From owner-freebsd-current Thu Jan 8 12:39:33 1998 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id MAA27048 for current-outgoing; Thu, 8 Jan 1998 12:39:33 -0800 (PST) (envelope-from owner-freebsd-current) Received: from nomis.simon-shapiro.org (nomis.i-Connect.Net [206.190.143.100]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id MAA26954 for ; Thu, 8 Jan 1998 12:38:29 -0800 (PST) (envelope-from shimon@nomis.Simon-Shapiro.ORG) Received: (qmail 28231 invoked by uid 1000); 7 Feb 2036 11:29:14 -0000 Message-ID: X-Mailer: XFMail 1.3-alpha-010198 [p0] on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199801072203.OAA16705@implode.root.com> Date: Thu, 07 Feb 2036 03:29:13 -0800 (PST) Reply-To: shimon@simon-shapiro.org Organization: The Simon Shapiro Foundation From: Simon Shapiro To: David Greenman Subject: Re: Panic: kmem_malloc: kmem_map too small Cc: freebsd-current@FreeBSD.ORG Sender: owner-freebsd-current@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On 07-Jan-98 David Greenman wrote: >>I receive that mainly under SMP, but looking at vm/vm_kern.c I do not see > > Sounds like you need to increase the kmem_map size - you're running > out > of malloc space. This can be done by changing VM_KMEM_SIZE in > /sys/i386/include/vmparam.h. The problem might indicate that you have a > very large system, a mistuned system, or perhaps could even indicate a > kernel memory leak. I suggest that you carefully analyze the output of > "vmstat -m" to determine where it's all going. I doubled this (VM_KMEM_SIZE) variable and gained a bounce buffers out of range panic. I disabled the bounce buffers and the problem is gone. This indicates to me (the trivially obvious) that certain kernel tunables are related to each other. Critically so. I would have loved to help in building a configuration system that would take these inter-dependancies into account but am totally unqualified for the job. Do you think a qualified core member could either do the job, or create (as comprehensive as possible) list of these symbols, and their interactions. If the latter is done, I will device the mechanism, if so desired. Thanx again for your help. Simon