Date: Tue, 19 Apr 2011 22:16:53 +0100
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
To: Anton Vorontsov <cbouatmailru@gmail.com>
Cc: Grant Likely <grant.likely@secretlab.ca>,
        Jean Delvare <khali@linux-fr.org>, LKML <linux-kernel@vger.kernel.org>,
        Jamie Iles <jamie@jamieiles.com>,
        Mark Brown <broonie@opensource.wolfsonmicro.com>, arnd@arndb.de
Subject: Re: [PATCH] gpio: New driver for the Intel 82801 (ICH) GPIO pins
Message-ID: <20110419221653.585a132e@lxorguk.ukuu.org.uk>
In-Reply-To: <20110419203031.GA6494@oksana.dev.rtsoft.ru>
References: <20110419145303.111aead7@endymion.delvare>
	<BANLkTinbO_p=sKtFJtjgpo4yrEVopq9vrw@mail.gmail.com>
	<20110419155406.1da0e7fb@lxorguk.ukuu.org.uk>
	<BANLkTi=tzXyozCck9H03-2RErGsvSxrv1Q@mail.gmail.com>
	<20110419165746.2857c56f@lxorguk.ukuu.org.uk>
	<20110419164048.GA30346@oksana.dev.rtsoft.ru>
	<20110419180829.56daccd0@lxorguk.ukuu.org.uk>
	<20110419203031.GA6494@oksana.dev.rtsoft.ru>
Face: iVBORw0KGgoAAAANSUhEUgAAADAAAAAwBAMAAAClLOS0AAAAFVBMVEWysKsSBQMIAwIZCwj///8wIhxoRDXH9QHCAAABeUlEQVQ4jaXTvW7DIBAAYCQTzz2hdq+rdg494ZmBeE5KYHZjm/d/hJ6NfzBJpp5kRb5PHJwvMPMk2L9As5Y9AmYRBL+HAyJKeOU5aHRhsAAvORQ+UEgAvgddj/lwAXndw2laEDqA4x6KEBhjYRCg9tBFCOuJFxg2OKegbWjbsRTk8PPhKPD7HcRxB7cqhgBRp9Dcqs+B8v4CQvFdqeot3Kov6hBUn0AJitrzY+sgUuiA8i0r7+B3AfqKcN6t8M6HtqQ+AOoELCikgQSbgabKaJW3kn5lBs47JSGDhhLKDUh1UMipwwinMYPTBuIBjEclSaGZUk9hDlTb5sUTYN2SFFQuPe4Gox1X0FZOufjgBiV1Vls7b+GvK3SU4wfmcGo9rPPQzgIabfj4TYQo15k3bTHX9RIw/kniir5YbtJF4jkFG+dsDK1IgE413zAthU/vR2HVMmFUPIHTvF6jWCpFaGw/A3qWgnbxpSm9MSmY5b3pM1gvNc/gQfwBsGwF0VCtxZgAAAAASUVORK5CYII=
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1686
Lines: 38

On Wed, 20 Apr 2011 00:30:31 +0400
Anton Vorontsov <cbouatmailru@gmail.com> wrote:

> On Tue, Apr 19, 2011 at 06:08:29PM +0100, Alan Cox wrote:
> > > How about exporting some bus/device-type neutral probe function, like
> > > we do in drivers/ata/pata_platform.c ?
> > 
> > Not sure you want to be using resources as the basic currency, nor do you
> > want your generic code doing the request_region stuff. That was a nasty
> > mistake we made in the IDE code.
> > 
> > Maybe the platform code should do that bit, but the generic stuff not - in
> > the PCI case the caller will have done pci_request_* itself and chances
> > are several of the resources are the same pci device resource and
> > different offsets so doing request_resource on them will blow up.
> 
> Oh, right. How about this:
> 
> int __devinit bgpio_probe(struct device *dev,
>                           unsigned long sz,
>                           void __iomem *dat,
>                           void __iomem *set,
>                           void __iomem *clr,
>                           void __iomem *dirout,
>                           void __iomem *dirin,
>                           bool big_endian);
> 
> I.e. PCI driver can now map the whole area with a single ioremap(),
> and then call bgpio_probe(), which will provide accessors and
> common MMIO GPIO registration stuff.

Looks sane and providing its all using iomap/ioread/iowrite it covers all
cases nicely.
--
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/