lkml.org 
[lkml]   [2016]   [Nov]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] x86/topology: Document cpu_llc_id
On Wed, Sep 07, 2016 at 12:36:41PM +0200, Peter Zijlstra wrote:
> On Wed, Sep 07, 2016 at 11:22:19AM +0200, Borislav Petkov wrote:
> > From: Borislav Petkov <bp@suse.de>
> >
> > It means different things on Intel and AMD so write it down so that
> > there's no confusion.
> >
> > Signed-off-by: Borislav Petkov <bp@suse.de>
> > Cc: Peter Zijlstra <peterz@infradead.org>
> > Cc: Thomas Gleixner <tglx@linutronix.de>
> > ---
> > Documentation/x86/topology.txt | 6 ++++++
> > 1 file changed, 6 insertions(+)
> >
> > diff --git a/Documentation/x86/topology.txt b/Documentation/x86/topology.txt
> > index 06afac252f5b..7a5485730476 100644
> > --- a/Documentation/x86/topology.txt
> > +++ b/Documentation/x86/topology.txt
> > @@ -63,6 +63,12 @@ The topology of a system is described in the units of:
> > The maximum possible number of packages in the system. Helpful for per
> > package facilities to preallocate per package information.
> >
> > + - cpu_llc_id:
> > +
> > + A per-CPU variable containing:
> > + - On Intel, the first APIC ID of the list of CPUs sharing the Last Level
> > + Cache
> > + - On AMD, the Node ID containing the Last Level Cache.
>
> And there are no AMD parts where there are multiple LLCs on a Node? Like
> where there isn't an L3 and the L2 is only per cluster?

Yes there are (now), see:

http://git.kernel.org/tip/b0b6e86846093c5f8820386bc01515f857dd8faa

But those LLC ID numbers are still increasing IDs of either nodes or
core complexes in Zen's case. So I think the text should say:

"- On AMD, the Node ID or Core Complex ID containing the Last Level Cache. In
general, numbers identifying an LLC uniquely on the system."

How's that?

--
Regards/Gruss,
Boris.

Good mailing practices for 400: avoid top-posting and trim the reply.

\
 
 \ /
  Last update: 2016-11-14 07:36    [W:0.557 / U:0.600 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site