From mboxrd@z Thu Jan  1 00:00:00 1970
From: Linus Walleij <linus.walleij@linaro.org>
Subject: Re: [PATCH 1/6] gpio: make the gpiochip a real device
Date: Wed, 4 Nov 2015 11:48:47 +0100
Message-ID: <CACRpkda6Ad4SGW5H4ZDLn0VLZzq0LqLxSC7f2gcCBDNWH5FMnA@mail.gmail.com>
References: <1445502750-22672-1-git-send-email-linus.walleij@linaro.org>
	<1445502750-22672-2-git-send-email-linus.walleij@linaro.org>
	<20151102103116.GE8676@localhost>
	<CACRpkdaVu_1pAPd-7sUeGLEYymynA10WtuxJcnVPKZVYQOeruw@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Return-path: <linux-gpio-owner@vger.kernel.org>
Received: from mail-ob0-f174.google.com ([209.85.214.174]:33511 "EHLO
	mail-ob0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S932467AbbKDKss (ORCPT
	<rfc822;linux-gpio@vger.kernel.org>); Wed, 4 Nov 2015 05:48:48 -0500
Received: by obbww6 with SMTP id ww6so9903164obb.0
        for <linux-gpio@vger.kernel.org>; Wed, 04 Nov 2015 02:48:47 -0800 (PST)
In-Reply-To: <CACRpkdaVu_1pAPd-7sUeGLEYymynA10WtuxJcnVPKZVYQOeruw@mail.gmail.com>
Sender: linux-gpio-owner@vger.kernel.org
List-Id: linux-gpio@vger.kernel.org
To: Johan Hovold <johan@kernel.org>
Cc: "linux-gpio@vger.kernel.org" <linux-gpio@vger.kernel.org>, Alexandre Courbot <acourbot@nvidia.com>, Arnd Bergmann <arnd@arndb.de>, Michael Welling <mwelling@ieee.org>, Markus Pargmann <mpa@pengutronix.de>, Mark Brown <broonie@kernel.org>, Amit Kucheria <amit.kucheria@linaro.org>

On Tue, Nov 3, 2015 at 10:24 PM, Linus Walleij <linus.walleij@linaro.org> wrote:
> On Mon, Nov 2, 2015 at 11:31 AM, Johan Hovold <johan@kernel.org> wrote:
>> On Thu, Oct 22, 2015 at 10:32:25AM +0200, Linus Walleij wrote:
>
>>> +      * We memset the struct to zero to avoid reentrance issues.
>>> +      */
>>> +     memset(&chip->device, 0, sizeof(chip->device));
>>
>> This is an indication of a larger problem.
>>
>> First of all, you must never register the same device structure twice.
>>
>> And the larger problem is: With the current interface where a struct
>> gpio_chip is passed and registered, how would you prevent the device
>> from going away while in use?
>
> OK I see. What about the design pattern elsewhere to pass a
> gpiochip desc and have a new gpiochip_register() create the
> gpiochip and delete gpiochip_add()?
>
> That's the usual pattern in subsystems.

Thinking about it maybe it's simplest to just make ->dev a pointer
and kzalloc() it at gpiochip_add().

That should solve this.

Yours,
Linus Walleij