Date:   Wed, 22 Feb 2023 10:36:40 +0100
From:   Peter Zijlstra <peterz@infradead.org>
To:     Benjamin Segall <bsegall@google.com>
Cc:     shrikanth hegde <sshegde@linux.vnet.ibm.com>, mingo@redhat.com,
        Vincent Guittot <vincent.guittot@linaro.org>,
        dietmar.eggemann@arm.com, Thomas Gleixner <tglx@linutronix.de>,
        Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
        Arjan van de Ven <arjan@linux.intel.com>,
        svaidy@linux.ibm.com, linux-kernel@vger.kernel.org
Subject: Re: [RFC PATCH] sched/fair: Interleave cfs bandwidth timers for
 improved single thread performance at low utilization
Message-ID: <Y/XiKKDLcejSYc6+@hirez.programming.kicks-ass.net>
References: <9c57c92c-3e0c-b8c5-4be9-8f4df344a347@linux.vnet.ibm.com>
 <Y/Ov+DzOtA0iHfFO@hirez.programming.kicks-ass.net>
 <0c4d7bbb-3fef-031e-e9a1-a678ab68ade7@linux.vnet.ibm.com>
 <xm267cwa4ruo.fsf@google.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <xm267cwa4ruo.fsf@google.com>
Precedence: bulk

On Tue, Feb 21, 2023 at 01:43:27PM -0800, Benjamin Segall wrote:
> The value should never come up, so it's just a question of if it's fine
> to call get_random_* in early contexts, which I don't know offhand.

Should be, scheduler init is quite late as things go and people have
been pushing the random init earlier and earlier.