From mboxrd@z Thu Jan  1 00:00:00 1970
From: b-cousson@ti.com (Cousson, Benoit)
Date: Mon, 28 Feb 2011 15:55:00 +0100
Subject: [PATCH v10 05/18] OMAP2, 3 DSS2 Change driver name to	omap_display
In-Reply-To: <20110228140617.GA2176@n2100.arm.linux.org.uk>
References: <1295850125-21405-1-git-send-email-sumit.semwal@ti.com>
	<1295850125-21405-6-git-send-email-sumit.semwal@ti.com>
	<1298539643.16119.54.camel@deskari>
	<1298876015.2096.36.camel@deskari> <4D6B4A1A.9000304@ti.com>
	<1298877549.2096.47.camel@deskari> <4D6B88BB.5010200@ti.com>
	<1298895026.9809.27.camel@deskari>
	<20110228121321.GB30781@n2100.arm.linux.org.uk>
	<4D6BA558.5050700@ti.com>
	<20110228140617.GA2176@n2100.arm.linux.org.uk>
Message-ID: <4D6BB744.1010205@ti.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On 2/28/2011 3:06 PM, Russell King - ARM Linux wrote:
> On Mon, Feb 28, 2011 at 02:38:32PM +0100, Cousson, Benoit wrote:
>> On 2/28/2011 1:13 PM, Russell King - ARM Linux wrote:
>>> On Mon, Feb 28, 2011 at 02:10:26PM +0200, Tomi Valkeinen wrote:
>>>> On Mon, 2011-02-28 at 05:36 -0600, Cousson, Benoit wrote:
>>>>> Cannot you use a device hierarchy then to do that?
>>>>>     omap_dss/core
>>>>>     omap_dss/dsi
>>>>>     omap_dss/venc
>>>>>
>>>>> This is moreover the way the HW is done.
>>>>
>>>> Hmm, how would that work? The devices are platform devices, and they
>>>> have a unique global name, which is used to match the driver for the
>>>> device.
>>>
>>> And not forgetting that '/' is not permitted in device or driver names.
>>
>> I was not really considering naming the device with a '/'. The idea is
>> that the devices will not longer be flat platform devices but something
>> like dss devices that will have have a omap_dss parent.
>>
>> I do not know if that makes sense to store some kind of hierarchy in the
>> device model, but encoding the hierarchy in the name like it is proposed
>> looks like a hack to me.
> 
> Try:
> 
> tree -d /sys/devices/pci0000\:00
> 
> on your PC, taking special note of how pci-to-pci bridges are handled.

Here it is, but I'm not sure to understand how it is really done :-(

/sys/devices/pci0000:00
??? 0000:00:00.0
?   ??? firmware_node -> ../../LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/PNP0C02:00
?   ??? power
?   ??? subsystem -> ../../../bus/pci
??? 0000:00:02.0
?   ??? 0000:00:02.0:pcie01
?   ?   ??? power
?   ?   ??? subsystem -> ../../../../bus/pci_express
?   ??? 0000:00:02.0:pcie02
?   ?   ??? driver -> ../../../../bus/pci_express/drivers/aer
?   ?   ??? power
?   ?   ??? subsystem -> ../../../../bus/pci_express
?   ??? 0000:06:00.0
?   ?   ??? 0000:06:00.0:pcie12
?   ?   ?   ??? power
?   ?   ?   ??? subsystem -> ../../../../../bus/pci_express
?   ?   ??? 0000:07:00.0
?   ?   ?   ??? 0000:07:00.0:pcie22
?   ?   ?   ?   ??? power
?   ?   ?   ?   ??? subsystem -> ../../../../../../bus/pci_express
?   ?   ?   ??? 0000:08:00.0
?   ?   ?   ?   ??? 0000:09:00.0
?   ?   ?   ?   ?   ??? driver -> ../../../../../../../bus/pci/drivers/bnx2
?   ?   ?   ?   ?   ??? net
?   ?   ?   ?   ?   ?   ??? eth1
?   ?   ?   ?   ?   ?       ??? device -> ../../../0000:09:00.0
?   ?   ?   ?   ?   ?       ??? power
?   ?   ?   ?   ?   ?       ??? statistics
?   ?   ?   ?   ?   ?       ??? subsystem -> ../../../../../../../../../class/net
?   ?   ?   ?   ?   ??? power
?   ?   ?   ?   ?   ??? subsystem -> ../../../../../../../bus/pci
?   ?   ?   ?   ??? pci_bus
?   ?   ?   ?   ?   ??? 0000:09
?   ?   ?   ?   ?       ??? device -> ../../../0000:08:00.0
?   ?   ?   ?   ?       ??? power
?   ?   ?   ?   ?       ??? subsystem -> ../../../../../../../../class/pci_bus
?   ?   ?   ?   ??? power
?   ?   ?   ?   ??? subsystem -> ../../../../../../bus/pci
?   ?   ?   ??? driver -> ../../../../../bus/pci/drivers/pcieport
?   ?   ?   ??? pci_bus
?   ?   ?   ?   ??? 0000:08
?   ?   ?   ?       ??? device -> ../../../0000:07:00.0
?   ?   ?   ?       ??? power
?   ?   ?   ?       ??? subsystem -> ../../../../../../../class/pci_bus
?   ?   ?   ??? power
?   ?   ?   ??? subsystem -> ../../../../../bus/pci
...