From owner-freebsd-alpha Thu Sep 21 16:23:15 2000 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.du.gtn.com (mail.du.gtn.com [194.77.9.57]) by hub.freebsd.org (Postfix) with ESMTP id C0B7737B42C; Thu, 21 Sep 2000 16:23:06 -0700 (PDT) Received: from mail.cicely.de (cicely.de [194.231.9.142]) by mail.du.gtn.com (8.11.0.Beta3/8.11.0.Beta3) with ESMTP id e8LNN2423565 (using TLSv1/SSLv3 with cipher EDH-RSA-DES-CBC3-SHA (168 bits) verified OK); Fri, 22 Sep 2000 01:23:04 +0200 (MET DST) Received: from cicely5.cicely.de (cicely5.cicely.de [fec0::104:200:92ff:fe9b:20e7]) by mail.cicely.de (8.11.0.Beta1/8.11.0.Beta1) with ESMTP id e8LNNGT09365; Fri, 22 Sep 2000 01:23:16 +0200 (CEST) Received: (from ticso@localhost) by cicely5.cicely.de (8.11.0/8.9.2) id e8LNNCM58742; Fri, 22 Sep 2000 01:23:12 +0200 (CEST) (envelope-from ticso) Date: Fri, 22 Sep 2000 01:23:11 +0200 From: Bernd Walter To: John Baldwin Cc: alpha@FreeBSD.ORG Subject: Re: Ithreads kernel weirdness Message-ID: <20000922012311.A58579@cicely5.cicely.de> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from jhb@FreeBSD.ORG on Tue, Sep 19, 2000 at 11:31:08PM -0700 Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Sep 19, 2000 at 11:31:08PM -0700, John Baldwin wrote: > I suppose many of you are wondering why there haven't been any > recent commits of ithreads for the alpha as promised. The reason > is that I can't get a plain kernel to boot. My test kernel with > several debugging kernel options added in the config file boots > fine. If I remove some of those debugging options, the kernel > will boot sometimes, and hang others. If I remove all the > debugging options, the kernel always hangs. When it hangs, the > kernel is always hanging just after the 'timecounter' device > probe printf's. Normally after this I get the acd0 probe > message. Booting verbose ends with two messages from > ata-master0, so I'm assuming it is during ata's probe of its > busses, which use the await()/asleep() code. At this point in > time, my guess is that this another weird manifestation of the > possible BSS corruption bug that a few people are seeing > (like static hints not working because hints_loaded is getting > trounced). I would have committed this 2 days ago, but I can't > get a normal kernel to boot. It might also be weirdness with > spl()'s (although it used to work fine, so no reason that it > doesn't work now, to be honest). Regardless, my next step is > to make the softinterrupt code mostly MI which will move SWI's > to the softinterrupt thread as a side effect. Then we should > be able to kill spl's (I hope), which may clear up some problems > as well. The current ithreads patchset as always is at: > http://www.FreeBSD.org/~jhb/patches/alpha.ithreads.patch. Pressing the halt key on the hanging system showed me that it is always in hardclock() -- B.Walter COSMO-Project http://www.cosmo-project.de ticso@cicely.de Usergroup info@cosmo-project.de To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message