Future of linux-arch

Future of linux-arch

[Russell King]

It seems that me posting these two patches just to linux-arch has caused
some upset in the kernel community.  In particular, Christoph Hellwig
is talking about publicising the address for this list.

11:48 < hch> rmk: why don't you simply CC lkml?
11:49 < hch> or actually subsribe lkml to linux-arch
11:49 < hch> that would be easiest
11:49 < rmk> because davem doesn't want the address getting into spammers
             databases.
11:49 < rmk> so we're not allowed to crosspost between linux-arch and lkml
11:49 < hch> well, I can get that one into it even without subsribing
11:50 < hch> and if doesn't stop beeing silly I'll just scatter websites
             containing mailto:linux-arch@vger.kernel.org all over google

and apparantly according to David Woodhouse, this is somehow my problem.

Sigh.  Just warning people here when this becomes a complete spam haven.
Sometimes I wonder why the fsck I bother following and defending peoples
rules.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 PCMCIA      - http://pcmcia.arm.linux.org.uk/
                 2.6 Serial core

Re: Future of linux-arch

[David S. Miller]

On Sun, 28 Mar 2004 12:01:33 +0100
Russell King <rmk@arm.linux.org.uk> wrote:

> It seems that me posting these two patches just to linux-arch has caused
> some upset in the kernel community.  In particular, Christoph Hellwig
> is talking about publicising the address for this list.

If he can't respect people's privacy, that shouldn't become our problem.

If a group of people interested in 'foo' wish to discuss 'foo' on a mailing
list amongst themselves, I don't see what can be wrong with that.

Maybe I should feel compelled to CC: lkml every time I come up with an
idea I wish to discuss with Linux or Andrew or one of the networking
co-maintainers?  See how rediculious such thinking is?

Now, on the other hand, it may have in fact been more appropriate to discuss
this particular change on lkml, flesh out the API issues publicly, _then_
move what you end up with to linux-arch so that arch people can code up
implementations for their platforms for you.

See, because this patch goes beyond arch issues, driver folks have to use
these interfaces, so logically it appears that folks outside of the arch
community should contribute to the design and review of said interfaces.

Right?

Re: Future of linux-arch

[Russell King]

On Sun, Mar 28, 2004 at 12:29:20PM -0800, David S. Miller wrote:
> See, because this patch goes beyond arch issues, driver folks have to use
> these interfaces, so logically it appears that folks outside of the arch
> community should contribute to the design and review of said interfaces.

So this is the thanks for taking hold of an issue which someone else
raised and trying to get it sorted.  Fine.

But that's no good if we can't even present what we think is a good
interface from the architecture side.  Which it appears we still
can't.

Remember that it was Andrea who raised it on lkml and it moved to
to linux-arch because all the architecture people needed to be
involved?

Well, I'm sick to death with this issue.  Someone else can sort out
this problem.  I'm going to just do what's necessary to sort it on
ARM and leave it at that.  If that screws someone else, tough shit.
I've made the effort and done my best to find a reasonable solution.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 PCMCIA      - http://pcmcia.arm.linux.org.uk/
                 2.6 Serial core

Re: Future of linux-arch

[David S. Miller]

On Sun, 28 Mar 2004 21:48:19 +0100
Russell King <rmk@arm.linux.org.uk> wrote:

> Well, I'm sick to death with this issue.  Someone else can sort out
> this problem.

The reality is that akpm asked me to create this alias so that
two things could happen:

1) Andrew could contact all arch maintainers when he was messing with
   something that could break platforms or needed explicit platform support.

2) Arch maintainers could notify each other of problems they've found
   in their own arch code which applies to other platforms as well.

So far it's worked wonderfully at both of these things.

You're worked up over a seperate issue and the fact that hch is being
his usual fucknut self about things like this, just ignore him and do
the fantastic job you always do Russell.