Return-Path: <foo00@h08.hostsharing.net>
Date: Tue, 2 Jan 2018 17:58:29 +0100
From: Lukas Wunner <lukas@wunner.de>
To: Marcel Holtmann <marcel@holtmann.org>
Cc: Linus Walleij <linus.walleij@linaro.org>,
	Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	Johan Hedberg <johan.hedberg@gmail.com>,
	Mika Westerberg <mika.westerberg@linux.intel.com>,
	Frederic Danis <frederic.danis.oss@gmail.com>,
	Loic Poulain <loic.poulain@linaro.org>,
	Hans de Goede <hdegoede@redhat.com>, Max Shavrick <mxms@me.com>,
	Leif Liddy <leif.liddy@gmail.com>,
	Daniel Roschka <danielroschka@phoenitydawn.de>,
	Ronald Tschalaer <ronald@innovation.ch>,
	"Peter Y. Chuang" <peteryuchuang@gmail.com>,
	linux-bluetooth@vger.kernel.org
Subject: Re: [PATCH 3/3] Bluetooth: Avoid WARN splat due to missing GPIOLIB
Message-ID: <20180102165829.GA6735@wunner.de>
References: <9ef8be4ef80c38b693e5755c738a88de9c907944.1514143015.git.lukas@wunner.de>
 <1514450477.7000.302.camel@linux.intel.com>
 <20171228091805.GA1559@wunner.de>
 <1514453177.7000.320.camel@linux.intel.com>
 <CACRpkdb=Wj9dEfu7YEfQCA94MdBNQmp5UwHOgRsTZX=bf5d-Hw@mail.gmail.com>
 <1514464858.7000.334.camel@linux.intel.com>
 <CACRpkdYRojbfCC_b9vO0MquvLeenru0rrYNv7WzMg3d3b=iF+g@mail.gmail.com>
 <20171229095135.GA25623@wunner.de>
 <CACRpkda1uzW2A=13tobpNrNNgTmv-ri6kb628TYS7rv3xPsx4A@mail.gmail.com>
 <0668F24B-610E-4239-8470-DEB96D533D96@holtmann.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
In-Reply-To: <0668F24B-610E-4239-8470-DEB96D533D96@holtmann.org>
List-ID: <linux-bluetooth.vger.kernel.org>

On Tue, Jan 02, 2018 at 04:27:02PM +0100, Marcel Holtmann wrote:
> >> By the way, what is the rationale for this rule that consumers shall
> >> depend on rather than select GPIOLIB?
> > 
> > I think consumers should depend on GPIOLIB and producers
> > should select GPIOLIB.
> 
> so what do we do here? Should I revert the patch or can I get an updated
> patch that uses the agreed upon select vs depends.

I've prepared this branch which still needs to be tested & debugged,
it contains a patch to change the "select" to a "depends on":

https://github.com/l1k/linux/commits/hci_bcm_v2/
https://github.com/l1k/linux/commit/71aa06c610d0

If you prefer reverting 27378f4c1b92 or rebase your branch and drop it,
I'll be happy to provide a rectified version instead (basically
27378f4c1b92 and 71aa06c610d0 squashed together).  I was assuming that
bluetooth-next/master is a non-rebasing branch, hence prepared
71aa06c610d0 as a fix of the existing commit.

Thanks,

Lukas