On Mon, Nov 07, 2016 at 08:29:06AM +0100, Ingo Molnar wrote:
> Wondering what the practical effect of this is - is any current
> hardware affected, and if yes, what are the effects of this patch?
No affect on current hw - just a cleanup. It can go in 4.10. Only the
previous one is tagged for
On Mon, Nov 07, 2016 at 08:29:06AM +0100, Ingo Molnar wrote:
> Wondering what the practical effect of this is - is any current
> hardware affected, and if yes, what are the effects of this patch?
No affect on current hw - just a cleanup. It can go in 4.10. Only the
previous one is tagged for
* Borislav Petkov wrote:
> Ditto, clean it up and add tags:
>
> ---
> From: Yazen Ghannam
> Date: Tue, 1 Nov 2016 11:51:03 -0500
> Subject: [PATCH] x86/AMD: Group cpu_llc_id assignment
>
> Currently, we assume that a system has multiple last level caches
* Borislav Petkov wrote:
> Ditto, clean it up and add tags:
>
> ---
> From: Yazen Ghannam
> Date: Tue, 1 Nov 2016 11:51:03 -0500
> Subject: [PATCH] x86/AMD: Group cpu_llc_id assignment
>
> Currently, we assume that a system has multiple last level caches only
> if there are multiple nodes,
Ditto, clean it up and add tags:
---
From: Yazen Ghannam
Date: Tue, 1 Nov 2016 11:51:03 -0500
Subject: [PATCH] x86/AMD: Group cpu_llc_id assignment
Currently, we assume that a system has multiple last level caches only
if there are multiple nodes, and that the cpu_llc_id
Ditto, clean it up and add tags:
---
From: Yazen Ghannam
Date: Tue, 1 Nov 2016 11:51:03 -0500
Subject: [PATCH] x86/AMD: Group cpu_llc_id assignment
Currently, we assume that a system has multiple last level caches only
if there are multiple nodes, and that the cpu_llc_id is equal to the
Currently, we assume that a system has multiple last level caches only if
there are multiple nodes, and that the cpu_llc_id is equal to the node_id.
This no longer applies since Fam17h can have multiple last level caches
within a node.
So group the cpu_llc_id assignment by topology feature and
Currently, we assume that a system has multiple last level caches only if
there are multiple nodes, and that the cpu_llc_id is equal to the node_id.
This no longer applies since Fam17h can have multiple last level caches
within a node.
So group the cpu_llc_id assignment by topology feature and
8 matches
Mail list logo