From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753018AbXC1MQS (ORCPT ); Wed, 28 Mar 2007 08:16:18 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753031AbXC1MQS (ORCPT ); Wed, 28 Mar 2007 08:16:18 -0400 Received: from ebiederm.dsl.xmission.com ([166.70.28.69]:55103 "EHLO ebiederm.dsl.xmission.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753018AbXC1MQR (ORCPT ); Wed, 28 Mar 2007 08:16:17 -0400 From: ebiederm@xmission.com (Eric W. Biederman) To: Michael Ellerman Cc: linux-pci@atrey.karlin.mff.cuni.cz, Greg Kroah-Hartman , "Eric W. Biederman" , "David S. Miller" , Benjamin Herrenschmidt , , Andrew Morton , Subject: Re: [PATCH 16/21] MSI: arch should connect the irq and the msi_desc desc when it's ready References: <20070322105344.312FEDDF70@ozlabs.org> Date: Wed, 28 Mar 2007 06:15:37 -0600 In-Reply-To: <20070322105344.312FEDDF70@ozlabs.org> (Michael Ellerman's message of "Thu, 22 Mar 2007 21:51:42 +1100") Message-ID: User-Agent: Gnus/5.110006 (No Gnus v0.6) Emacs/21.4 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Michael Ellerman writes: > set_irq_msi() currently connects an irq_desc to an msi_desc. The archs call > it at some point in their setup routine, and then the generic code sets up the > reverse mapping from the msi_desc back to the irq. > > set_irq_msi() should really do both connections, making it the one and only > call required to connect an irq with it's MSI desc. Hmm. This comment should be a little more forceful than a suggestion to make ti clear you have made that change somewhere in this patch. > It should be the arch code that calls set_irq_msi(), and it should do so > only once it's sure it's not going to fail the irq allocation. > > Given that there's no need for the arch to return the irq anymore, the return > value from the arch setup routine just becomes 0 for success and !0 for > failure. Call it negative for failure, or possibly nonzero. !0 == 1 Which isn't what you meant, and is confusing comparing it to the code. Ok. This is likely me limit of picking on your patches for tonight. > Index: msi-new/arch/ia64/kernel/msi_ia64.c > =================================================================== > --- msi-new.orig/arch/ia64/kernel/msi_ia64.c > +++ msi-new/arch/ia64/kernel/msi_ia64.c > @@ -74,7 +74,6 @@ int ia64_setup_msi_irq(struct pci_dev *p > if (irq < 0) > return irq; > > - set_irq_msi(irq, desc); > dest_phys_id = cpu_physical_id(first_cpu(cpu_online_map)); > vector = irq; > > @@ -91,11 +90,13 @@ int ia64_setup_msi_irq(struct pci_dev *p > MSI_DATA_DELIVERY_FIXED | > MSI_DATA_VECTOR(vector); > > + desc->irq = irq; Setting desc->irq just before set_irq_msi appears redundant here. > + set_irq_msi(irq, desc); > set_irq_chip_and_handler(irq, &ia64_msi_chip, handle_edge_irq); > get_irq_chip(irq)->mask(irq); > write_msi_msg(irq, &msg); > > - return irq; > + return 0; > } > > void ia64_teardown_msi_irq(unsigned int irq) Eric