From owner-freebsd-current@FreeBSD.ORG Mon May 9 13:47:08 2011 Return-Path: Delivered-To: current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 1A4941065673; Mon, 9 May 2011 13:47:08 +0000 (UTC) (envelope-from jhb@freebsd.org) Received: from cyrus.watson.org (cyrus.watson.org [65.122.17.42]) by mx1.freebsd.org (Postfix) with ESMTP id E564E8FC1B; Mon, 9 May 2011 13:47:07 +0000 (UTC) Received: from bigwig.baldwin.cx (66.111.2.69.static.nyinternet.net [66.111.2.69]) by cyrus.watson.org (Postfix) with ESMTPSA id 83CF046B1A; Mon, 9 May 2011 09:47:07 -0400 (EDT) Received: from jhbbsd.localnet (unknown [209.249.190.124]) by bigwig.baldwin.cx (Postfix) with ESMTPSA id 1C8D08A027; Mon, 9 May 2011 09:47:07 -0400 (EDT) From: John Baldwin To: freebsd-current@freebsd.org Date: Mon, 9 May 2011 09:35:53 -0400 User-Agent: KMail/1.13.5 (FreeBSD/8.2-CBSD-20110325; KDE/4.5.5; amd64; ; ) References: <4DC512D6.9070904@FreeBSD.org> In-Reply-To: <4DC512D6.9070904@FreeBSD.org> MIME-Version: 1.0 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Message-Id: <201105090935.54264.jhb@freebsd.org> X-Greylist: Sender succeeded SMTP AUTH, not delayed by milter-greylist-4.2.6 (bigwig.baldwin.cx); Mon, 09 May 2011 09:47:07 -0400 (EDT) Cc: current@freebsd.org, Andriy Gapon Subject: Re: COUNT_IPIS vs CPU_FOREACH X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 09 May 2011 13:47:08 -0000 On Saturday, May 07, 2011 5:37:26 am Andriy Gapon wrote: > > I believe that the following change is needed to fix COUNT_IPIS option. > Right now it seems to be a noop. > > > mp_ipi_intrcnt: CPU_FOREACH can't be used this early > > ... because all_cpus is not set yet. Have you tested this? all_cpus is set by start_all_aps() from cpu_mp_start() which runs at SI_SUB_CPU. This SYSINIT runs later at SI_SUB_INTR. So I think CPU_FOREACH() should be fine here. -- John Baldwin