From owner-svn-src-projects@FreeBSD.ORG Sun Jul 1 17:15:51 2012 Return-Path: Delivered-To: svn-src-projects@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 11C3B106566C; Sun, 1 Jul 2012 17:15:51 +0000 (UTC) (envelope-from marcel@FreeBSD.org) Received: from svn.freebsd.org (svn.freebsd.org [IPv6:2001:4f8:fff6::2c]) by mx1.freebsd.org (Postfix) with ESMTP id F12F38FC0C; Sun, 1 Jul 2012 17:15:50 +0000 (UTC) Received: from svn.freebsd.org (localhost [127.0.0.1]) by svn.freebsd.org (8.14.4/8.14.4) with ESMTP id q61HFoec014224; Sun, 1 Jul 2012 17:15:50 GMT (envelope-from marcel@svn.freebsd.org) Received: (from marcel@localhost) by svn.freebsd.org (8.14.4/8.14.4/Submit) id q61HFoN6014222; Sun, 1 Jul 2012 17:15:50 GMT (envelope-from marcel@svn.freebsd.org) Message-Id: <201207011715.q61HFoN6014222@svn.freebsd.org> From: Marcel Moolenaar Date: Sun, 1 Jul 2012 17:15:50 +0000 (UTC) To: src-committers@freebsd.org, svn-src-projects@freebsd.org X-SVN-Group: projects MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Cc: Subject: svn commit: r237933 - projects/altix2/sys/ia64/ia64 X-BeenThere: svn-src-projects@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: "SVN commit messages for the src " projects" tree" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 01 Jul 2012 17:15:51 -0000 Author: marcel Date: Sun Jul 1 17:15:50 2012 New Revision: 237933 URL: http://svn.freebsd.org/changeset/base/237933 Log: Fire up our APs as soon as we kick the scheduler. We've woken them a long time back anyway, so they've been sitting idle for long enough. More importantly: the Altix PROM has assigned interrupts to CPUs and having our APs on hold means that we can't handle inteerupts that go to CPUs other than the APs -- this happens when we're doing interrupt based configuration (SI_SUB_INT_CONFIG_HOOKS). On NUMA machines like the Altix, it's infinitely better to have each brick probe and attach its own devices, so waking the APs sooner is generally a good idea. This is but a small step towards full SMP booting though... Modified: projects/altix2/sys/ia64/ia64/mp_machdep.c Modified: projects/altix2/sys/ia64/ia64/mp_machdep.c ============================================================================== --- projects/altix2/sys/ia64/ia64/mp_machdep.c Sun Jul 1 17:06:21 2012 (r237932) +++ projects/altix2/sys/ia64/ia64/mp_machdep.c Sun Jul 1 17:15:50 2012 (r237933) @@ -533,4 +533,4 @@ ipi_send(struct pcpu *cpu, int xiv) CTR3(KTR_SMP, "ipi_send(%p, %d): cpuid=%d", cpu, xiv, PCPU_GET(cpuid)); } -SYSINIT(start_aps, SI_SUB_SMP, SI_ORDER_FIRST, cpu_mp_unleash, NULL); +SYSINIT(start_aps, SI_SUB_KICK_SCHEDULER, SI_ORDER_ANY, cpu_mp_unleash, NULL);