From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Thu, 8 Oct 2009 15:24:43 +0200 (CEST) From: Kjetil Oftedal To: David Miller Subject: Re: [RFC PATCH 00/12] Merge common OpenFirmware device tree code In-Reply-To: <20091007.213915.37481688.davem@davemloft.net> Message-ID: References: <20091007.140938.255120957.davem@davemloft.net> <20091007.213915.37481688.davem@davemloft.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: sammy@sammy.net, sfr@canb.auug.org.au, julian.calaby@gmail.com, devicetree-discuss@lists.ozlabs.org, sparclinux@vger.kernel.org, crn@netunix.com, microblaze-uclinux@itee.uq.edu.au, wmb@firmworks.com, linuxppc-dev@lists.ozlabs.org List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Wed, 7 Oct 2009, David Miller wrote: > From: Chris Newport > Date: Thu, 8 Oct 2009 02:29:25 +0100 (BST) > >> Sun4d has never had SMP support and > > Wrong. > >> this is apparantly problematic due to Cray interlectual property >> causing a lack of bus documentation. > > XBUS documentation is not available, but we fully know how to > program the SBUS interrupt controller and whatnot. It's all > there in the sun4d interrupt and SMP support and it did work > just fine at one point. > > Amusingly the SBUS interrupt stuff on sun4d is a very close > sibling to the IMAP/ICLR scheme used on sun4u. > The Sun4d SMP support exists, but is broken in 2.6. And the UP support is buggy. At least on my test-setup, the kernel is unable to load userland from scsi-drives. Dropping the sun4 32-bit SPARC is a bit counterproductive when considering the resent effort to unify sparc64 and sparc arch-branches ?