Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20200703155749.GA6255@pc636> <CAEJqkgiFFh8CvXkM4ZzXxNQmOJLL7WcgDL6rM83safNgEewZ9w@mail.gmail.com>
 <20200703174539.GA4800@hirez.programming.kicks-ass.net>
In-Reply-To: <20200703174539.GA4800@hirez.programming.kicks-ass.net>
From:   Gabriel C <nix.or.die@googlemail.com>
Date:   Fri, 3 Jul 2020 20:36:02 +0200
Message-ID: <CAEJqkgh0samT7TkLqhBLw5x8-bV1iAX1a-PHxtgruMgpdsfvBQ@mail.gmail.com>
Subject: Re: nr_cpu_ids vs AMD 3970x(32 physical CPUs)
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Uladzislau Rezki <urezki@gmail.com>,
        LKML <linux-kernel@vger.kernel.org>, linux-mm@kvack.org,
        Andrew Morton <akpm@linux-foundation.org>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        "Rafael J. Wysocki" <rjw@rjwysocki.net>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Am Fr., 3. Juli 2020 um 19:45 Uhr schrieb Peter Zijlstra <peterz@infradead.org>:
>
> On Fri, Jul 03, 2020 at 07:07:39PM +0200, Gabriel C wrote:
>
> > I boot all the boxes restricting the cores to the correct count on the
> > command line.
>
> This, because you're right about the wasted memory.
>
> > Wasted resource or not, this is still a bug IMO.
>
> Yeah, but not one we can do much about I think. It is the BIOS saying it
> wants more because it expects someone to come along and stick another
> CPU in.
>
> Possible we could say that for single socket machines overprovisioning
> is 'silly', but then, I've no idea how to detect that. You'll need to
> find an ACPI person.

I know the EPYC box got that problem too initially, it reported twice
the cores and twice the sockets,
but got fixed in some kernel versions.

https://lkml.org/lkml/2018/5/20/102

I never really looked at how this is calculated but I still believe
there is a bug somewhere.