Re: [PATCH v1 1/5] ARM: imx28: add basic dt support

["Lothar Waßmann" <LW@KARO-electronics.de>]

Hi,

Grant Likely writes:
> On Fri, 16 Mar 2012 11:01:35 +0800, Dong Aisheng <aisheng.dong@freescale.com> wrote:
> > On Thu, Mar 15, 2012 at 07:22:04PM +0800, Lothar Waßmann wrote:
> > > Hi,
> > > 
> > > Dong Aisheng writes:
> > > > On Thu, Mar 15, 2012 at 02:53:29PM +0800, Lothar Waßmann wrote:
> > > > > Hi,
> > > > > 
> > > > > Dong Aisheng writes:
> > > > > > On Wed, Mar 14, 2012 at 10:16:43PM +0800, s.hauer@pengutronix.de wrote:
> > > > > > > On Wed, Mar 14, 2012 at 08:45:23PM +0800, Dong Aisheng wrote:
> > > > > > > > On Wed, Mar 14, 2012 at 01:23:51AM +0800, Grant Likely wrote:
> > > > > [...]
> > > > > > > > But it seems this needs pass mac address to fec driver via platforom data which is
> > > > > > > > not friendly to dt.
> > > > > > > > 
> > > > > > > > Another way may be changing fec driver to get the fixed part of mac address(first
> > > > > > > > two bytes) from device tree and read the left dynamical part from otp which i'm not
> > > > > > > > sure is better enough.
> > > > > > > > 
> > > > > > > > BTW, filling with zeros seems not work since it's invalid mac address.
> > > > > > > 
> > > > > > > Yes, that's the idea of using this value...
> > > > > > > 
> > > > > > But comparing to use an invalid value, why not just do not define that
> > > > > > local-mac-address property?
> > > > > > 
> > > > > Because a MAC address is by definition a GLOBALLY UNIQUE IDENTIFIER
> > > > > which is contrary to coding a "valid" default value for it somewhere!
> > > > > The owner of a certain MAC address range (OUI) is responsible for
> > > > > ensuring the uniqueness. Thus only they may assign a specific MAC
> > > > > address to a specific device.
> > > > > 
> > > > Yes, you're correct.
> > > > So i propose to read the MAC address from MX28 OTP in fec driver instead of
> > > > define it in device tree in my last mail.
> > > > http://www.spinics.net/lists/arm-kernel/msg165040.html
> > > > Do you have any comment on that way?
> > > > 
> > > That patch sets the OUI to the Freescale owned MAC range. Thus only
> > > Freescale products may be able to use the driver.
> > > 
> > No.
> > My proposal is only set the fixed part(first two octets) in board dts file,
> > each board knows it's value, and read the left 4 octets from OCOTP dynamically.
> > Obviously the freeescale board dts file is only for FSL platforms.
> > Other platforms can define their fix part of MAC address in their dts file.
> > 
> > > Anyway there is no definite spec how the MAC address(es) are stored
> > > in the fuse map. Thus reading the MAC from there is more or less
> > > platform specific.
> > > 
> > It's just provide one more option since there are customers storing the MAC
> > in the fuse map.
> 
> That should be straight forward to support; have a property that
> specifies the method used for fetching/calculating the MAC.
> 
Executable code stored inside a DT blob? ;)

> > > Currently I'm setting up the MAC address for our TX28 from the fuses
> > > in the platform code passed via platform_data, but that will obviously
> > > not work with DT.
> > > 
> > Non-dt can also use my proposal, then you only need to pass the fixed part of
> > MAC via platfrom data, the left will be read by fec driver.
> > 
> > > The correct way would probably be to pass the MAC from the bootloader
> > > via a DT blob. But that would require all bootloaders to be updated
> > > first to support DTS. :(
> > > 
> > But uboot will face the same problem that can not define a valid MAC
> > in dts, did i undertand correctly?
> > 
> > > An intermediate solution could be using OF_DEV_AUXDATA to pass a
> > > platform_data struct that may contain a MAC address set up by the
> > > platform code.
> > > 
> > Yes, it's a way but i'm afraid will not get accepted by dt people.
> 
> The problem remains; where does the platform code get the MAC address
> from?  The kernel has to devise it from *somewhere* and the best place
> for that should be in the MAC driver itself.
> 
The platform code knows how and where the MAC is stored on a specific
platform. The driver has no business in knowing platform details like
this. Thus only platform code (or the bootloader) can provide the MAC
to the driver.

> I've got no problem with the idea of only specifying the OUI in the DT
> and the driver extracting the remaining 3 bytes from the hardware.
> 
> In principle, that's the design intent for DT systems anyway; it
> describes things which cannot be probed from the hardware, or tells
> the OS how the data is encoded in the hardware.
> 
If all bootloaders were aware of DT this wouldn't be a problem,
because they could place the correct MAC inside the DT blob right
away.
But in the mean time (AKA forever) we need some other solution.


Lothar Waßmann
-- 
___________________________________________________________

Ka-Ro electronics GmbH | Pascalstraße 22 | D - 52076 Aachen
Phone: +49 2408 1402-0 | Fax: +49 2408 1402-10
Geschäftsführer: Matthias Kaussen
Handelsregistereintrag: Amtsgericht Aachen, HRB 4996

www.karo-electronics.de | info@karo-electronics.de
___________________________________________________________