Date: Mon, 03 Dec 2001 21:15:53 +0100 From: Uwe Doering <gemini@geminix.org> To: Christopher Farley <chris@northernbrewer.com> Cc: freebsd-stable@freebsd.org Subject: Re: VXLOCK interlock avoided Message-ID: <3C0BDD79.B3772ECB@geminix.org> References: <20011201155117.A61451@northernbrewer.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Christopher Farley wrote: > > I got a couple of weird messages the other day that I've never seen > before: > > VXLOCK interlock avoided > VXLOCK interlock avoided in vn_lock > > What are these? I've seen them as well. I run a recent 4.4-STABLE. In my case, according to my analysis, it is triggered by vlrureclaim() in "kern/vfs_subr.c". This routine was introduced a while ago by Matt Dillon, I think. It reclaims vnodes in order to curb the boundless growth of the vnode table once it reached a certain limit (depends on the amount of memory the system has). As soon as said table grows beyond that limit and vlrureclaim() kicks in the messages above start to appear. Very rarely, though. What I don't know yet is whether this is just a nuisance or whether something actually breaks in this case. I see it on a multi-processor machine. Maybe this is a contributing factor. Do you happen to run an SMP machine as well? Maybe Matt can shed some light on it. Uwe -- Uwe Doering <gemini@geminix.org> Berlin, Germany To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3C0BDD79.B3772ECB>