To: "Jonas Gorski" <jonas.gorski@gmail.com>
cc: "Arnd Bergmann" <arnd@arndb.de>,
	"zajec5@gmail.com" <zajec5@gmail.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-wireless@vger.kernel.org" <linux-wireless@vger.kernel.org>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>,
	"b43-dev@lists.infradead.org" <b43-dev@lists.infradead.org>,
	"George Kashperko" <george@znau.edu.ua>,
	"Hauke Mehrtens" <hauke@hauke-m.de>,
	"Russell King" <rmk@arm.linux.org.uk>,
	"Larry Finger" <Larry.Finger@lwfinger.net>,
	"Andy Botting" <andy@andybotting.com>, "Greg KH" <greg@kroah.com>,
	"Michael Buesch" <m@bues.ch>
Subject: Re: [PATCH] drivers: brcmaxi: provide amba axi functionality in
 separate module
References: <1303331669-31293-1-git-send-email-arend@broadcom.com>
 <201104211612.49805.arnd@arndb.de> <op.vuallvn63ri7v4@arend-laptop>
 <BANLkTinUWNvkigqkOQeYsyaAKnVXOCji8w@mail.gmail.com>
Date: Sat, 23 Apr 2011 17:07:19 +0200
MIME-Version: 1.0
From: "Arend van Spriel" <arend@broadcom.com>
Message-ID: <op.vuecahou3ri7v4@arend-laptop>
In-Reply-To: <BANLkTinUWNvkigqkOQeYsyaAKnVXOCji8w@mail.gmail.com>
Content-Type: text/plain;
 charset=utf-8;
 format=flowed;
 delsp=yes
Sender: linux-wireless-owner@vger.kernel.org

On Sat, 23 Apr 2011 14:33:21 +0200, Jonas Gorski <jonas.gorski@gmail.com>  
wrote:

> On 21 April 2011 16:38, Arend van Spriel <arend@broadcom.com> wrote:
>> Would it be possible to make chipcommon driver optional (not doing the
>> initialization)?
>
> This would need to be done on a per-device/bus basis, at least for
> embedded. Consider the following setup (which is quite common for dual
> band routers):
>
> BCM4718 (bus A)
   +- MIPS74k
> +- Common Core <- provides flash write access, GPIOs, watchdog, ...
> +- 802.11 Core <- for 2.4Ghz wifi
> +- PCIe Core
>   + BCM43224 (bus B)
>     +- Common Core
>     +- 802.11 Core <- for 5Ghz wifi
>
> (I omitted any cores not relevant for the example)

The MIPS may be relevant as well ;-)

In the example above I would expect two axi bus driver instances for bus A  
and bus B.

> So eventually you want to able to drive both 802.11 cores, but can't
> exclusively claim both common cores.

I would expect to be probed twice. One call for 2.4GHz 802.11 core  
referencing to bus A and one call for 5GHz 802.11 core referencing to bus  
B.

Regarding chipcommon I agree that it requires one init sequence per  
device. I would just like to have the option to provide a custom  
initialization function for chip common (and possibly pcie) somehow.

Gr. AvS
-- 
"The world is indeed comic, but the joke is on mankind." — H.P. Lovecraft