From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752617Ab2I2Exr (ORCPT ); Sat, 29 Sep 2012 00:53:47 -0400 Received: from smtp-vbr14.xs4all.nl ([194.109.24.34]:3722 "EHLO smtp-vbr14.xs4all.nl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751895Ab2I2Exp (ORCPT ); Sat, 29 Sep 2012 00:53:45 -0400 Date: Sat, 29 Sep 2012 06:53:42 +0200 From: jurriaan_lkml@rivierenland.xs4all.nl To: Kernel development list Subject: Etron usb3 device not detected on Qnap TS 119 P II Message-ID: <20120929045342.GA17904@qnap.fritz.box> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org I have a situation where a single-bay NAS, the Qnap TS 119 P II in the newest revision has 2 usb3 ports on the back. When using the 3.6-rc7 kernel (and earlier ones), these ports don't exist. There is no output with lsusb, and the device doesn't show up with lspci or dmesg. The directory /sys/devices/pci is empty. Qnap's own kernel, a patched variety of 2.6.33.2, adds a separate etcxhci-pci driver that does find the hardware, with vendorid PCI_VENDOR_ID_ETRON 0x1b6f and PCI_DEVICE_ID_ASROCK_P67 0x7023. I did notice these ids are in the 3.6-rc7 xhci-pci.c file, so the new kernel should support the hardware. Contact with Sarah Sharp, the author of xhci-pci.c, pointed out that Qnap likely did something in the pci-enumeration code to get the device activated. This is a armiv5tel device, running a Marvell Feroceon cpu. I have a serial console cable and can run experimental kernels/patches. What I would dearly like is some pointers about where to turn on debugging and what to post here to get find out where things go wrong. I'm posting it here, since I think it might be more of a general pci device problem than an usb problem. Thanks, Jurriaan