From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sudeep Holla Subject: Re: [PATCH] arm64: KVM: unregister notifiers in hyp mode teardown path Date: Mon, 4 Apr 2016 15:22:28 +0100 Message-ID: <570278A4.9060600@arm.com> References: <1459777611-22592-1-git-send-email-sudeep.holla@arm.com> <5702726D.7000505@arm.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: Received: from localhost (localhost [127.0.0.1]) by mm01.cs.columbia.edu (Postfix) with ESMTP id 61929407A5 for ; Mon, 4 Apr 2016 10:21:18 -0400 (EDT) Received: from mm01.cs.columbia.edu ([127.0.0.1]) by localhost (mm01.cs.columbia.edu [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6TGxycl8FVGg for ; Mon, 4 Apr 2016 10:21:16 -0400 (EDT) Received: from foss.arm.com (foss.arm.com [217.140.101.70]) by mm01.cs.columbia.edu (Postfix) with ESMTP id 9DA7A400E9 for ; Mon, 4 Apr 2016 10:21:15 -0400 (EDT) In-Reply-To: <5702726D.7000505@arm.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: kvmarm-bounces@lists.cs.columbia.edu Sender: kvmarm-bounces@lists.cs.columbia.edu To: Marc Zyngier , kvmarm@lists.cs.columbia.edu, linux-arm-kernel@lists.infradead.org Cc: Sudeep Holla List-Id: kvmarm@lists.cs.columbia.edu On 04/04/16 14:55, Marc Zyngier wrote: > Hi Sudeep, > > On 04/04/16 14:46, Sudeep Holla wrote: [...] >> @@ -1270,12 +1279,7 @@ static int init_hyp_mode(void) >> free_boot_hyp_pgd(); >> #endif >> >> - cpu_notifier_register_begin(); >> - >> - err = __register_cpu_notifier(&hyp_init_cpu_nb); >> - >> - cpu_notifier_register_done(); >> - >> + err = register_cpu_notifier(&hyp_init_cpu_nb); > > We went from something like this to the cpu_notifier_register_begin/end > with 8146875de ("arm, kvm: Fix CPU hotplug callback registration"). > > What makes it more acceptable now? > Correct, but in the initial code even init_hyp_mode was protected under cpu_notifier_register_begin, but IIUC recent re-org eliminated the need for that and the above code exactly resembles what register_cpu_notifier does. If that's not the case then we need to move cpu_notifier_register_begin further up and retain __register_cpu_notifier I mainly changed it to keep it consistent with unregister call. -- Regards, Sudeep