Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752808AbbBSRsY (ORCPT ); Thu, 19 Feb 2015 12:48:24 -0500 Received: from cantor2.suse.de ([195.135.220.15]:37931 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751996AbbBSRsX (ORCPT ); Thu, 19 Feb 2015 12:48:23 -0500 User-Agent: Kaiten Mail In-Reply-To: <20150219173255.GC15980@treble.redhat.com> References: <20150217212532.GJ11861@treble.redhat.com> <20150218152100.GZ5029@twins.programming.kicks-ass.net> <20150218171256.GA28553@treble.hsd1.ky.comcast.net> <20150219002058.GD5029@twins.programming.kicks-ass.net> <20150219041753.GA13423@treble.redhat.com> <20150219101607.GG5029@twins.programming.kicks-ass.net> <20150219162429.GA15980@treble.redhat.com> <20150219163359.GA25438@suse.cz> <20150219170353.GB15980@treble.redhat.com> <20150219171929.GA13178@suse.cz> <20150219173255.GC15980@treble.redhat.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=UTF-8 Subject: Re: [PATCH 1/3] sched: add sched_task_call() From: Vojtech Pavlik Date: Thu, 19 Feb 2015 18:48:15 +0100 To: Josh Poimboeuf , Vojtech Pavlik CC: Peter Zijlstra , Andrew Morton , Ingo Molnar , Jiri Kosina , Seth Jennings , linux-kernel@vger.kernel.org Message-ID: <7696686c-f714-42bf-9fb2-7a100c4c66d7@email.android.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1128 Lines: 26 On February 19, 2015 6:32:55 PM CET, Josh Poimboeuf wrote: >> Yes. I'm saying that rather than guaranteeing they don't enter the >> kernel (by having them spin) you can flip them in case they try to do >> that instead. That solves the race condition just as well. > >Ok, gotcha. > >We'd still need a safe way to check if they're in user space though. > >How about with a TIF_IN_USERSPACE thread flag? It could be cleared/set >right at the border. Then for running tasks it's as simple as: > >if (test_tsk_thread_flag(task, TIF_IN_USERSPACE)) > klp_switch_task_universe(task); The s390x arch used to have a TIF_SYSCALL, which was doing exactly that (well, negated). I think it would work well, but it isn't entirely for free: two instructions per syscall and an extra TIF bit, which are (near) exhausted on some archs. -- Vojtech Pavlik Director SuSE Labs -- 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/