[FFmpeg-devel] [PATCH] build: fix make checkheaders in out-of-tree builds

[Andreas Cadhalpun]

Hi Ronald,

On 24.01.2016 14:06, Ronald S. Bultje wrote:
> On Sun, Jan 24, 2016 at 6:52 AM, Andreas Cadhalpun <
> andreas.cadhalpun at googlemail.com> wrote:
> 
>> Thus I object to reverting this before the regression caused by 31741ae is
>> fixed.
> 
> 
> This is ridiculous, I (who didn't break it and don't even have hw
> supporting any of our hwaccel implementatons) already proposed a method to
> fix it _properly_ (without introducing alternate regressions) that everyone
> agreed to.

I'm not sure if the method you proposed would work. I don't know the hwaccel
implementation very well, as I'm a mere user of that feature.

> That is, everyone that responded. After a re-poke. And you
> weren't one of them.

And thus I haven't responded, yet. If you're confident that it will work,
just give it a go. I'm happy to test patches.

> Which makes me wonder, you're very good at objecting,

Thanks for the compliment.

> but you never give +1s on good ideas that are different (but better) than
> your ideas.

I never give '+1s', because I prefer to write full sentences, including
some sort of explanation. Doing that requires to first figure out whether it
could work, which can take a lot of time.

> That's a terrible attitude. You're not perfect. Other people's ideas can be better
> than yours.

Agreed.

> Can you please look at my RFC in the related thread and comment
> on the proposed fix? If you like it, that includes +1'ing it so I can
> assume we have general consensus before I implement it and get bikeshedded
> after doing work on it.

I'm fine with any solution that makes VLC's hwaccels work again.
So you can assume there is a general consensus for your suggestion, if it works.

> And after you've done that, can we not bring this subject and 31741ae up
> again?

Fine with me.

Best regards,
Andreas