain because it represents all cpus that are in the
Zijlstra
> Subject: Re: [PATCH 1/2] x86/CPU/AMD: Present package as die instead of
> socket
>
> On Tue, Jun 27, 2017 at 04:42:32PM +, Duran, Leo wrote:
>
> First of all, please do not top-post.
>
> > Are you saying that "amd.c' should be scheduler-awa
Boris,
On 6/27/17 20:42, Borislav Petkov wrote:
On Tue, Jun 27, 2017 at 08:07:10PM +0700, Suravee Suthikulpanit wrote:
What we are trying point out here is that (NUMA) "node" and "die" are the
same thing in most of AMD processors, not necessary trying to introduce
another term here.
So don't
On Tue, Jun 27, 2017 at 04:42:32PM +, Duran, Leo wrote:
First of all, please do not top-post.
> Are you saying that "amd.c' should be scheduler-aware?.. Really?
Please read again what I said.
> If so, are you saying that information returned by kernel-defined
> terms like 'Package', 'Core',
eo.
> -Original Message-
> From: Borislav Petkov [mailto:b...@alien8.de]
> Sent: Tuesday, June 27, 2017 11:32 AM
> To: Duran, Leo
> Cc: 'Thomas Gleixner' ; Suthikulpanit, Suravee
> ; x...@kernel.org; linux-
> ker...@vger.kernel.org; Ghannam, Yazen ;
> Peter
kernel@vger.kernel.org;
> Ghannam, Yazen ; Peter Zijlstra
>
> Subject: RE: [PATCH 1/2] x86/CPU/AMD: Present package as die instead of
> socket
>
> On Tue, 27 Jun 2017, Duran, Leo wrote:
> > > From: Thomas Gleixner [mailto:t...@linutronix.de] On Tue, 27 Jun
> > > 2
On Tue, Jun 27, 2017 at 04:23:16PM +, Duran, Leo wrote:
> My understating is that it is *not* our job at the "amd.c" level to
> return information that influences the scheduler in any prescribed
> way.
Your understanding is wrong.
The abstractions we present to the rest of the kernel is so th
vger.kernel.org; Ghannam, Yazen ;
> Peter Zijlstra
> Subject: Re: [PATCH 1/2] x86/CPU/AMD: Present package as die instead of
> socket
>
> On Tue, Jun 27, 2017 at 03:48:55PM +, Duran, Leo wrote:
> > Basically, in our case a Package may contain more than one L3 (i.e.,
> >
On Tue, 27 Jun 2017, Duran, Leo wrote:
> > From: Thomas Gleixner [mailto:t...@linutronix.de]
> > On Tue, 27 Jun 2017, Suravee Suthikulpanit wrote:
> > > On 6/27/17 17:48, Borislav Petkov wrote:
> > > > On Tue, Jun 27, 2017 at 01:40:52AM -0500, Suravee Suthikulpanit wrote:
> > > > > However, this is
On Tue, Jun 27, 2017 at 10:55:52PM +0700, Suravee Suthikulpanit wrote:
> The reason we are trying to present "package == NUMA node (die)" here is
> because the topology.txt defines package to contain a number of cores plus
> shared resources (e.g. DRAM controller, shared caches, etc). Since the
> c
On Tue, Jun 27, 2017 at 03:48:55PM +, Duran, Leo wrote:
> Basically, in our case a Package may contain more than one L3 (i.e.,
> in hardware terms, there may more than one 'Core complex' in a 'Die').
> The important point is that all logical processors (threads) that
> share an L3 have a common
On 6/27/17 21:21, Thomas Gleixner wrote:
On Tue, 27 Jun 2017, Suravee Suthikulpanit wrote:
On 6/27/17 17:48, Borislav Petkov wrote:
On Tue, Jun 27, 2017 at 01:40:52AM -0500, Suravee Suthikulpanit wrote:
However, this is not the case on AMD family17h multi-die processor
platforms, which can hav
ran, Leo ; Ghannam,
> Yazen ; Peter Zijlstra
> Subject: Re: [PATCH 1/2] x86/CPU/AMD: Present package as die instead of
> socket
>
> On Tue, 27 Jun 2017, Suravee Suthikulpanit wrote:
> > On 6/27/17 17:48, Borislav Petkov wrote:
> > > On Tue, Jun 27, 2017
Le 27/06/2017 16:21, Thomas Gleixner a écrit :
> On Tue, 27 Jun 2017, Suravee Suthikulpanit wrote:
>> On 6/27/17 17:48, Borislav Petkov wrote:
>>> On Tue, Jun 27, 2017 at 01:40:52AM -0500, Suravee Suthikulpanit wrote:
However, this is not the case on AMD family17h multi-die processor
pl
On Tue, 27 Jun 2017, Suravee Suthikulpanit wrote:
> On 6/27/17 17:48, Borislav Petkov wrote:
> > On Tue, Jun 27, 2017 at 01:40:52AM -0500, Suravee Suthikulpanit wrote:
> > > However, this is not the case on AMD family17h multi-die processor
> > > platforms, which can have up to 4 dies per socket as
On Tue, Jun 27, 2017 at 08:07:10PM +0700, Suravee Suthikulpanit wrote:
> What we are trying point out here is that (NUMA) "node" and "die" are the
> same thing in most of AMD processors, not necessary trying to introduce
> another term here.
So don't use it then. The whole topology topic is confus
Hi Boris,
On 6/27/17 17:48, Borislav Petkov wrote:
On Tue, Jun 27, 2017 at 01:40:52AM -0500, Suravee Suthikulpanit wrote:
According to the Documentation/x86/topology.txt, AMD nomenclature for
package is NUMA node (or die).
You guys keep talking about die so you should add the definition of
"d
Remove stable from CC - this is not how you do a stable submission.
See Documentation/process/stable-kernel-rules.rst
On Tue, Jun 27, 2017 at 01:40:52AM -0500, Suravee Suthikulpanit wrote:
> According to the Documentation/x86/topology.txt, AMD nomenclature for
> package is NUMA node (or die).
Yo
23 matches
Mail list logo
