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 S1751775AbdJNJVd (ORCPT <rfc822;w@1wt.eu>);
        Sat, 14 Oct 2017 05:21:33 -0400
Received: from mail-wm0-f67.google.com ([74.125.82.67]:56014 "EHLO
        mail-wm0-f67.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751337AbdJNJVb (ORCPT
        <rfc822;linux-kernel@vger.kernel.org>);
        Sat, 14 Oct 2017 05:21:31 -0400
X-Google-Smtp-Source: ABhQp+QvBkK7xQNKYxs61wTEVbn6wpXl8tluK36eVA3TyGiwcehz6G3KHmjYMtxWmbXM2opfOj/rEw==
Date: Sat, 14 Oct 2017 18:21:29 +0900
From: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: LKML <linux-kernel@vger.kernel.org>, Petr Mladek <pmladek@suse.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Subject: Re: [PATCH] printk: Remove superfluous memory barriers from
 printk_safe
Message-ID: <20171014092129.GD2973@tigerII.localdomain>
References: <20171011124647.7781f98f@gandalf.local.home>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20171011124647.7781f98f@gandalf.local.home>
User-Agent: Mutt/1.9.1 (2017-09-22)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On (10/11/17 12:46), Steven Rostedt wrote:
> From: Steven Rostedt (VMware) <rostedt@goodmis.org>
> 
> The variable printk_safe_irq_ready is set and never cleared at system
> boot up, when there's only one CPU active. It is set before other
> CPUs come on line. Also, it is extremely unlikely that an NMI would
> trigger this early in boot up (which I wonder why we even have this
> variable at all).

it's not only NMI related, printk() recursion can happen at any stages,
including... um... wait a second. ... including the "before we set up
per-CPU areas" stage? hmm... smells like a bug?

do we need to move per-CPU printk_safe buffers out of per-CPU and turn
it into a global static buffer? like logbuf, and just give every CPU a
starting offset of its printk_safe_logbuf part.

IOW,

char printk_safe_logbuf[number of cpus * sizeof printk safe buffer];

cpu0 offset 0, up to sizeof printk safe buffer
cpu1 offset sizeof printk safe buffer, up to 2 * sizeof printk safe buffer

etc.

or... at least. avoid stoing to per-CPU printk-safe/printk-nmi buffers
unless we've got per-CPU areas set up? or am I hallucinating?

	-ss