From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [PATCH 2/2] [libata] sata_mv: Support integrated controllers Date: Tue, 18 Dec 2007 17:02:01 -0500 Message-ID: <47684359.9030706@garzik.org> References: <11966101971581-git-send-email-saeed.bishara@gmail.com> <1196610199229-git-send-email-saeed.bishara@gmail.com> <11966102001423-git-send-email-saeed.bishara@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from srv5.dvmed.net ([207.36.208.214]:45132 "EHLO mail.dvmed.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754478AbXLRWCD (ORCPT ); Tue, 18 Dec 2007 17:02:03 -0500 In-Reply-To: <11966102001423-git-send-email-saeed.bishara@gmail.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: saeed.bishara@gmail.com Cc: linux-ide@vger.kernel.org, Saeed Bishara saeed.bishara@gmail.com wrote: > From: Saeed Bishara > > Marvell's Orion SoC includes SATA controllers based on Marvell's > PCI-to-SATA 88SX controllers. The integrated SATA unit is connected > directly to the internal bus of the Orion SoC, and not via PCI. > This patch extends the libATA sata_mv driver to support those > controllers. > > Signed-off-by: Saeed Bishara Overall, looks pretty good. My main comment is that several changes, like - void (*reset_bus)(struct pci_dev *pdev, void __iomem *mmio); + void (*reset_bus)(struct ata_host *host, void __iomem *mmio); and - if (unlikely(irq_stat & PCI_ERR)) { + if (unlikely(irq_stat & PCI_ERR) && HAS_PCI(host)) { belong in patch #1, the PCI isolation patch, rather than this patch implementing support for integrated controllers. Finally, don't hardcode a 32-bit DMA mask in your platform_device setup, pass that mask in via mv_sata_platform_data. Jeff