Date: Tue, 17 May 2011 11:11:18 +0200 From: Marius Strobl <marius@alchemy.franken.de> To: Peter Jeremy <peterjeremy@acm.org> Cc: freebsd-sparc64@freebsd.org Subject: Re: "panic: cpu_initclocks: could not determine STICK frequency" booting V890 Message-ID: <20110517091118.GA75596@alchemy.franken.de> In-Reply-To: <20110517010554.GA75440@server.vk2pj.dyndns.org> References: <20110517010554.GA75440@server.vk2pj.dyndns.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, May 17, 2011 at 11:05:55AM +1000, Peter Jeremy wrote: > I've found an unused V890 (64GB RAM, 8 dual-core 1.35GHz US-IV CPUs, > 12 146GB FC disks) and thought I'd try FreeBSD on it (especially since > I had a copy of FreeBSD-8.2-RC2 lying around on CD). My initial try > failed with "panic: kmem_suballoc: bad status return of 3" but > limiting the RAM to 4GB with hw.physmem worked around that. > Unfortunately, it then panic's as above. That should no longer be necessary with recent versions of head and stable/8. > > Looking at the sources, the only recent change in that area of > sparc64/sparc64/tick.c is related to Serengeti-class machines (which > the V890 isn't). > > Does anyone have any suggestions? A verbose boot is available at > http://pastie.org/1913906 > I'd need an OFW device tree dump (`prtconf -Ppv` with Solaris) of that machine in order to check what's wrong with it and to decide how to solve it. All dumps of V890 I got so far show no difference in the layout of the stick timer property, i.e. the existing code should work just fine there. Marius
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20110517091118.GA75596>