On Mon, Apr 07, 2014 at 09:31:39AM -0700, Olof Johansson wrote:
On Mon, Apr 7, 2014 at 7:40 AM, Russell King - ARM Linux linux@arm.linux.org.uk wrote:
On Mon, Apr 07, 2014 at 07:30:50AM -0700, Olof Johansson wrote:
On April 7, 2014 7:25:20 AM PDT, Russell King - ARM Linux linux@arm.linux.org.uk wrote:
I haven't applied anything from the patch system for the last month or more as I've been soo bogged down with the l2c and fec changes.
A good question at this point is what change introduced this regression, and why did that change go through a different tree to that which is being asked to carry the fixes?
Good point. I'll apply the fix to arm-soc this morning.
Not quite - the two changes "Add cpu_is_pj4 to distinguish PJ4 because it has some differences with V7" and "Check cpu id in pj4_cp0_init." are both clearly core code changes, so should be applied via my tree.
Hence my question about what introduced this regression.
You know... Arnd moaned at me over the weekend having changes to arch/arm/boot/dts in my tree, saying that they should go via arm-soc, but it seems that it's perfectly fine for arm-soc to take core ARM code changes on a whim.
I don't care which stops: either arm-soc stops taking changes which should come via my tree, or arm-soc maintainers accept that from time to time I will be carrying changes to arch/arm/boot/dts which may conflict with changes in arm-soc.
What is unacceptable is both complaining and also take core ARM changes.
Awesome!
I'm strongly in favor of code going in through the appropriate tree since it makes life easier for maintainers, and I am looking forward to you starting sending non-core patches over to us.
I didn't know we had a resolution to that problem, and I'm glad to see that we do. This made my morning!
That's not what I said. We have a *big* problem in that many of our patches affect each other's trees, that's because we work in the same area.
We both have to accept that we're going to keep conflicting - there are always going to be core ARM changes which affect SoC specifics, and there's always going to be SoC specific stuff which affects core ARM stuff. There's no getting away from that.
That's why I believe it's totally wrong that arm-soc has sole ownership of anything under arch/arm. That just doesn't work.
Plus, let's not forget that I'm the primary maintainer for a bunch of ARM platforms (though you wouldn't ever know it given how arm-soc behaves...) and that's been *really* pissing me off over the last few months...