Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Vineeth Remanan Pillai <vpillai@digitalocean.com>
To:     Peter Zijlstra <peterz@infradead.org>
Cc:     Vineeth Remanan Pillai <vpillai@digitalocean.com>,
        Nishanth Aravamudan <naravamudan@digitalocean.com>,
        Julien Desfossez <jdesfossez@digitalocean.com>,
        Aubrey Li <aubrey.intel@gmail.com>,
        Tim Chen <tim.c.chen@linux.intel.com>, mingo@kernel.org,
        tglx@linutronix.de, pjt@google.com, torvalds@linux-foundation.org,
        linux-kernel@vger.kernel.org, subhra.mazumdar@oracle.com,
        fweisbec@gmail.com, keescook@chromium.org, kerrnel@google.com,
        Aaron Lu <aaron.lwe@gmail.com>,
        Valentin Schneider <valentin.schneider@arm.com>,
        Mel Gorman <mgorman@techsingularity.net>,
        Pawan Gupta <pawan.kumar.gupta@linux.intel.com>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Phil Auld <pauld@redhat.com>
Subject: Re: [RFC PATCH v2 09/17] sched: Introduce sched_class::pick_task()
Date:   Fri, 26 Apr 2019 16:10:56 +0000
Message-Id: <20190426161056.4624-1-vpillai@digitalocean.com>
In-Reply-To: <20190426140209.GA2623@hirez.programming.kicks-ass.net>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

>
> I didn't get around to reading the original discussion here, but how can
> that possibly be?
>
> I can see !se, in that case curr is still selected.
>
> I can also see !curr, in that case curr is put.
>
> But I cannot see !se && !curr, per the above check we know
> cfs_rq->nr_running != 0, so there must be a cfs task to find. This means
> either curr or se must exist.

This fix was suggested as a quick fix for a crash seen in v1. But
I agree with you that this should be a bug if it happens and should
be investigated. I have tried in v2 and can no longer reproduce the
crash. Will remove the check in v3.

Thanks