From mboxrd@z Thu Jan  1 00:00:00 1970
From: tony@atomide.com (Tony Lindgren)
Date: Fri, 2 Mar 2012 09:06:01 -0800
Subject: [PATCH 2/4] mmc: omap_hsmmc: Use gpio_find_by_chip_name() for
 omap_hsmmc_gpio_init()
In-Reply-To: <4F5060B0.7010703@ti.com>
References: <20120301185044.29210.44521.stgit@kaulin.local>
 <20120301185528.29210.85854.stgit@kaulin.local>
 <4F5060B0.7010703@ti.com>
Message-ID: <20120302170600.GG18901@atomide.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

* Rajendra Nayak <rnayak@ti.com> [120301 21:23]:
> On Friday 02 March 2012 12:25 AM, Tony Lindgren wrote:
> >Use gpio_find_by_chip_name() to find the GPIO pins as they can
> >be dynamically allocated on various gpio_chips.
> >
> >Note that we don't want to touch the platform data as it can
> >now specify the GPIO offset on a named gpio_chip.
> >
> >This removes the need to use callbacks to set the GPIO pins
> >in platform data.
> 
> While one of the reasons for those callbacks was to set the GPIO
> pins in platform data, I guess the other and more important one
> was to make sure the init sequencing between twl4030-gpio and the
> mmc device depending on it was done rightly. Doesn't this patch now
> leave the sequencing to work by luck (in the absence of something
> like deferred probe being in place) like is the case of twl6030 and
> mmc init sequence on OMAP4 already?

Nope :) The difference is that  we can exit and produce a sensible
error to the user about the particular gpio_chip not being available.

Regards,

Tony