From owner-freebsd-hackers Fri May 25 11:29:37 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from smtp10.phx.gblx.net (smtp10.phx.gblx.net [64.211.219.59]) by hub.freebsd.org (Postfix) with ESMTP id CB6AF37B423 for ; Fri, 25 May 2001 11:29:33 -0700 (PDT) (envelope-from tlambert@usr05.primenet.com) Received: (from daemon@localhost) by smtp10.phx.gblx.net (8.9.3/8.9.3) id LAA54944 for ; Fri, 25 May 2001 11:29:32 -0700 Received: from usr05.primenet.com(206.165.6.205) via SMTP by smtp10.phx.gblx.net, id smtpdn3EKMa; Fri May 25 11:29:24 2001 Received: (from tlambert@localhost) by usr05.primenet.com (8.8.5/8.8.5) id LAA11814 for hackers@freebsd.org; Fri, 25 May 2001 11:35:29 -0700 (MST) From: Terry Lambert Message-Id: <200105251835.LAA11814@usr05.primenet.com> Subject: Re: panics with 4GB on an IBM xSeries 330 To: hackers@freebsd.org Date: Fri, 25 May 2001 18:35:29 +0000 (GMT) X-Mailer: ELM [version 2.5 PL2] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG ] We have a 4GB IBM xSeries 330 (1GHz PIII) and I can't get 4.3-RELEASE to ] boot on it. I did set NKPT to 64 as suggested by DG about a week ago on ] this list (this is also the reason I take this to -hackers rather than ] -questions). Still, I get ] panic: swap_pager_swap_init: swap_zone=NULL ] when booting (both the modified kernel and GENERIC behave the same). An ] identical machine with 1GB works like a champ. Anything else other than ] NKPT I should set? Personally, I think NKPT is a red herring. I am running several 4G machines with the default of 32, and have not had problems (any problems will be automatically fixed for you by grow_kernel()). You aren't changing the KVA space tuning, are you? There are some really magic off-by-one numbers in the bowels of the VM startup on x86 processors. By default, the kernel is mapped into a 4G page as part of the startup; I have notices on some SMP borads (notably, Tyan Tiger LE's with v1.4 and 1.5 AMI BIOS) that the use of 4G pages will survive what is supposed to be a TLB shootdown. This was especially problematic with two or more processors in the box. In theory, you could fix this by reloading CR3, but the kernel also sets the PG_G (global) bit on the 4M page, so it's a bit of a problem (invltlb doesn't work right, IMO; under SMP, it uses a rendesvous, which seems to me to have a race condition). The easiest thing to do is to: options DISABLE_PSE Which may help you out of your current mess. If not, set "kern.maxfiles=120000" at the boot prompt, and if it works for you, add it to the loader.conf. This really sucks, because it eats memory for sockets, TCP connections, UDP connections, etc., all proportional to the maxfiles... Other than that... good luck. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message