From owner-freebsd-emulation@FreeBSD.ORG Fri Mar 21 18:08:35 2008 Return-Path: Delivered-To: freebsd-emulation@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id B99691065675 for ; Fri, 21 Mar 2008 18:08:35 +0000 (UTC) (envelope-from nox@saturn.kn-bremen.de) Received: from gwyn.kn-bremen.de (gwyn.kn-bremen.de [212.63.36.242]) by mx1.freebsd.org (Postfix) with ESMTP id 693978FC23 for ; Fri, 21 Mar 2008 18:08:35 +0000 (UTC) (envelope-from nox@saturn.kn-bremen.de) Received: by gwyn.kn-bremen.de (Postfix, from userid 10) id 523792A7997; Fri, 21 Mar 2008 19:08:34 +0100 (CET) Received: from saturn.kn-bremen.de (nox@localhost [127.0.0.1]) by saturn.kn-bremen.de (8.14.2/8.13.8) with ESMTP id m2LI875I094952 for ; Fri, 21 Mar 2008 19:08:07 +0100 (CET) (envelope-from nox@saturn.kn-bremen.de) Received: (from nox@localhost) by saturn.kn-bremen.de (8.14.2/8.13.6/Submit) id m2LI87Z3094951 for freebsd-emulation@FreeBSD.org; Fri, 21 Mar 2008 19:08:07 +0100 (CET) (envelope-from nox) From: Juergen Lock Date: Fri, 21 Mar 2008 19:08:07 +0100 To: freebsd-emulation@FreeBSD.org Message-ID: <20080321180807.GA93940@saturn.kn-bremen.de> Mail-Followup-To: freebsd-emulation@FreeBSD.org MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.16 (2007-06-09) Cc: Subject: new cpu affinity syscalls and (k)qemu on amd64 SMP X-BeenThere: freebsd-emulation@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Development of Emulators of other operating systems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 21 Mar 2008 18:08:35 -0000 Hi! There's a (remote?) possibility that the amd64 SMP kqemu crashes could be cured by pinning the qemu process to a single cpu (tho, if its that, I wonder why it works on i386 without doing that...) Anyway, could the new syscalls be used to pin a process to the cpu it was started on, and if so, how? Or, what would be the best way to ensure multiple qemu instances don't all end up on the same cpu and still won't be moved between cpus once they are running? Wondering... Juergen PS: Yeah I tried debugging this via running qemu in qemu but got nowhere, I'm beginning to suspect the cpus end up scribbling over each other's stack...