From owner-freebsd-stable Thu Jan 10 16:45:33 2002 Delivered-To: freebsd-stable@freebsd.org Received: from resnet.uoregon.edu (resnet.uoregon.edu [128.223.122.47]) by hub.freebsd.org (Postfix) with ESMTP id 7C30D37B400 for ; Thu, 10 Jan 2002 16:45:28 -0800 (PST) Received: from localhost (dwhite@localhost) by resnet.uoregon.edu (8.11.3/8.10.1) with ESMTP id g0B0XD998345; Thu, 10 Jan 2002 16:33:13 -0800 (PST) Date: Thu, 10 Jan 2002 16:33:13 -0800 (PST) From: Doug White To: Dmitry Morozovsky Cc: Josh Paetzel , Subject: Re: Reproducable system hangs in -STABLE In-Reply-To: <20020110211312.C15653-100000@woozle.rinet.ru> Message-ID: <20020110163043.S95050-100000@resnet.uoregon.edu> X-All-Your-Base: are belong to us MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Thu, 10 Jan 2002, Dmitry Morozovsky wrote: > Hmm... Is it related to SMP machines only or to uni-processor machines > too? I have no SMP machines with dnetc's around to test, but every UP > machine with dnetc starts it reniced -20, and there were no lockups at > all... I've mainly seen it on SMP. > Also, would you please give me a link to thread/article duscussing > "classic priority inversion proble"? The classic priority inversion problem is when you have two processes with differing priorities. The lower-priority process will lock a resource that a higher-priority process needs. Since the higher-priority process will block the lower-priority process from running, the lock is never released. Thus the deadlock. If you run dnetc at something other than nice 20 it should avoid the deadlock. Doug White | FreeBSD: The Power to Serve dwhite@resnet.uoregon.edu | www.FreeBSD.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message