Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757431Ab0GPSHI (ORCPT ); Fri, 16 Jul 2010 14:07:08 -0400 Received: from terminus.zytor.com ([198.137.202.10]:49446 "EHLO mail.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756846Ab0GPSHG (ORCPT ); Fri, 16 Jul 2010 14:07:06 -0400 Message-ID: <4C409F62.6030303@zytor.com> Date: Fri, 16 Jul 2010 11:05:22 -0700 From: "H. Peter Anvin" User-Agent: Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.1.10) Gecko/20100621 Fedora/3.0.5-1.fc13 Thunderbird/3.0.5 MIME-Version: 1.0 To: Avi Kivity CC: Mathieu Desnoyers , LKML , Linus Torvalds , Andrew Morton , Ingo Molnar , Peter Zijlstra , Steven Rostedt , Steven Rostedt , Frederic Weisbecker , Thomas Gleixner , Christoph Hellwig , Li Zefan , Lai Jiangshan , Johannes Berg , Masami Hiramatsu , Arnaldo Carvalho de Melo , Tom Zanussi , KOSAKI Motohiro , Andi Kleen , akpm@osdl.org, Jeremy Fitzhardinge , "Frank Ch. Eigler" Subject: Re: [patch 2/2] x86 NMI-safe INT3 and Page Fault References: <20100714154923.947138065@efficios.com> <20100714155804.252253097@efficios.com> <4C405078.20707@redhat.com> <20100716144927.GA22516@Krystal> <4C408D0C.5050709@redhat.com> <20100716165855.GA3836@Krystal> <4C409CBA.1050709@redhat.com> In-Reply-To: <4C409CBA.1050709@redhat.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1272 Lines: 32 On 07/16/2010 10:54 AM, Avi Kivity wrote: > On 07/16/2010 07:58 PM, Mathieu Desnoyers wrote: >> >>> Why is that kernel code calling vmalloc_sync_all()? If it is only NMI >>> which cannot take vmalloc faults, why bother? If not, why not? >>> >> Modules come as yet another example of stuff that is loaded in vmalloc'd space >> and can be accesses from NMI context. That would include oprofile, tracers, and >> probably others I'm forgetting about. >> > > Module loading can certainly take a vmalloc_sync_all() (though I agree > it's unpleasant). Anything else? > > Note perf is not modular at this time, but could be made so with > preempt/sched notifiers to hook the context switch. > Actually, module loading is already a performance problem; a lot of distros load sometimes hundreds of modules on startup, and it's heavily serialized, so I can see this being desirable to skip. I really hope noone ever gets the idea of touching user space from an NMI handler, though, and expecting it to work... -hpa -- 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/