On 05. 11. 18 14:20, Borislav Petkov wrote:

On Mon, Nov 05, 2018 at 02:06:11PM +0100, Michal Simek wrote:

quoted

I don't think that driver will be broken. You can build them, use them
on out of tree HW. And when this patch is merged to mainline it will be
enabled for xilinx soc.

But if the DT entries are missing, the driver won't load, would it?

you don't have that HW anyway.

quoted

TBH I can't see any reason to do merges but if you want to do that way
we can also do it.

The reason is because there's a separate DT tree and all those arm
drivers need DT.

I have already acked EDAC patches to go through other trees too, FWIW.

I looked at v10 and I can't see your ack there. Can you please give me a
link?
I see Rob's reviewed by in v10 2/6

Which is not optimal either if someone sends fixes ontop but I cannot
apply them yet because the dependent patches are in a different tree.

So yes, there are at least two good reasons for merging a shared branch.

I have not a problem with that. I can simply take patch (process via
arm-soc) with pointing to reviewed binding doc and you will take the
rest when this is in arm-soc tree.

Thanks,
Michal