Device tree aware EMAC driver

Device tree aware EMAC driver

[David Gibson]

This is BenH's rewritten, device tree aware driver for the IBM/AMCC
EMAC, ethernet MAC controller ASIC found on various 4xx CPUs and also
the Axon bridge.
	- Updated to build on current kernels
	- Removed device_type matching requirement on MAL and ZMII, to
match current best practice of omitting device_type unless a
well-defined class binding exists.
	- Added documentation to booting-without-of.txt for the EMAC
binding.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>

Download from:

http://ozlabs.org/~dgibson/home/emac/powerpc-emac-new-20070530.patch

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

Device tree aware EMAC driver

[David Gibson]

Based on BenH's earlier work, this is a new version of the EMAC driver
for the built-in ethernet found on PowerPC 4xx embedded CPUs.  The
same ASIC is also found in the Axon bridge chip.  This new version is
designed to work in the arch/powerpc tree, using the device tree to
probe the device, rather than the old and ugly arch/ppc OCP layer.

This driver is designed to sit alongside the old driver (it lies in
drivers/net/ibm_emac and this one in drivers/net/ibm_newemac).  The
old driver is left in place to support arch/ppc until arch/ppc itself
reaches its final demise (not too long now, with luck).

This driver still has a number of things that could do with cleaning
up, but I think they can be fixed up after merging.  Specifically:
	- Should be adjusted to properly use the dma mapping API.
Axon needs this.
	- Probe logic needs reworking, in conjuction with the general
probing code for of_platform devices.  The dependencies here between
EMAC, MAL, ZMII etc. make this complicated.  At present, it usually
works, because we initialize and register the sub-drivers before the
EMAC driver itself, and (being in driver code) runs after the devices
themselves have been instantiated from the device tree.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
---

Paul, I discussed with BenH, and although there are some problems with
the driver still, we think it's good enough to merge in 2.6.24, the
warts can be fixed up later.

Too big for the list, patch is at:
	http://ozlabs.org/~dgibson/home/emac/powerpc-emac-new-20070807.patch

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

Re: Device tree aware EMAC driver

[Josh Boyer]

On Tue, 7 Aug 2007 16:22:31 +1000
David Gibson <david@gibson.dropbear.id.au> wrote:

> Based on BenH's earlier work, this is a new version of the EMAC driver
> for the built-in ethernet found on PowerPC 4xx embedded CPUs.  The
> same ASIC is also found in the Axon bridge chip.  This new version is
> designed to work in the arch/powerpc tree, using the device tree to
> probe the device, rather than the old and ugly arch/ppc OCP layer.
> 
> This driver is designed to sit alongside the old driver (it lies in
> drivers/net/ibm_emac and this one in drivers/net/ibm_newemac).  The
> old driver is left in place to support arch/ppc until arch/ppc itself
> reaches its final demise (not too long now, with luck).
> 
> This driver still has a number of things that could do with cleaning
> up, but I think they can be fixed up after merging.  Specifically:
> 	- Should be adjusted to properly use the dma mapping API.
> Axon needs this.
> 	- Probe logic needs reworking, in conjuction with the general
> probing code for of_platform devices.  The dependencies here between
> EMAC, MAL, ZMII etc. make this complicated.  At present, it usually
> works, because we initialize and register the sub-drivers before the
> EMAC driver itself, and (being in driver code) runs after the devices
> themselves have been instantiated from the device tree.
> 
> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>

Hm.  Should this go through Jeff Garzik or Paul?  If it's the latter,
I'll pull this into my git tree soon.

josh

Re: Device tree aware EMAC driver

[David Gibson]

On Tue, Aug 07, 2007 at 07:15:39AM -0500, Josh Boyer wrote:
> On Tue, 7 Aug 2007 16:22:31 +1000
> David Gibson <david@gibson.dropbear.id.au> wrote:
> 
> > Based on BenH's earlier work, this is a new version of the EMAC driver
> > for the built-in ethernet found on PowerPC 4xx embedded CPUs.  The
> > same ASIC is also found in the Axon bridge chip.  This new version is
> > designed to work in the arch/powerpc tree, using the device tree to
> > probe the device, rather than the old and ugly arch/ppc OCP layer.
> > 
> > This driver is designed to sit alongside the old driver (it lies in
> > drivers/net/ibm_emac and this one in drivers/net/ibm_newemac).  The
> > old driver is left in place to support arch/ppc until arch/ppc itself
> > reaches its final demise (not too long now, with luck).
> > 
> > This driver still has a number of things that could do with cleaning
> > up, but I think they can be fixed up after merging.  Specifically:
> > 	- Should be adjusted to properly use the dma mapping API.
> > Axon needs this.
> > 	- Probe logic needs reworking, in conjuction with the general
> > probing code for of_platform devices.  The dependencies here between
> > EMAC, MAL, ZMII etc. make this complicated.  At present, it usually
> > works, because we initialize and register the sub-drivers before the
> > EMAC driver itself, and (being in driver code) runs after the devices
> > themselves have been instantiated from the device tree.
> > 
> > Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
> 
> Hm.  Should this go through Jeff Garzik or Paul?  If it's the latter,
> I'll pull this into my git tree soon.

