From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
        id S1753955AbcI1SAt (ORCPT <rfc822;w@1wt.eu>);
        Wed, 28 Sep 2016 14:00:49 -0400
Received: from smtprelay4.synopsys.com ([198.182.47.9]:55158 "EHLO
        smtprelay.synopsys.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1753866AbcI1SAg (ORCPT
        <rfc822;linux-kernel@vger.kernel.org>);
        Wed, 28 Sep 2016 14:00:36 -0400
Subject: Re: NMI for ARC
To: Peter Zijlstra <peterz@infradead.org>
References: <1445286926.3913.13.camel@synopsys.com>
 <C2D7FE5348E1B147BCA15975FBA23075F44D210D@IN01WEMBXA.internal.synopsys.com>
 <20151117110749.GT3816@twins.programming.kicks-ass.net>
 <564B0E18.3040207@synopsys.com>
 <20151117122401.GY3816@twins.programming.kicks-ass.net>
 <20151117122540.GL11639@twins.programming.kicks-ass.net>
 <564B2341.4030409@synopsys.com>
 <20151117131538.GZ3816@twins.programming.kicks-ass.net>
 <3b23d64c-2579-b72e-645a-e7ece3c1a3c6@synopsys.com>
 <20160928071642.GU2794@worktop>
CC: arcml <linux-snps-arc@lists.infradead.org>,
        Alexey Brodkin <Alexey.Brodkin@synopsys.com>,
        lkml <linux-kernel@vger.kernel.org>,
        "Andy Lutomirski" <luto@amacapital.net>
Newsgroups: gmane.linux.kernel.arc,gmane.linux.kernel
From: Vineet Gupta <Vineet.Gupta1@synopsys.com>
Message-ID: <3b5cbc7c-ed5c-af95-8dc9-25fd92277e98@synopsys.com>
Date: Wed, 28 Sep 2016 10:58:06 -0700
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101
 Thunderbird/45.3.0
MIME-Version: 1.0
In-Reply-To: <20160928071642.GU2794@worktop>
Content-Type: text/plain; charset="windows-1252"
Content-Transfer-Encoding: 7bit
X-Originating-IP: [10.9.131.247]
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On 09/28/2016 12:16 AM, Peter Zijlstra wrote:
>> 1. Is it OK in general to short-circuit preemption off irq checks for NMI style
>> > interrupts. 
> Yes. If the NMI returns to kernel space you must not attempt preemption
> for reasons you found :-), if the NMI returns to userspace you should do
> the normal return to user bits, I think.

Just to add, eliding preemption check in return path might not be sufficient as we
could still have timer intr leading to scheduler_tick() - whcih is what i found as
well ;-) So scheduler nevertheless needs to be told to not reschedule in this code
path.

So eliding preempt-off-irq just becomes an optimization for NMI code paths IMHO.

Now I was a stupid fool, fudging preemption counts in low level code, to achieve
this, whereas we have nice generic nmi_{enter,exit}() and in_nmi() helpers which
can be used.

Thx,
-Vineet