[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20130408101530.GC9243@opensource.wolfsonmicro.com>
Date: Mon, 8 Apr 2013 11:15:30 +0100
From: Mark Brown <broonie@...nsource.wolfsonmicro.com>
To: Girish KS <girishks2000@...il.com>
Cc: spi-devel-general@...ts.sourceforge.net,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>,
Grant Likely <grant.likely@...retlab.ca>,
Tomasz Figa <t.figa@...sung.com>
Subject: Re: [PATCH V3 4/5] spi: s3c64xx: Added provision for dedicated cs pin
On Mon, Apr 08, 2013 at 03:21:03PM +0530, Girish KS wrote:
> On Mon, Apr 1, 2013 at 6:27 PM, Mark Brown
> > It's also a bit odd that we end up checking cs_gpio and then using line
> > in the code, it'd be more idiomatic if cs_gpio were the GPIO number.
> In the original driver it was assumed that the cs line is always a gpio pin.
> But the current controller that i am working on has no gpio pin for cs
> selection.
> All the lines to the device are internally connected. There is no
> option to select
> the cs signal. So cs-gpio property parsing has to skipped for this
> controller, that means
> cs_gpio cannot be a GPIO number. If it has to be a number then it has
> to be < 0 to say
> it is not gpio. Any >= 0 number implies it is a valid gpio (in reality
> for this controller it is not.)
Two options here, one is to just assume nobody will use GPIO 0 and the
other is to set the number appopriately during probe so that only probe
needs to worry about the issue.
Download attachment "signature.asc" of type "application/pgp-signature" (837 bytes)
Powered by blists - more mailing lists