Re: [Qemu-devel] [PATCH V2] booke_206_tlbwe: Discard invalid bits in MAS2

[Alexander Graf]

On 30.05.2012, at 14:38, Andreas Färber wrote:

> Am 30.05.2012 14:16, schrieb Alexander Graf:
>> 
>> On 30.05.2012, at 13:54, Andreas Färber wrote:
>> 
>>> Am 30.05.2012 13:29, schrieb Alexander Graf:
>>>> 
>>>> On 30.05.2012, at 13:22, Andreas Färber wrote:
>>>> 
>>>>> Am 30.05.2012 10:13, schrieb Alexander Graf:
>>>>>> 
>>>>>> No reply from Blue, so I applied this patch to ppc-next :). Thanks a lot!
>>>>> 
>>>>> I did reply though, so I'm not okay. Please rebase on top of the patches
>>>>> that I have supplied you with. Rebasing that set is no fun.
>>>> 
>>>> Of course it's no fun, but it's neither crucial nor are you Blue. I'm wary 
>>>> enough with a big change like this, but your mail didn't sound like you 
>>>> were 100% confident that you took everything into account :(.
>>>> 
>>>> I'd really prefer to have Blue just resend a set that converts the target 
>>>> and be done with it. Unless you're 100% sure that you did everything 
>>>> correctly. Then please resend the whole patch set with your own SOB lines 
>>>> and declare your own ownership (or blameship rather) ;).
>>> 
>>> It is obvious that you haven't even looked at it or you would've seen my
>>> SoB. :(
>>> 
>>> I am confident that I did the rebasing of your series right, and I
>>> pointed out that your series was better than Blue's latest before it
>>> vanished.
>>> 
>>> I am not confident that Blue's original conversion was fully correct,
>>> but since it worked and had your SoB I didn't have to worry. ;)
>> 
>> Heh :). Yeah, my main concern was the resend to the ML. And yes, I never 
>> look at git trees. A patch set is either a mail or doesn't exist to me :). 
>> So just resend it, I'll read through it, and then I'll merge things in my 
>> tree.
> 
> You didn't post your fixed-up version to the list either!

Yes, because it would've gotten posted on the next pull request :).

> But I can
> easily do so, whatever you two do with it. You could've simply replied
> to my mail requesting me to send it rather than pretending as if you
> received nothing and pushing something else instead. That's the issue
> I'm having here.

Noted.

> 
>>> P.S. Don't understand what is not supposed to be crucial here?
>> 
>> That one's easily explained. If the patch set isn't applied, the world 
>> doesn't fall apart :).
>> 
>>> I do see
>>> a working qemu.git master branch and progress with merging QOM into
>>> qemu.git as crucial. Whereas pulling logging workarounds into ppc-next
>>> is not crucial at all and could be done by Anthony/Blue just as well.
>> 
>> I don't see your point.
> 
> Quoting agraf: "If the patch set isn't applied, the world doesn't fall
> apart :)."
> 
> So I don't see why you want to push non-crucial, non-ppc patches to
> ppc-next and force people (Blue/me) to rebase large series with code
> movements onto them rather than working towards applying the large
> series first and rebasing small patches on top of it, which is much
> easier for everyone involved and much more fair.

I can certainly handle putting the 2 patches in my queue on top of the big 
patch set myself, no worries ;).


Alex