linux-next: interesting merges in the arm-soc tree

linux-next: interesting merges in the arm-soc tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 714 bytes --]

Hi all,

I noticed that the arm-soc tree has merged in the v4l-dvb and the
slave-dma trees today.   Is there some good reason for this?  It fixes a
few conflicts (but that is not excuse) and there may be dependencies in a
driver on the v4l-dvb tree (but maybe that means that that driver should
be merged via the v4l-dvb tree - it looks like the "at91/drivers" is
based on the v4l-dvb tree, so probably doesn't depend on anything else on
the arm-soc tree).

If nothing else, are you sure that neither of those merged trees will
rebase?  You have also just inherited any bugs in those two trees.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

RE: linux-next: interesting merges in the arm-soc tree

[Kukjin Kim]

Stephen Rothwell wrote:
> 
> Hi all,
> 
Hi,

> I noticed that the arm-soc tree has merged in the v4l-dvb and the
> slave-dma trees today.   Is there some good reason for this?  It fixes a

I did slave-dma...

Yes, I have created branch for samsung dma on top of slave-dma tree. Because
samsung tree touched drivers/dma/pl330.c for samsung device tree then the
arch/arm/common/pl330.c file have been merged into drivers/dma/pl330.c with
touching samsung stuff and of course there are changes of
drivers/dma/pl330.c in slave-dma as well. So I did...

> few conflicts (but that is not excuse) and there may be dependencies in a
> driver on the v4l-dvb tree (but maybe that means that that driver should
> be merged via the v4l-dvb tree - it looks like the "at91/drivers" is
> based on the v4l-dvb tree, so probably doesn't depend on anything else on
> the arm-soc tree).
> 
> If nothing else, are you sure that neither of those merged trees will
> rebase?  You have also just inherited any bugs in those two trees.

I know, Vinod knows I merged his tree for above situation so he will not
rebase that and Arnd will send that after merging of slave-dma into mainline
during merge window. So I think, we will not see any problem. :) But if
wrong, please let me know.

Thanks.

Best regards,
Kgene.
--
Kukjin Kim <kgene.kim@samsung.com>, Senior Engineer,
SW Solution Development Team, Samsung Electronics Co., Ltd.

Re: linux-next: interesting merges in the arm-soc tree

[Nicolas Ferre]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 01/04/2012 12:47 AM, Stephen Rothwell :
> Hi all,
> 
> I noticed that the arm-soc tree has merged in the v4l-dvb and the 
> slave-dma trees today.   Is there some good reason for this?  It 
> fixes a few conflicts (but that is not excuse) and there may be 
> dependencies in a driver on the v4l-dvb tree (but maybe that means 
> that that driver should be merged via the v4l-dvb tree - it looks 
> like the "at91/drivers" is based on the v4l-dvb tree, so probably 
> doesn't depend on anything else on the arm-soc tree).
> 
> If nothing else, are you sure that neither of those merged trees 
> will rebase?  You have also just inherited any bugs in those two 
> trees.

We discussed this with Guennadi and Olof and decided to include this
dependency in arm-soc. Anrd and Olof said that the at91/drivers branch
could be merged late during the merge window and will make sure that
the corresponding v4l-dvb branch is actually upstream before sending
the pull request.

Here is the conversation about this:
http://thread.gmane.org/gmane.linux.ports.arm.kernel/144547/focus=144743

Best regards,
- -- 
Nicolas Ferre
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJPBCBqAAoJEAf03oE53VmQTtgH/2h9WBxjnwgcUrKOykHrXPPT
apB4y7FQA0EIaO6RWPi6LpXuBtEz+tV2mGyaXahUETgKQ9Oc459+FxoSoOmJyssh
vbaA4uWrPhdHXVVo464SLwxr/flUeb09iY3C+nIHn9JlYtqUriYOVUNCa+k/Oszp
AoaeDHPfjkkZ+07vZdjpYu9BC/NidijJvQDWCZ+Z4Jsfi7cygIB+HZGNAptPFmyq
jILk4LIdJcqsKmD0UA3i43WkTdqJQ4reZQk96POAX6151p7Ieh0PXNn68+oCYnwO
c6JAaM1FVsEhZAV1G/w0EpfDa0JNHep8h87GOHernrU6KuOVrWB0SCp5l0zERiw=
=mJTb
-----END PGP SIGNATURE-----

Re: linux-next: interesting merges in the arm-soc tree

[Arnd Bergmann]

On Wednesday 04 January 2012, Kukjin Kim wrote:
> > few conflicts (but that is not excuse) and there may be dependencies in a
> > driver on the v4l-dvb tree (but maybe that means that that driver should
> > be merged via the v4l-dvb tree - it looks like the "at91/drivers" is
> > based on the v4l-dvb tree, so probably doesn't depend on anything else on
> > the arm-soc tree).
> > 
> > If nothing else, are you sure that neither of those merged trees will
> > rebase?  You have also just inherited any bugs in those two trees.
> 
> I know, Vinod knows I merged his tree for above situation so he will not
> rebase that and Arnd will send that after merging of slave-dma into mainline
> during merge window. 

Right. The same is true for the other branch (depends/v4l) in next/drivers2.
The next/drivers2 branch will get submitted when all three of its dependencies
are merged.

Stephen, we have introduced a file arch/arm/arm-soc-for-next-contents.txt
to help us track those dependencies, and maybe it can also help you
understand our merge plans when you see potential problems like this.

Right now, there are 13 next/* branches that we plan to submit to Linus
in order. All of those are merged into the for-next branch that you pull,
and the contents.txt file tells you about the dependencies between those
branches and to external branches.

My understanding about the at91 changes from Nicolas Ferre is that they
depend on both the v4l and previous arm-soc changesets, so it was decided
to merge those through arm-soc after the v4l branch.

	Arnd