Arnd Bergmann wrote:

On Monday 30 June 2014 16:22:25 Thierry Reding wrote:

quoted

On Mon, Jun 30, 2014 at 03:34:27PM +0200, Arnd Bergmann wrote:

quoted

On Tuesday 01 July 2014 06:32:10 Kukjin Kim wrote:

quoted

This series removes S5P64X0 and S5PC100 related codes in mainline,
because no more user now. And if its supporting is required later,
it will be done with using device tree.

Acked-by: Arnd Bergmann <arnd@arndb.de>

It seems like I'm not completely up-to-date. What happened to our prime
directive never to remove support for boards? Why is this special? What
about people that still use this hardware? The commit message says that
there are no longer any users. How was this determined?

I asked whether we may remove its supporting in mainline before not just one
time and then no objection at that time...

From what I can tell, the supported machines on these two platforms are
all the official development prototypes, not production boards, and they
came out in rather small quantities, probably Samsung-internal for the
most part. The maintainers have previously assured me they are not aware
of anybody using it, and I trust their judgment.

Yeah, I think so ;-)

I'm totally fine with removing support for these machines, and we have
removed a number of similar platforms in the past. The rule is not that

S5P6442. I did commit a73ddc61 'ARM: S5P6442: Removing ARCH_S5P6442' in 2011
because that could not be mass-produced.

we never remove stuff, the rule is that we never break things for
active users.

YES.

As always, if it turns out that we missed some active users, we will
revert these patches, or bring support back another way.

Sure, so please let us know if you have any objection on this removing.

This should probably all be spelled out explicitly in the changelog
for patch 1, which I admit I have not seen.

It's my fault, sorry I forgot it and I'm resending 1/17 and 8/17 with -D.

Thanks,
Kukjin