From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gate.crashing.org (gate.crashing.org [63.228.1.57]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTP id 5012067B18 for ; Fri, 30 Jun 2006 17:51:00 +1000 (EST) Received: from localhost.localdomain (localhost [127.0.0.1]) by gate.crashing.org (8.12.8/8.12.8) with ESMTP id k5U7vIiH012073 for ; Fri, 30 Jun 2006 02:57:19 -0500 Subject: new irq works status (#2) From: Benjamin Herrenschmidt To: linuxppc-dev@ozlabs.org Content-Type: text/plain Date: Fri, 30 Jun 2006 17:50:55 +1000 Message-Id: <1151653855.13194.9.camel@localhost.localdomain> Mime-Version: 1.0 List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , I uploaded a new version at http://gate.crashing.org/~benh/irq-WIP/ This one applies on top of a linus git checkout from today and thus should also apply on top of a recent powerpc.git tree. It contains various fixes & cleanups over the previous ones, significant fixes in the mpic code (mpic based pseries work now) and powermac 64 bits is ported over. Remaining bits are: - PowerMac 32 bits. There is a bit of work to replace the old workarounds for broken device-trees here, those used to be done by changing device_node->intrs which is now gone - Maple (should be trivial, +/- a small nit for some SLOF version) - CHRP (should be really trivial :) Just need to be careful of Pegasos lack of a valid interrupt-tree and properly default to the PCI_INTERRUPT_LINE config space register in that case, which shall match 8259 interrupts - embedded6xx/* platforms. I'll leave the pleasure of fixing these to the maintainers for now unless I decide to beat them to it, but since I'm not too familiar, I'd rather have them do the job. It should be fairly trivial to just make them work. The next step though is to make them _use_ the new stuff which will probably involve device-tree updates. I expect to have PowerMac32 and possibly CHRP fixed tomorrow and will proceed to the merge asap, leaving Maple and embedded6xx as "to be fixed before 2.6.18 is finished". I'd appreciate review and comments. Cheers, Ben.