From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nishanth Menon Subject: Re: [PATCH] OMAP4: Pandaboard: Fixing MMC card detect gpio line Date: Fri, 17 Dec 2010 12:06:47 -0600 Message-ID: <4D0BA6B7.3090406@ti.com> References: <1291723986-24658-1-git-send-email-raghuveer.murthy@ti.com> <4CFFB704.3030905@ti.com> <20101217180251.GQ5941@atomide.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from na3sys009aog102.obsmtp.com ([74.125.149.69]:44663 "EHLO na3sys009aog102.obsmtp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755317Ab0LQSGu (ORCPT ); Fri, 17 Dec 2010 13:06:50 -0500 Received: by yxm34 with SMTP id 34so803861yxm.25 for ; Fri, 17 Dec 2010 10:06:50 -0800 (PST) In-Reply-To: <20101217180251.GQ5941@atomide.com> Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: Tony Lindgren Cc: David Anders , "Murthy, Raghuveer" , "linux-omap@vger.kernel.org" Tony Lindgren had written, on 12/17/2010 12:02 PM, the following: > * Menon, Nishanth [101217 09:50]: >> On Wed, Dec 8, 2010 at 10:49, David Anders wrote: >>> On 12/07/2010 06:13 AM, Murthy, Raghuveer wrote: >>>> The .gpio_cd member of omap2_hsmmc_info is not initialized. This >>>> will default to zero. On Pandaboard this interferes with gpio line >>>> assigned for powering TFP410 DVI chip. >>>> >>>> This fix was missed out in the previous commit bf56f0a6668cd, from >>>> Nishanth Menon >>>> >>>> Signed-off-by: Kishore Kadiyala >>>> Signed-off-by: Raghuveer Murthy >>>> Acked-by: Nishanth Menon >>>> >>> Tested-by: David Anders >> Tony, >> this is a regression introduced in my commit for .37-rc1. any chance >> of this making to one of 37 rc cycles? I can re-post this patch >> including l-a if that is the path you'd like. > > Yes please, and also explain what exactly this breaks. > > If it just breaks MMC card detect with some DVI chip driver that's > not merged yet, then this would be in the "it never worked before". The regression is card detect ends up refering to a valid GPIO -> yeah as a result the upcoming DVI driver is broken without this patch. given that commit bf56f0a6668cd was intended to fix cd, it did not do it cleanly - hence my claim of regression. DVI in opinion is just a symptom of the wiring of GPIO0 on the board. ok with considering this a regression? -- Regards, Nishanth Menon