[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAHp75Vc-frdJSAMxK1YpHwmPa_-0fTpRxq=QObcux3Jt=5+9kw@mail.gmail.com>
Date: Mon, 11 Apr 2022 18:07:45 +0300
From: Andy Shevchenko <andy.shevchenko@...il.com>
To: Qing Wang <wangqing@...o.com>
Cc: Sudeep Holla <sudeep.holla@....com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
"Rafael J. Wysocki" <rafael@...nel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] arch_topology: Do not set llc_sibling if llc_id is invalid
On Mon, Apr 11, 2022 at 12:10 PM Qing Wang <wangqing@...o.com> wrote:
>
> From: Wang Qing <wangqing@...o.com>
>
> When ACPI is not enabled, cpuid_topo->llc_id = cpu_topo->llc_id = -1, which
> will set llc_sibling 0xff(...), this is misleading.
Shouldn't it be a Fixes tag then?
> Don't set llc_sibling(default 0) if we don't know the cache topology.
...
> - if (cpuid_topo->llc_id == cpu_topo->llc_id) {
> + if (cpu_topo->llc_id != -1 && cpuid_topo->llc_id == cpu_topo->llc_id) {
I'm wondering if more strict check is better here, i.e.
if (cpu_topo->llc_id >= 0 && cpuid_topo->llc_id ==
cpu_topo->llc_id) {
> cpumask_set_cpu(cpu, &cpuid_topo->llc_sibling);
> cpumask_set_cpu(cpuid, &cpu_topo->llc_sibling);
> }
--
With Best Regards,
Andy Shevchenko
Powered by blists - more mailing lists