On Tuesday 04 November 2014 23:41:26 Alexandre Belloni wrote:

On 31/10/2014 at 22:36:55 +0100, Arnd Bergmann wrote :

quoted

On Friday 31 October 2014 21:57:56 Alexandre Belloni wrote:

quoted

On 31/10/2014 at 21:50:05 +0100, Arnd Bergmann wrote :

quoted

On Friday 31 October 2014 21:45:58 Alexandre Belloni wrote:

quoted

To be able to make the watchdog driver independent from the mach/ includes, pass
the system timer register space as a resource.

Also, change the name to avoid conflicting with the at91sam9 watchdog driver.

Signed-off-by: Alexandre Belloni <redacted>

Doing this change as a separate patch breaks bisection because now the device
name no longer matches untile the other patch is applied too.

Yeah, I was not sure how important that was as there is no user of the
watchdog in the kernel. My thinking was that both patch can then go
through different trees.

I can definitely squash them.

AFAICT, arch/arm/configs/at91rm9200_defconfig enables the device and it
gets registered through at91_add_standard_devices. You definitely have
my Ack to merge the mach-at91 patch through the watchdog tree.

You're right, I missed that one. I was expecting it to be called from
board files.

So, I'll squash both patches, add your SoB and your Ack and get it
merged through the watchdog tree, tell me if that is not what you
expect.

I definitely *don't* expect you to add my Signed-off-by, that would
be against the procedures we have in Documentation/SubmittingPatches.

Other than that, it sounds good, thanks!

	Arnd