From mboxrd@z Thu Jan 1 00:00:00 1970 From: "H. Peter Anvin" Subject: Re: start_kernel(): bug: interrupts were enabled early Date: Wed, 31 Mar 2010 15:36:26 -0700 Message-ID: <4BB3CE6A.2080406@zytor.com> References: <20100325194100.GA2364@debian> <20100331134048.da4e35a7.akpm@linux-foundation.org> <4BB3B4DB.7040904@kernel.org> <1270074687.7101.74.camel@pasglop> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Return-path: Received: from terminus.zytor.com ([198.137.202.10]:58322 "EHLO mail.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755213Ab0CaWke (ORCPT ); Wed, 31 Mar 2010 18:40:34 -0400 In-Reply-To: <1270074687.7101.74.camel@pasglop> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Benjamin Herrenschmidt Cc: Yinghai Lu , Andrew Morton , Rabin Vincent , lkml , penberg@cs.helsinki.fi, cl@linux-foundation.org, linux-arch@vger.kernel.org On 03/31/2010 03:31 PM, Benjamin Herrenschmidt wrote: > On Wed, 2010-03-31 at 13:47 -0700, Yinghai Lu wrote: >>> perhaps the second one isn't needed? Perhaps no architecture >> requires >>> that local interrupts be disabled across the above initialisations? >> >> spin_unlock_irq from arm is different from other archs? > > No, it's not, it will enable IRQs and thats illegal to do so early > during boot. We've been over that one again and again, the problem is > that people want to keep using that instead of irqsave/restore because > it's a nano-optimisation on x86... oh well... > Well, guess what... the particular user in this case *isn't used at all on x86* so it is a non-issue here. So I take it we have your particular vote to use irqsave/irqrestore in lib/rwsem-spinlock.c? -hpa