Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753725AbdCMVF7 (ORCPT ); Mon, 13 Mar 2017 17:05:59 -0400 Received: from mail.kernel.org ([198.145.29.136]:49204 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751229AbdCMVFx (ORCPT ); Mon, 13 Mar 2017 17:05:53 -0400 MIME-Version: 1.0 In-Reply-To: <20170313165507.GJ3312@twins.programming.kicks-ass.net> References: <20170313165507.GJ3312@twins.programming.kicks-ass.net> From: Andy Lutomirski Date: Mon, 13 Mar 2017 14:05:27 -0700 X-Gmail-Original-Message-ID: Message-ID: Subject: Re: perf: race with automatic rdpmc() disabling To: Peter Zijlstra Cc: Andy Lutomirski , Vince Weaver , "linux-kernel@vger.kernel.org" , Ingo Molnar , Arnaldo Carvalho de Melo Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1030 Lines: 27 On Mon, Mar 13, 2017 at 9:55 AM, Peter Zijlstra wrote: > On Mon, Mar 13, 2017 at 09:44:02AM -0700, Andy Lutomirski wrote: >> static void x86_pmu_event_mapped(struct perf_event *event) >> { >> if (!(event->hw.flags & PERF_X86_EVENT_RDPMC_ALLOWED)) >> return; >> >> if (atomic_inc_return(¤t->mm->context.perf_rdpmc_allowed) == 1) >> >> <-- thread 1 stalls here >> >> on_each_cpu_mask(mm_cpumask(current->mm), refresh_pce, NULL, 1); >> } >> >> Suppose you start with perf_rdpmc_allowed == 0. Thread 1 runs >> x86_pmu_event_mapped and gets preempted (or just runs slowly) where I >> marked. Then thread 2 runs the whole function, does *not* update CR4, >> returns to userspace, and GPFs. >> >> The big hammer solution is to stick a per-mm mutex around it. Let me >> ponder whether a smaller hammer is available. > > Reminds me a bit of what we ended up with in kernel/jump_label.c:static_key_slow_inc(). > > One thing I don't get: isn't mmap_sem held for write the whole time?