From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932361Ab2CSKxk (ORCPT ); Mon, 19 Mar 2012 06:53:40 -0400 Received: from mail1.bemta7.messagelabs.com ([216.82.255.50]:35592 "EHLO mail1.bemta7.messagelabs.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932177Ab2CSKxW (ORCPT ); Mon, 19 Mar 2012 06:53:22 -0400 X-Greylist: delayed 390 seconds by postgrey-1.27 at vger.kernel.org; Mon, 19 Mar 2012 06:53:22 EDT X-Env-Sender: darryl.degraff@licor.com X-Msg-Ref: server-7.tower-172.messagelabs.com!1332154009!2153835!1 X-Originating-IP: [208.82.105.217] X-StarScan-Version: 6.5.7; banners=-,-,- X-VirusChecked: Checked Message-ID: <4F670E6A.9060403@licor.com> Date: Mon, 19 Mar 2012 05:46:02 -0500 From: Darryl User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:10.0.2) Gecko/20120216 Thunderbird/10.0.2 MIME-Version: 1.0 To: , CC: , , Subject: arm: davinci: USB ethernet Content-Type: text/plain; charset="ISO-8859-1"; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org We are currently using a DA850 eval board but will eventually use a custom board to interface host system to davinci using ethernet over USB. We are using 2.6.37 with real-time patches, but a cursory look (have not been able to build yet) at 3.3-rc7 suggests that the issues noted below still exist. There are currently two issues: - Documentation/usb/hotplug.txt suggests that the drivers should be built as modules so that such things as loading drivers and, in this case, bringing up the interface, can ge moderated in user-space. However, though the code can be build modular, the drivers are never loaded. I suspect that one or more of the several relevant configs (I see at least these: USB_MUSB_HDRC, USB_MUSB_DA8XX_GLUE, USB_GADGET, USB_ETH, USB_GPIO_VBUS, NOP_USB_XCEIV, USB_ETH_RNDIS, USB_OTG_UTILS) are required to be built in, but cannot seem to determine the correct ones. - Even when compiling all the driver code built-in, neither the plug-in nor plug-out event is propagated to user space, again, so that the interface can be brought up. I am happy enough to build-in the driver code, but I still need the ability to bring up the interface. I know that I can use a call to call_usermodehelper to get to use-space, but I'd still be faced with where that call would work best. So, I'm wondering if you have suggestions regarding how to proceed. Thanks, Darryl