Thread (14 messages) 14 messages, 5 authors, 2010-07-24

Re: [PATCH 2/2] mpc85xx_edac: change to use new definitions for PCI EDAC regspace

From: Dmitry Eremin-Solenikov <hidden>
Date: 2010-07-22 19:03:05
Also in: linux-edac

Hello,

On Thu, Jul 22, 2010 at 10:40 PM, Kumar Gala [off-list ref] wr=
ote:
On Jul 22, 2010, at 1:25 PM, Scott Wood wrote:
quoted
On Thu, 22 Jul 2010 20:48:15 +0400
Dmitry Eremin-Solenikov [off-list ref] wrote:
quoted
Hello,

On Thu, Jul 22, 2010 at 7:38 PM, Kumar Gala [off-list ref]=
 wrote:
quoted
quoted
quoted
On Jul 21, 2010, at 7:03 PM, Dmitry Eremin-Solenikov wrote:
quoted
Currently (as mpc8540-pci) devices are not created on of_platform bus=
,
quoted
quoted
quoted
quoted
mpc85xx_edac can't probe to them. Follow the change to dts trees to b=
ind
quoted
quoted
quoted
quoted
not to the main mpc8540-pci node but to special mpc85xx-pci-error nod=
es,
quoted
quoted
quoted
quoted
present on soc bus.

Signed-off-by: Dmitry Eremin-Solenikov <redacted>
---
drivers/edac/mpc85xx_edac.c | =A0 18 +++++++++---------
1 files changed, 9 insertions(+), 9 deletions(-)
Nak.

We already have a node in the dts for the PCI controller. =A0Lets upda=
te the platform code to add the pci controller to the of_platform_bus_probe=
 list.
quoted
quoted
I've had that idea. However it's really look strange to me to call
of_platform_bus_probe() on the bus node, for which we (IMO) explicitly
won't like for
child devices (PCI devices) to be added to of_platform bus.
Right, and it's also not great for a driver for one aspect of PCI to
claim to be the driver for the whole thing.

But changing the device tree because of this Linux-internal concern is
also not good.

How about keeping the error stuff as a separate device from Linux's
perspective, but have the main Freescale PCI code create it as a
platform device instead of putting it in the device tree?
I'd be good with that solution.
Then we come back to the question that was raised before (during initial
review of edac driver): as PCI code is probbed long before other parts
of the kernel and mpc85xx_edac code can be compiled as module,
it's not possible to directly call mpc85xx_edac code from fsl_pci.c

Two initial suggestions were:
1) creating special platform device
2) creating special of_platform device from dts

Which approach should I choose? Did i miss any other opportunities?

--=20
With best wishes
Dmitry
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help