From owner-freebsd-hackers Wed Feb 17 11:31:46 1999 Delivered-To: freebsd-hackers@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (Postfix) with ESMTP id 9C65B1159B for ; Wed, 17 Feb 1999 11:31:15 -0800 (PST) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id LAA10322; Wed, 17 Feb 1999 11:31:12 -0800 (PST) (envelope-from dillon) Date: Wed, 17 Feb 1999 11:31:12 -0800 (PST) From: Matthew Dillon Message-Id: <199902171931.LAA10322@apollo.backplane.com> To: Tor.Egge@fast.no Cc: hackers@FreeBSD.ORG Subject: Re: Problems in VM structure ? References: <199902171744.JAA06913@apollo.backplane.com> <199902171910.UAA17721@midten.fast.no> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG :All zones with ZONE_INTERRUPT flag has preallocated KVA :and a nonzero max. --> 78 MB. : :SMP --> 4 MB per processor private kva :last page dir entry reserved for alternate page table --> 4 MB :Kernel text + data + bss --> 1.5 MB : : 78 MB + 4 MB + 4 MB + 1.5 MB + 170 MB --> 257.5 MB : :When looking at current kernel_map on a 512 MB SMP machine with :maxusers set to 256, VM_KMEM_SIZE and VM_KMEM_SIZE_MAX set to :128 MB and NMBCLUSTERS set to 12288: : : (kgdb) print kernel_map->size : $1 = 334737408 : :- Tor Egge Ach. So why isn't it panicing on boot? Why does the machine destabilize and die later on with all sorts of problems? Is it trying to use more page table entries then were allocated for the kernel_map? Is the submap allocator broken? Is there some test we can add to panic the machine at boot time if KVM is over-reserved? -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message