From owner-freebsd-ia64 Fri Jan 3 20:34:16 2003 Delivered-To: freebsd-ia64@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BBECA37B405 for ; Fri, 3 Jan 2003 20:34:15 -0800 (PST) Received: from kayak.xcllnt.net (209-128-86-226.BAYAREA.NET [209.128.86.226]) by mx1.FreeBSD.org (Postfix) with ESMTP id 5607743EE5 for ; Fri, 3 Jan 2003 20:34:14 -0800 (PST) (envelope-from marcel@xcllnt.net) Received: from dhcp01.pn.xcllnt.net (dhcp01.pn.xcllnt.net [192.168.4.201]) by kayak.xcllnt.net (8.12.6/8.12.6) with ESMTP id h044Y8rT034192; Fri, 3 Jan 2003 20:34:09 -0800 (PST) (envelope-from marcel@kayak.pn.xcllnt.net) Received: from dhcp01.pn.xcllnt.net (localhost [127.0.0.1]) by dhcp01.pn.xcllnt.net (8.12.6/8.12.6) with ESMTP id h044ZP6P002120; Fri, 3 Jan 2003 20:35:25 -0800 (PST) (envelope-from marcel@dhcp01.pn.xcllnt.net) Received: (from marcel@localhost) by dhcp01.pn.xcllnt.net (8.12.6/8.12.6/Submit) id h044ZPci002119; Fri, 3 Jan 2003 20:35:25 -0800 (PST) (envelope-from marcel) Date: Fri, 3 Jan 2003 20:35:24 -0800 From: Marcel Moolenaar To: Arun Sharma Cc: freebsd-ia64@FreeBSD.ORG Subject: Re: unaligned fault in pmap_find_vhpt Message-ID: <20030104043524.GA2059@dhcp01.pn.xcllnt.net> References: <200301032303.gBQJBOs00863@unix-os.sc.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200301032303.gBQJBOs00863@unix-os.sc.intel.com> User-Agent: Mutt/1.5.1i Sender: owner-freebsd-ia64@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Fri, Jan 03, 2003 at 03:03:14PM -0800, Arun Sharma wrote: > > I saw a kernel mode unaligned fault during a compilation workload > yesterday on an SMP 5.0-RC1 kernel. The fault happened here: > > 0xe000000000aad660 : > > More info below. It looks like the pte_chain is getting corrupted > somehow. What is the locking scheme being used to protect pte > collision chains on an SMP kernel ? We don't really have a consistent locking scheme. We walk and update the VHPT from IVA interrupt code as well. Under high load, a SMP kernel corrupts process space. I haven't seen the unaligned fault you mention. As for locking: I've been thinking about a VHPT per CPU. With processor affinity, massive clustering, NUMA architectures and other complexities, there may be an advantage to pay for the additional faults if we can keep a flexible implementation and without so much lock contention. I haven't thought about hyper- threaded setups (ie a VHPT per thread or per CPU package) Thoughts? I'll play with SMP some more this weekend (also WRT you patch)... -- Marcel Moolenaar USPA: A-39004 marcel@xcllnt.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ia64" in the body of the message