Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.34 as permitted sender) client-ip=23.128.96.34;
Date:   Thu, 12 Oct 2023 10:05:46 +0200
From:   Peter Zijlstra <peterz@infradead.org>
To:     Josh Don <joshdon@google.com>
Cc:     Ankit Jain <ankitja@vmware.com>, yury.norov@gmail.com,
        andriy.shevchenko@linux.intel.com, linux@rasmusvillemoes.dk,
        qyousef@layalina.io, pjt@google.com, bristot@redhat.com,
        vschneid@redhat.com, linux-kernel@vger.kernel.org,
        namit@vmware.com, amakhalov@vmware.com, srinidhir@vmware.com,
        vsirnapalli@vmware.com, vbrahmajosyula@vmware.com,
        akaher@vmware.com, srivatsa@csail.mit.edu
Subject: Re: [PATCH RFC] cpumask: Randomly distribute the tasks within
 affinity mask
Message-ID: <20231012080546.GI6307@noisy.programming.kicks-ass.net>
References: <20231011071925.761590-1-ankitja@vmware.com>
 <20231011105329.GA17066@noisy.programming.kicks-ass.net>
 <20231011114642.GA36521@noisy.programming.kicks-ass.net>
 <20231011135238.GG6337@noisy.programming.kicks-ass.net>
 <CABk29NvHx2saNLdYmQgt31R8W28p7=GUtXiiupgE5czXRBAx5g@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CABk29NvHx2saNLdYmQgt31R8W28p7=GUtXiiupgE5czXRBAx5g@mail.gmail.com>
Precedence: bulk

On Wed, Oct 11, 2023 at 04:55:35PM -0700, Josh Don wrote:
> Hey Peter,
> 
> > +static struct cpumask *root_domain_allowed(struct cpumask *newmask,
> > +                                          struct cpumask *scratch,
> > +                                          struct cpumask *valid)
> > +{
> > +       struct root_domain *rd;
> > +       struct cpumask *mask;
> > +       struct rq *rq;
> > +
> > +       int first = cpumask_first_and(newmask, valid);
> > +       if (first >= nr_cpu_ids)
> > +               return NULL;
> 
> This picks the first arbitrarily, but isn't it possible for the mask
> to span both isolated and non-isolated cpus? In which case, the rest
> of this just boils down to chance (ie. whatever the span happens to be
> for the first cpu here)?

Yes, but it matches historical behaviour :/

Like I said, ideally we'd flat out reject masks like this, but...