From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ralf Baechle Subject: Re: [PATCH 10/15] MIPS: Add code for new system 'paravirt'. Date: Fri, 23 May 2014 23:47:36 +0200 Message-ID: <20140523214736.GI334@linux-mips.org> References: <1400597236-11352-1-git-send-email-andreas.herrmann@caviumnetworks.com> <1400597236-11352-11-git-send-email-andreas.herrmann@caviumnetworks.com> <537CAC74.4030800@imgtec.com> <20140523202855.GL11800@alberich> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: James Hogan , linux-mips@linux-mips.org, David Daney , kvm@vger.kernel.org, David Daney To: Andreas Herrmann Return-path: Received: from eddie.linux-mips.org ([78.24.191.182]:54799 "EHLO cvs.linux-mips.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751258AbaEWVry (ORCPT ); Fri, 23 May 2014 17:47:54 -0400 Received: from localhost.localdomain ([127.0.0.1]:45407 "EHLO linux-mips.org" rhost-flags-OK-OK-OK-FAIL) by eddie.linux-mips.org with ESMTP id S6855076AbaEWVrvSvBgY (ORCPT ); Fri, 23 May 2014 23:47:51 +0200 Content-Disposition: inline In-Reply-To: <20140523202855.GL11800@alberich> Sender: kvm-owner@vger.kernel.org List-ID: On Fri, May 23, 2014 at 10:28:55PM +0200, Andreas Herrmann wrote: > > > + wmb(); > > > > Wouldn't smp_wmb() be more accurate? > > ... use smp_wmb there ... A few years ago I reviewed the use of mb()/rmb()/wmb() as opposed to smp_mb()/smp_rmb()/smp_wmb() throughout the kernel. Every single use was a bug should either have been replaced by the smp_* variant because it was not necessary on uniprocessors, was pure cargocult programming or was used for a purpose such I/O where other solutions were required. Ralf