Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751446AbbFMGrS (ORCPT ); Sat, 13 Jun 2015 02:47:18 -0400 Received: from mail-wi0-f175.google.com ([209.85.212.175]:34432 "EHLO mail-wi0-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750765AbbFMGrK (ORCPT ); Sat, 13 Jun 2015 02:47:10 -0400 Date: Sat, 13 Jun 2015 08:47:05 +0200 From: Ingo Molnar To: Oleg Nesterov Cc: linux-kernel@vger.kernel.org, linux-mml@vger.kernel.org, Andy Lutomirski , Andrew Morton , Denys Vlasenko , Brian Gerst , Peter Zijlstra , Borislav Petkov , "H. Peter Anvin" , Linus Torvalds , Thomas Gleixner , Waiman Long Subject: Re: [PATCH 06/12] x86/mm: Enable and use the arch_pgd_init_late() method Message-ID: <20150613064705.GA13835@gmail.com> References: <1434031637-9091-1-git-send-email-mingo@kernel.org> <1434031637-9091-7-git-send-email-mingo@kernel.org> <20150612225000.GA24699@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150612225000.GA24699@redhat.com> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1526 Lines: 53 * Oleg Nesterov wrote: > On 06/11, Ingo Molnar wrote: > > > > +void arch_pgd_init_late(struct mm_struct *mm, pgd_t *pgd) > > +{ > > + /* > > + * This is called after a new MM has been made visible > > + * in fork() or exec(). > > + * > > + * This barrier makes sure the MM is visible to new RCU > > + * walkers before we initialize it, so that we don't miss > > + * updates: > > + */ > > + smp_wmb(); > > I can't understand the comment and the barrier... > > Afaics, we need to ensure that: > > > + if (pgd_val(*pgd_src)) > > + WRITE_ONCE(*pgd_dst, *pgd_src); > > either we notice the recent update of this PGD, or (say) the subsequent > sync_global_pgds() can miss the child. > > How the write barrier can help? So the real thing this pairs with is the earlier: tsk->mm = mm; plus the linking of the new task in the task list. _that_ write must become visible to others before we do the (conditional) copy ourselves. Granted, it happens quite a bit earlier, and the task linking's use of locking is a natural barrier - but since this is lockless I didn't want to leave a silent assumption in. Perhaps remove the barrier and just leave a comment in that describes the assumption on task-linking being a full barrier? Thanks, Ingo -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/