Tail pad optimization, cache friendlyness and C++ interrop

[Iain Buclaw via Digitalmars-d]

On 22 June 2014 08:21, SomeDude via Digitalmars-d
<digitalmars-d at puremagic.com> wrote:
> On Saturday, 21 June 2014 at 10:49:57 UTC, Artur Skawina via
>
> Digitalmars-d wrote:
>>
>>
>> It's not about being able to contribute to DMD, it is about being able
>> to work on /other/ projects. If contributing to DMD carries the risk of
>> affecting the latter then it's simply best to avoid it; it's not a risk
>> worth taking, just for a few small improvements. Significant work often
>> starts with simple and trivial fixes; if scratching-an-itch is too costly
>> then major contributions suffer too. Note that whether the risk is
>> significant, or even real, doesn't really matter much -- it's the cost of
>> making the decision that matters. Just-submit-a-small-patch-to-a-boost-
>> -licensed-project turns into investigate-the-licensing-and-evaluate-all-
>> -the-potential-legal-implications. It's enough to discourage submissions
>> *even in the cases where there is no problem*.
>>
>> artur
>
>
> I really don't see what the issue is. If the projects are
> unrelated, there is no  reason there could be a "contamination".
>
> And even with that, nothing prevents you from working on the
> front end with LDC or GDC.


And if your reasoning for not working on the front-end for GDC or LDC
is that they are behind current development.  Then why don't you make
your first contribution as updating the chosen compiler to be aligned
with DMD development!

This is on the projects page for GDC at least, and soemwhere hidden on
my long TODO list, but there are too many other important things to
look after for the time being.  :)

Regards
Iain.