Through Jeff Garzik, I think.  I was planning to send it to him today
or tomorrow if there are no screams from the linuxppc-dev list.

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

Re: Device tree aware EMAC driver

[Tony Breeds]

On Tue, Aug 07, 2007 at 04:22:31PM +1000, David Gibson wrote:
> Based on BenH's earlier work, this is a new version of the EMAC driver
> for the built-in ethernet found on PowerPC 4xx embedded CPUs.  The
> same ASIC is also found in the Axon bridge chip.  This new version is
> designed to work in the arch/powerpc tree, using the device tree to
> probe the device, rather than the old and ugly arch/ppc OCP layer.

<snip>

Hi David,
	I had a look over the patch FWIW, asside from the points below
looks good.

Minor nits:
 * Shouldn't ndev->priv accesses, use netdev_priv() ?  if not a comment
   somewhere about why not will keep the janators off your back.
 * c++ style comments
 * in emac_probe():
	+	/* Wait for dependent devices */
	+	err = -ENODEV;
	+	err = emac_wait_deps(dev);
    The initialisation to -ENODEV is pointless right?
 * s/get_property/of_get_property/g  or you'll make sfr grumpy ;P
 * In drivers/net/ibm_newemac/Makefile, I think the preferred method is
   to use ibm_newemac-y rather than ibm_newemac-objs.

Is there any reason to leave config IBM_NEW_EMAC (and IBM_EMAC) in
drivers/net/Kconfig ?  It seems to me they'd be nicer to in the
appropriate drivers/net/*Kconfig.


Yours Tony

  linux.conf.au        http://linux.conf.au/ || http://lca2008.linux.org.au/
  Jan 28 - Feb 02 2008 The Australian Linux Technical Conference!

Re: Device tree aware EMAC driver

[David Gibson]

On Wed, Aug 08, 2007 at 01:16:43PM +1000, Tony Breeds wrote:
> On Tue, Aug 07, 2007 at 04:22:31PM +1000, David Gibson wrote:
> > Based on BenH's earlier work, this is a new version of the EMAC driver
> > for the built-in ethernet found on PowerPC 4xx embedded CPUs.  The
> > same ASIC is also found in the Axon bridge chip.  This new version is
> > designed to work in the arch/powerpc tree, using the device tree to
> > probe the device, rather than the old and ugly arch/ppc OCP layer.
> 
> <snip>
> 
> Hi David,
> 	I had a look over the patch FWIW, asside from the points below
> looks good.
> 
> Minor nits:
>  * Shouldn't ndev->priv accesses, use netdev_priv() ?  if not a comment
>    somewhere about why not will keep the janators off your back.

netdev_priv() hey?  Hadn't heard of that before.  I guess
so... Changed.

>  * c++ style comments

Eh, all the c++ comments are FIXMEs anyway.  I'm inclined to leave
them.

>  * in emac_probe():
> 	+	/* Wait for dependent devices */
> 	+	err = -ENODEV;
> 	+	err = emac_wait_deps(dev);
>     The initialisation to -ENODEV is pointless right?

Yes, I think so.  Removed.

>  * s/get_property/of_get_property/g  or you'll make sfr grumpy ;P

Oops, yes.

>  * In drivers/net/ibm_newemac/Makefile, I think the preferred method is
>    to use ibm_newemac-y rather than ibm_newemac-objs.
> 
> Is there any reason to leave config IBM_NEW_EMAC (and IBM_EMAC) in
> drivers/net/Kconfig ?  It seems to me they'd be nicer to in the
> appropriate drivers/net/*Kconfig.

Yes, and that also seems to be how it's done for tulip.  Moved.

