[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAO9wTFitMEzPXNqrMs7U=+M3jAghZtO1iXYbw5YBL1bwpB_Ecg@mail.gmail.com>
Date: Thu, 24 Jul 2025 23:27:33 +0530
From: Suchit K <suchitkarunakaran@...il.com>
To: "Russell King (Oracle)" <linux@...linux.org.uk>
Cc: ardb@...nel.org, linus.walleij@...aro.org, ebiggers@...gle.com,
nathan@...nel.org, lumag@...nel.org, dave@...ilevsky.ca,
ruanjinjie@...wei.com, kees@...nel.org, liuyuntao12@...wei.com,
linux-arm-kernel@...ts.infradead.org, skhan@...uxfoundation.org,
linux-kernel-mentees@...ts.linux.dev, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] arm: allow single CPU configuration by adjusting NR_CPUS
range and defaults
On Thu, 24 Jul 2025 at 23:16, Russell King (Oracle)
<linux@...linux.org.uk> wrote:
>
> On Thu, Jul 24, 2025 at 10:56:03PM +0530, Suchit Karunakaran wrote:
> > Previously, the NR_CPUS config for ARM required a minimum of 2 CPUs.
> > This patch changes the minimum NR_CPUS to 1 when SMP is not enabled,
> > sets range to exactly 1 if !SMP, allowing only a single CPU setting and
> > adds conditional defaults:
> > - default to 1 if SMP is disabled (uniprocessor)
> > - default to 4 if SMP is enabled (multiprocessor)
> >
> > Signed-off-by: Suchit Karunakaran <suchitkarunakaran@...il.com>
>
> This hasn't been tested. NR_CPUS depends on SMP, so if SMP is not
> enabled, then NR_CPUS doesn't exist in the configuration. Therefore,
> providing a default for !SMP is meaningless.
>
Oops, I think I misunderstood the intent of commit
278d1ed65e25d80af7c3a112d707b3f70516ddb4(cpumask: make CONFIG_NR_CPUS
always valid). And sorry for not testing it, I wasn't exactly sure how
to test it. Could you please point out the issues so I can fix them
properly? I appreciate your patience. I’m still learning my way around
kernel development and trying to improve. Thanks!
Powered by blists - more mailing lists