From owner-freebsd-arm@freebsd.org Thu May 5 16:18:02 2016 Return-Path: Delivered-To: freebsd-arm@mailman.ysv.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by mailman.ysv.freebsd.org (Postfix) with ESMTP id AE231B2EC92 for ; Thu, 5 May 2016 16:18:02 +0000 (UTC) (envelope-from fbsd@www.zefox.net) Received: from www.zefox.net (www.zefox.net [69.239.235.194]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "www.zefox.org", Issuer "www.zefox.org" (not verified)) by mx1.freebsd.org (Postfix) with ESMTPS id 7213D1332 for ; Thu, 5 May 2016 16:18:02 +0000 (UTC) (envelope-from fbsd@www.zefox.net) Received: from www.zefox.net (localhost [127.0.0.1]) by www.zefox.net (8.15.2/8.15.2) with ESMTPS id u45GI5qQ005688 (version=TLSv1.2 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO); Thu, 5 May 2016 16:18:06 GMT (envelope-from fbsd@www.zefox.net) Received: (from fbsd@localhost) by www.zefox.net (8.15.2/8.15.2/Submit) id u45GI4Ju005687; Thu, 5 May 2016 09:18:04 -0700 (PDT) (envelope-from fbsd) Date: Thu, 5 May 2016 09:18:04 -0700 From: bob prohaska To: Svatopluk Kraus Cc: "freebsd-arm@freebsd.org" Subject: Re: Another reboot hang Message-ID: <20160505161804.GS71221@www.zefox.net> References: <20160426183030.GN71221@www.zefox.net> <20160426204016.GO71221@www.zefox.net> <20160426214309.GP71221@www.zefox.net> <20160429185209.GQ71221@www.zefox.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.24 (2015-08-30) X-BeenThere: freebsd-arm@freebsd.org X-Mailman-Version: 2.1.22 Precedence: list List-Id: "Porting FreeBSD to ARM processors." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 05 May 2016 16:18:02 -0000 Hi Svata, On Thu, May 05, 2016 at 04:38:36PM +0200, Svatopluk Kraus wrote: > On Fri, Apr 29, 2016 at 8:52 PM, bob prohaska wrote: > > > > So far the reboot hang hasn't repeated in half a dozen or so attempts..... > > not sure if that's good or bad. > > > > In the meantime the system ran stress2 (old version) for 12 hours 49 minutes, > > the previous record was around ten hours. Interestingly, the crash did _not_ > > report "translation fault (L2)", but rather > > "panic: pmap_growkernel: no memory to grow kernel". > > > This panic is saying that there is no free page to alloc but the pmap > really needs it. I have no experience with such thing, but total > exhaustion of free page pool may happen probably. And when it happens > in some critical place, kernel has no possibility, but panic. BTW, it > would be nice to see output of 'show uma" and 'show malloc' from ddb. I'll try to capture the output of "show uma" and "show malloc" the next time the machine crashes. So far, no recurrence of the "panic: pmap_grow....." crash, just the usual "translation fault L2...." variety or silent hang. In the "panic: pmap_grow...." instance (April 29) it appears I mistakenly captured the wrong top output, from an earlier crash. Can a date stamp be added to the top output timestamp? The man page doesn't seem to say. bob prohaska