46dcfc8 (ARM: Update mach-types) removes used machine types

On Mon, Aug 22, 2011 at 9:48 PM, Russell King - ARM Linux
[off-list ref] wrote:

On Mon, Aug 22, 2011 at 03:36:12PM +0200, Marek Vasut wrote:

quoted

On Monday, August 22, 2011 03:32:42 PM Russell King - ARM Linux wrote:

quoted

On Mon, Aug 22, 2011 at 03:30:41PM +0200, Marek Vasut wrote:

quoted

On Monday, August 22, 2011 03:27:19 PM Russell King - ARM Linux wrote:

quoted

On Mon, Aug 22, 2011 at 03:16:00PM +0200, Marek Vasut wrote:

quoted

On Monday, August 22, 2011 03:01:19 PM Russell King - ARM Linux wrote:

quoted

For further reading, see:
http://lists.arm.linux.org.uk/lurker/message/20110815.085325.ae6ee0
7d.e n.ht ml

Russell, calm down please.

Maybe it'd be better to just email the people with broken stuff, then
wait ... if they don't reply in a week (two weeks?) then remove their
stuff. And make that a policy.

What do you think ?

Who do I email? ?Which entries are causing problems? ?That's the whole
bloody point.

Isn't the email in your database? You can't just script that ?

quoted

There's just far too much of it. ?The amount of effort required to sort
through this file each time it needs to be updated has become
*excessive*. The amount of effort required to go through the file and
identify which entries are broken is *excessive*. ?The amount of
effort required to find who to email is *excessive*.

Why not write a script for that ?

No, I don't have the time at present.

You're the head maintainer here and ARM is a rising architecture. If you can't
cope with that, you have a growing community of very competent people here.
Maybe someone will be willing to take over eventually? It'll relieve some
pressure from you too, what do you think?

Hello. ?Do you even understand the issue you're talking about?

There is a problem _right_ _now_ with the mach-types file.

That problem was caused by gplugd being merged without its entry being in
the mach-types file - which was omitted - by way of editing by hand, along
with shit loads of other entries - because of the fuckup with the
machine_is_xxx() name not matching the configuration or MACH_TYPE symbol.

Grrrr.... as the original guilty one who stirred this whole pile of
sh*t up, should
I keep my mouth shut up on this?  ;-)

True that it's a stupid mistake for code to use non-matched machine entries
and it's fair to simply remove them.

Yet I guess what us sub-maintainers would like to know is the correct way
to amend this (after the incorrect entry being removed), and to better help
with "getting it right"?

So, rather than wasting a *significant* amount of time hand editing the
file at every update (which is precisely why I haven't been updating the
bloody thing) I've built it into the script so that entries which do not
conform are automatically omitted.

Problem solved.

Any remaining problems are down to the maintainers not talking to me and,
if you haven't realised yet, I have _zero_ sympathy for that especially
when it doesn't take much effort _on their part_ to do.

It's got nothing to do with ARM being a rising architecture. ?It's about
maintainers doing the right thing, behaving in the right way and taking
responsibility for their fuckups. ?Putting the pain where the pain should
be - in the platform maintainers lap, not in my lap.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel at lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel