Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752037AbbFNH75 (ORCPT ); Sun, 14 Jun 2015 03:59:57 -0400 Received: from mail-wi0-f181.google.com ([209.85.212.181]:36854 "EHLO mail-wi0-f181.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751494AbbFNH7t (ORCPT ); Sun, 14 Jun 2015 03:59:49 -0400 Date: Sun, 14 Jun 2015 09:59:43 +0200 From: Ingo Molnar To: Oleg Nesterov Cc: linux-kernel@vger.kernel.org, linux-mm@kvack.org, Andy Lutomirski , Andrew Morton , Denys Vlasenko , Brian Gerst , Peter Zijlstra , Borislav Petkov , "H. Peter Anvin" , Linus Torvalds , Thomas Gleixner , Waiman Long Subject: Re: why do we need vmalloc_sync_all? Message-ID: <20150614075943.GA810@gmail.com> References: <1434188955-31397-1-git-send-email-mingo@kernel.org> <20150613185828.GA32376@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20150613185828.GA32376@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: 1913 Lines: 48 * Oleg Nesterov wrote: > I didn't read v2 yet, but I'd like to ask a question. > > Why do we need vmalloc_sync_all()? > > It has a single caller, register_die_notifier() which calls it without > any explanation. IMO, this needs a comment at least. Yes, it's used to work around crashes in modular callbacks: if the callbacks happens to be called from within the page fault path, before the vmalloc page fault handler runs, then we have a catch-22 problem. It's rare but not entirely impossible. > I am not sure I understand the changelog in 101f12af correctly, but at first > glance vmalloc_sync_all() is no longer needed at least on x86, do_page_fault() > no longer does notify_die(DIE_PAGE_FAULT). And btw DIE_PAGE_FAULT has no users. > DIE_MNI too... > > Perhaps we can simply kill it on x86? So in theory we could still have it run from DIE_OOPS, and that could turn a survivable kernel crash into a non-survivable one. Note that all of this will go away if we also do the vmalloc fault handling simplification that I discussed with Andy: - this series already makes the set of kernel PGDs strictly monotonically increasing during the lifetime of the x86 kernel - if in a subsequent patch we can synchronize new PGDs right after the vmalloc code creates it, before the area is used - so we can remove vmalloc_fault() altogether [or rather, turn it into a debug warning initially]. vmalloc_fault() is a clever but somewhat fragile complication. - after that we can simply remove vmalloc_sync_all() from x86, because all active vmalloc areas will be fully instantiated, all the time, on x86. 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/