Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1758076Ab3CNPbZ (ORCPT ); Thu, 14 Mar 2013 11:31:25 -0400 Received: from mail-lb0-f171.google.com ([209.85.217.171]:56021 "EHLO mail-lb0-f171.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757582Ab3CNPbW convert rfc822-to-8bit (ORCPT ); Thu, 14 Mar 2013 11:31:22 -0400 MIME-Version: 1.0 In-Reply-To: References: <1363219179-14900-1-git-send-email-agraf@suse.de> <6F88F86A-8C8E-4C2E-BD15-14C788FC6E48@suse.de> Date: Thu, 14 Mar 2013 08:31:20 -0700 X-Google-Sender-Auth: 8HfgGVaemzGgNmJqe-fnhK8RwI4 Message-ID: Subject: Re: [PATCH v2] USB: ehci-s5p: Fix phy reset From: Doug Anderson To: Thomas Abraham Cc: Alexander Graf , "linux-kernel@vger.kernel.org" , linux-usb@vger.kernel.org, linux-samsung-soc@vger.kernel.org, dmueller@suse.de, Vivek Gautam , Jingoo Han , Alan Stern , Kukjin Kim , Felipe Balbi , Greg Kroah-Hartman , Vincent Palatin , jwerner@chromium.org Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7BIT Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1535 Lines: 30 Hi, On Thu, Mar 14, 2013 at 7:58 AM, Thomas Abraham wrote: >> I can see your point, but as I mentioned earlier there seems to be some timing issue here. By simply doing the reset a few ms earlier (in the first probe, before the driver detects that it needs to defer probing), I already can't find the hub on the bus later. >> >> So I'm assuming that the same thing would also happen if I put it even earlier in machine init. > > True, I missed that point. The usb hub connected over hsic interface, > after power-on-reset, might have initiated the 'connect' state on > seeing the idle condition on the bus and since the host/phy controller > is not ready yet, the connect might have failed. > > So the correct sequence would be, after the usb host controller and > the phy controllers are initialized, the 'reset' pin on the on-board > usb hub should be asserted. Upon releasing that reset, the usb hub > would initiate the 'connect' state on the HSIC bus. I think we ran into similar issues on one of our boards that had a hub attached to the HSIC port. We had to make our fix in a version of the code that's nowhere near what landed upstream (so our change is not relevant), but we're definitely interested in whatever solution you come up with. :) -Doug -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/