From owner-freebsd-arch@FreeBSD.ORG Mon Nov 8 15:34:35 2010 Return-Path: Delivered-To: freebsd-arch@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E85741065670; Mon, 8 Nov 2010 15:34:34 +0000 (UTC) (envelope-from mdf356@gmail.com) Received: from mail-iw0-f182.google.com (mail-iw0-f182.google.com [209.85.214.182]) by mx1.freebsd.org (Postfix) with ESMTP id 7467B8FC18; Mon, 8 Nov 2010 15:34:34 +0000 (UTC) Received: by iwn39 with SMTP id 39so6226556iwn.13 for ; Mon, 08 Nov 2010 07:34:34 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:received:in-reply-to :references:date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=Q3Yh+Akq1iSym1e61foxgBYYRwPfKywY/6SkwmPImWI=; b=ITcUReiQ1Qq6+npAlu+MYy/C+tV+7hfn9dJFlXSRAgcVYAL6htMDDzjvdP9FBOyCJ3 xqHXQmmg3dZTLffKV2XfMU4aS7YJ0rveLpzy1waUIieiU81hiIUcDvvaoep+vLqDqmsm bSKPQAB0xnd504G9JKn3MLrWA7alfB9GXieq0= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; b=jz8/c6Ax+rXemiy3JRzfAya/R4KsAa/5OBWvkkJi0BUWqPqB+qkt78OocoYx//nSja TsKH6qhsaLl/2Pn6fGTLavzGBkQgub1do2HzdV5WikYJGCr8Gn7fEC1yAFZNvJU0E1dP 74SnrgWFZpxa8AfXuMvNqW4Fzf8Rnx12vmt70= MIME-Version: 1.0 Received: by 10.231.156.139 with SMTP id x11mr4170511ibw.22.1289230473154; Mon, 08 Nov 2010 07:34:33 -0800 (PST) Received: by 10.231.21.35 with HTTP; Mon, 8 Nov 2010 07:34:33 -0800 (PST) In-Reply-To: <201011080947.00550.jhb@freebsd.org> References: <201011012054.59551.hselasky@c2i.net> <201011061522.26533.hselasky@c2i.net> <201011080947.00550.jhb@freebsd.org> Date: Mon, 8 Nov 2010 07:34:33 -0800 Message-ID: From: Matthew Fleming To: John Baldwin Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: freebsd-usb@freebsd.org, Weongyo Jeong , freebsd-current@freebsd.org, Andrew Thompson , freebsd-arch@freebsd.org, Hans Petter Selasky Subject: Re: [RFC] Outline of USB process integration in the kernel taskqueue system X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 08 Nov 2010 15:34:35 -0000 On Mon, Nov 8, 2010 at 6:47 AM, John Baldwin wrote: > On Saturday, November 06, 2010 4:33:17 pm Matthew Fleming wrote: >> On Sat, Nov 6, 2010 at 7:22 AM, Hans Petter Selasky w= rote: >> > Hi, >> > >> > On Saturday 06 November 2010 14:57:50 Matthew Fleming wrote: >> >> >> >> I think you're misunderstanding the existing taskqueue(9) implementat= ion. >> >> >> >> As long as TQ_LOCK is held, the state of ta->ta_pending cannot change= , >> >> nor can the set of running tasks. =A0So the order of checks is >> >> irrelevant. >> > >> > I agree that the order of checks is not important. That is not the pro= blem. >> > >> > Cut & paste from suggested taskqueue patch from Fleming: >> > >> > =A0> +int >> >> > +taskqueue_cancel(struct taskqueue *queue, struct task *task) >> >> > +{ >> >> > + =A0 =A0 =A0 int rc; >> >> > + >> >> > + =A0 =A0 =A0 TQ_LOCK(queue); >> >> > + =A0 =A0 =A0 if (!task_is_running(queue, task)) { >> >> > + =A0 =A0 =A0 =A0 =A0 =A0 =A0 if ((rc =3D task->ta_pending) > 0) >> >> > + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 STAILQ_REMOVE(&queue-= >tq_queue, task, task, >> >> > ta_link); + =A0 =A0 =A0 =A0 =A0 =A0 =A0 task->ta_pending =3D 0; >> >> > + =A0 =A0 =A0 } else { >> >> > + =A0 =A0 =A0 =A0 =A0 =A0 =A0 rc =3D -EBUSY; >> > >> > What happens in this case if ta_pending > 0. Are you saying this is no= t >> > possible? If ta_pending > 0, shouldn't we also do a STAILQ_REMOVE() ? >> >> Ah! =A0I see what you mean. >> >> I'm not quite sure what the best thing to do here is; I agree it would >> be nice if taskqueue_cancel(9) dequeued the task, but I believe it >> also needs to indicate that the task is currently running. =A0I guess >> the best thing would be to return the old pending count by reference >> parameter, and 0 or EBUSY to also indicate if there is a task >> currently running. >> >> Adding jhb@ to this mail since he has good thoughts on interfacing. > > I agree we should always dequeue when possible. =A0I think it should retu= rn > -EBUSY in that case. =A0That way code that uses 'cancel' followed by a > conditional 'drain' to implement a blocking 'cancel' will DTRT. Do we not also want the old ta_pending to be returned? In the case where a task is pending and is also currently running (admittedly a narrow window), how would we do this? This is why I suggested returning the old ta_pending by reference. This also allows callers who don't care about the old pending to pass NULL and ignore it. Thanks, matthew