Re: [Pull Request] Eseries patches

[Ian Molton <spyro@xxxxxxx>]

Eric Miao dropped from the CC because I doubt he wants two copies of this.

I didn't particularly want to write it, but don't want to leave things unsaid, so...

...

Russell King - ARM Linux wrote:

>> Well when you dont reply for several days, what should I do? sit here
>> and wait longer?
> 
> How about you actually check my published tree once in a while to make
> sure what you want there is there, and it's there in the right state?

I do, something which should be obvious from my comment in:

Message-Id: <1218813813.30234.13.camel@wirenth>

Which reads:

   'Will you pull this version? I can't see it in your tree so I'm
    guessing you haven't yet.'

>> and its your own fault I had to do that twice because as you pointed
>> out to me, you willfully ignored one of my messages (your choice of word).
> 
> Of course it's my fault for you not checking whether it's been pulled.

I did check, and when I (admittedly mistakenly) didn't find the changes, I sent
you a message, which you ignored.

I could, if pushed, post the contents of the mail you sent me where you actually
state you ignored me, though, if you like?

> git diff mybranch...yourbranch
>
> If empty, it's been merged.

There doesn't seem to be a way to check if its been pulled without knowing which
branch its been pulled into, short of checking every single branch (which is how
come I managed to miss you pulling the patchset - I must have missed a branch).

> Five messages between 19th and 28th August concerning pulling these
> changes - that's an average of one every two days.  Don't you think
> that's excessive?

Nope.

Lets see...

19/08/08  Initial pull request, generated with git request-pull
     No response.
24/08/08  Ping sent privately.
     No response.
25/08/08  First public ping.
     Response to the effect that your PC is broken.
28/08/08  Second public ping, sent once it appeared you fixed your PC.
     No response.
05/09/08  Third public ping.
     <cue public roasting>

If you're going to ignore people, you can expect them to wonder what's
going on.

> Also, bear in mind that whenever I send a pull request to Linus, it's
> _always_ formatted how Linus wants it.  Never is a short "please pull
> this URL".  Missing out things like the diffstat is a recipe for being
> ignored, or being flamed by Linus.  Even if he missed a pull request.
> It still has to be formatted to his liking.

I have no problem with that, if you'd stated you want reminders to basically be
complete new requests, then that's fine.

As you've said no such thing (to me at least) then I reserve the right to have
'got it wrong'.

Obviously I'll not do it that way again, but if you'd have simply said what you
want instead of leaving me to guess, it'd have been much simpler.

> So why is it seemingly unreasonable for me to require the same thing
> about your pull requests?

It isn't. Just ask nicely instead of leaving people to guess.

> Why is it soo difficult for you to provide
> the information I ask for?

Its not. You asked for a proper pull request, and you got one in

Message-ID: <48AAC3B0.4080303@xxxxxxx>

> Why do you always have to question such requests?

To be sure I understand correctly. And when you fail to answer questions, or you
don't make something clear, then you don't get things the way you like.

Like now.

you never stated that reminders should be complete pull requests until just now,
so you didn't get reminders formatted as complete pull requests.

Now I know your preference, that's what you'll get - but it didn't have to be
this hard to find out.

> Oh, it's obvious is it?  I'm sorry, I should've consulted my crystal
> ball which would've told me.  Get real Ian.

Typically, one does not send reminders for completely new material.

Ergo, its the same patchset.

Which I stated quite clearly in:

Message-ID: <48AD424A.109@xxxxxxx>

and

Message-ID: <48B28788.7000703@xxxxxxx>

> As for your claim that you sent a pull request with a diffstat in earlier,
> I see *no* evidence of that having been done *ever* for this set, unlike
> your claims.  I've gone back to 21st August, no diffstat for the 'for_rmk'
> branch.  The only one I can find is on the 19th August which was for the
> 'for_rmk_17' branch.

Same patchset, different branch name. Either way, the above two mails state that
they are the same patchset, and the properly formatted pull request was sent in
Message-ID: <48AAC3B0.4080303@xxxxxxx> as mentioned above.

You also got an email from me explaining that all future patchsets for you would
be in for_rmk, now that I know about git push -f

You had all the pieces but didn't put them together, which is the same offence
you are roasting me over right now. I had all the information about wether you
pulled my patches, and I fucked up. I can admit that, but it seems you cannot.

> 	IF YOU WANT ME TO PULL YOUR TREE, PLEASE SEND PROPERLY
> 	FORMATTED PULL REQUESTS, EVERY TIME, AS I ALREADY REQUESTED.

You never said (or implied) the 'every time' bit (which would include reminders,
I guess) before, unless I'm mistaken. But as I said above, I know now.

> And yes, I've tried asking for that politely, and evidence is that it
> doesn't work.

Perhaps you weren't clear about it. I don't think you were, anyway.

> Finally, note that my original plan was to be away from 20th to 31st
> August.  Do you think that I want a mailbox full of people whining
> "RMK - Where are you?"

Do you honestly expect me to remember your holiday plans? If I applied that to
all the people I keep contact with I'd be spending more time consulting the
calendar than I would be actually doing anything productive.

Anyhow, I hope this is over now. Can we get back to something more useful now?


-------------------------------------------------------------------
List admin: http://lists.arm.linux.org.uk/mailman/listinfo/linux-arm-kernel
FAQ:        http://www.arm.linux.org.uk/mailinglists/faq.php
Etiquette:  http://www.arm.linux.org.uk/mailinglists/etiquette.php