Date: Mon, 23 Aug 1999 00:16:56 +0100 From: David Malone <dwmalone@maths.tcd.ie> To: Ollivier Robert <roberto@keltia.freenix.fr> Cc: freebsd-current@FreeBSD.ORG Subject: Re: Panic with NFSv3 on a CURRENT/SMP system Message-ID: <19990823001656.A39559@walton.maths.tcd.ie> In-Reply-To: <19990822222433.A11209@keltia.freenix.fr>; from Ollivier Robert on Sun, Aug 22, 1999 at 10:24:33PM %2B0200 References: <199908220230.WAA08616@lor.watermarkgroup.com> <19990822175059.H14964@freebie.lemis.com> <19990822222433.A11209@keltia.freenix.fr>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Aug 22, 1999 at 10:24:33PM +0200, Ollivier Robert wrote: > That's what I'm thinking but compiling NFS into the kernel "fixed" my > panic. The weird part is that I'm still using INVARIANT. I don't see why the > condition is not met when compiling all these together and is when using the > kld. As I understand it, compiling the kernel with INVARIENTS makes it uncompatable with modules compiled without INVARIENTS. The reason is probably to do with inline functions and the like - I see some inline functions in vm_zone.h which set and check certain variables only when INVARIANTS is defined. The variables seem also to be set and checked in vm_zone.c. So I suppose if you use an inline function to initialise something without INVARIENTS in a module, and then it is checked by the kernel which did have INVARIENTS defined things go boom... David. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990823001656.A39559>