[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZPI1IP38l/X7K/k9@swahl-home.5wahls.com>
Date: Fri, 1 Sep 2023 14:01:52 -0500
From: Steve Wahl <steve.wahl@....com>
To: Dave Hansen <dave.hansen@...el.com>
Cc: Steve Wahl <steve.wahl@....com>,
Dimitri Sivanich <dimitri.sivanich@....com>,
Russ Anderson <russ.anderson@....com>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
Dave Hansen <dave.hansen@...ux.intel.com>, x86@...nel.org,
"H. Peter Anvin" <hpa@...or.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] x86/platform/uv: Use alternate source for socket to node
data
On Fri, Sep 01, 2023 at 10:18:20AM -0700, Dave Hansen wrote:
> On 8/7/23 07:17, Steve Wahl wrote:
> > When nr_cpus is set to a smaller number than actually present, the
> > cpu_to_node() mapping information for unused CPUs is not available to
> > build_socket_tables(). This results in an incomplete table and will
> > later cause use of a -1 value for some array indexing, and eventual
> > kernel page faults.
> >
> > Switch to using the __apicid_to_node array, which still contains all
> > the information mapping apicids to nodes, even for CPUs disabled with
> > a reduced nr_cpus setting.
>
> Before, the lookup went:
>
> CPU => APICID => SOCKET
>
> But when the CPU wasn't present, there wasn't a way to start this lookup.
>
> So, instead of looping over all CPUs, looking up their APICIDs and
> mapping those to sockets, just take CPUs out of the equation entirely.
>
> Loop over all APICIDs which are mapped to a valid NUMA node. Then just
> extract the socket-id from the APICID.
>
> Right?
>
> That seems sane enough. It's also way less code than the previous approach.
Yes. That's it precisely. And, yes, way less code.
Thanks.
--> Steve Wahl
--
Steve Wahl, Hewlett Packard Enterprise
Powered by blists - more mailing lists