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

Re: Device tree aware EMAC driver

[Benjamin Herrenschmidt]

On Wed, 2007-08-08 at 15:09 +1000, David Gibson wrote:
> 
> Eh, all the c++ comments are FIXMEs anyway.  I'm inclined to leave
> them.

Ahah... maybe I -did- add some of those then :-) I do use C++ comments
for FIXME's on purpose, because they are ugly, it improves the changes
that I actually get to fix whatever needs fixing :-)

Ben.

Re: Device tree aware EMAC driver

[Benjamin Herrenschmidt]

>  * c++ style comments

Ouch... I haven't removed them all ? I promise I didn't -add- any :-)

>  * in emac_probe():
> 	+	/* Wait for dependent devices */
> 	+	err = -ENODEV;
> 	+	err = emac_wait_deps(dev);
>     The initialisation to -ENODEV is pointless right?

Leftover from when it was coded differently yes.

>  * s/get_property/of_get_property/g  or you'll make sfr grumpy ;)
>  * In drivers/net/ibm_newemac/Makefile, I think the preferred method is
>    to use ibm_newemac-y rather than ibm_newemac-objs.
> 
> Is there any reason to leave config IBM_NEW_EMAC (and IBM_EMAC) in
> drivers/net/Kconfig ?  It seems to me they'd be nicer to in the
> appropriate drivers/net/*Kconfig.

Probably.

Ben.

Re: Device tree aware EMAC driver

[Segher Boessenkool]

>  * In drivers/net/ibm_newemac/Makefile, I think the preferred method is
>    to use ibm_newemac-y rather than ibm_newemac-objs.

I thought it was the other way around, so I checked with the
Kbuild maintainer, and indeed you are correct.


Segher

Re: Device tree aware EMAC driver

[David Gibson]

On Fri, Aug 10, 2007 at 10:39:31PM +0200, Segher Boessenkool wrote:
> >  * In drivers/net/ibm_newemac/Makefile, I think the preferred method is
> >    to use ibm_newemac-y rather than ibm_newemac-objs.
> 
> I thought it was the other way around, so I checked with the
> Kbuild maintainer, and indeed you are correct.

Changed accordingly.

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

Re: Device tree aware EMAC driver

[Olof Johansson]

On Tue, Aug 07, 2007 at 04:22:31PM +1000, David Gibson wrote:

> This driver is designed to sit alongside the old driver (it lies in
> drivers/net/ibm_emac and this one in drivers/net/ibm_newemac).  The
> old driver is left in place to support arch/ppc until arch/ppc itself
> reaches its final demise (not too long now, with luck).

I would very much prefer if the patch was provided as first a pure copy
without changes, then the actual diffs on top of that. The way it is now
is just a huge blob that's harder to review than it would otherwise be.


-Olof

Re: Device tree aware EMAC driver

[Benjamin Herrenschmidt]

On Thu, 2007-08-23 at 13:56 +1000, David Gibson wrote:
> 
> 
> Jeff, I've discussed this with BenH, and although there are some
> problems with the driver still, we think it's good enough to merge in
> 2.6.24, the warts can be fixed up later.  Please apply...

Or to be more precise:

This driver will work well for 4xx platforms and will allow us to move a
big step toward getting rid of arch/ppc. The issues with DMA mapping
really only concern the case where this is used within the cell "Axon"
southbridge, which doesn't happen on any released product at this stage.

Ben.

Re: Device tree aware EMAC driver

[Josh Boyer]

On Thu, 2007-08-23 at 09:30 +0200, Benjamin Herrenschmidt wrote:
> On Thu, 2007-08-23 at 13:56 +1000, David Gibson wrote:
> > 
> > 
> > Jeff, I've discussed this with BenH, and although there are some
> > problems with the driver still, we think it's good enough to merge in
> > 2.6.24, the warts can be fixed up later.  Please apply...
> 
> Or to be more precise:
> 
> This driver will work well for 4xx platforms and will allow us to move a
> big step toward getting rid of arch/ppc. The issues with DMA mapping
> really only concern the case where this is used within the cell "Axon"
> southbridge, which doesn't happen on any released product at this stage.

Since we have 4 boards that could use this now, and at least one more on
the way, I've put this in the 'emac' branch of my git tree.  That branch
is based off of the current 'upstream' branch of the netdev-2.6 tree.

This is just intended to make it easier for those that want to use
ibm_newemac on their boards while we work the driver into the netdev
tree.

josh