[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20161107140746.GA20626@gmail.com>
Date: Mon, 7 Nov 2016 15:07:46 +0100
From: Ingo Molnar <mingo@...nel.org>
To: Borislav Petkov <bp@...e.de>
Cc: x86@...nel.org, Yazen Ghannam <Yazen.Ghannam@....com>,
linux-kernel@...r.kernel.org,
Peter Zijlstra <a.p.zijlstra@...llo.nl>
Subject: Re: [PATCH v3 1/2] x86/AMD: Fix cpu_llc_id for AMD Fam17h systems
* Borislav Petkov <bp@...e.de> wrote:
> On Mon, Nov 07, 2016 at 08:31:21AM +0100, Ingo Molnar wrote:
> > > cpu_llc_id (Last Level Cache ID) derivation on AMD Fam17h has an
> > > underflow bug when extracting the socket_id value. It starts from 0
>
> How's this...
>
> > > so subtracting 1 from it will result in an invalid value. This breaks
> > > scheduling topology later on since the cpu_llc_id will be incorrect.
> ^^^^^^^^^^^^^^^^^^^
> ... here?
>
> > Same question as for the previous patch: what are the effects of the bug:
>
> See above.
There's many ways the scheduling topology can 'break', resulting in different
effects:
- scheduling domains might be mixed up to the extent of crashing the bootup
- some cores might be missing altogether, reducing available CPUs in essence
- cache domains might be seriously mixed up, resulting in serious drop in
performance.
- or domains might be partitioned 'wrong' but not catastrophically
wrong, resulting in a minor performance drop (if at all)
... do we know which of these occurs in this situation?
Thanks,
Ingo
Powered by blists - more mailing lists