Date: Thu, 4 Mar 2004 14:30:28 -0500 (EST) From: Andrew Gallatin <gallatin@cs.duke.edu> To: Vincent Poy <vince@oahu.WURLDLINK.NET> Cc: current@freebsd.org Subject: Re: -CURRENT kernel panic Message-ID: <16455.33748.860265.157572@grasshopper.cs.duke.edu> In-Reply-To: <20040304084600.I8264-100000@oahu.WURLDLINK.NET> References: <16455.17105.820332.248316@grasshopper.cs.duke.edu> <20040304084600.I8264-100000@oahu.WURLDLINK.NET>
next in thread | previous in thread | raw e-mail | index | archive | help
Vincent Poy writes:
> On Thu, 4 Mar 2004, Andrew Gallatin wrote:
>
> > Just add vm.kmem_size=VALUE to /boot/loader.conf. That way you don't
> > need to rebuild a kernel. I run with vm.kmem_size=429391872
>
> Does that actually work since I thought unless you alter the
> kernel, it won't use more than 200KB for the kmem_map?
Sure.. read the code (kern/kern_malloc.c:kmeminit) The
VM_KMEM_SIZE_MAX comes first, then the tunable from the loader.
> > Even after bumping the limits, I'm seeing a repeatable panic when
> > allocating a large amount of kernel memory ("panic: pmap_enter:
> > attempted pmap_enter on 4MB page"). See yesterday's message
> > with that subject for details..
>
> Just checked the archives and yours seems to jhave more variables
> than mines from the panic message.
Can you describe the panic you see?
Drew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?16455.33748.860265.157572>