Tail pad optimization, cache friendlyness and C++ interrop

[Artur Skawina via Digitalmars-d]

On 06/20/14 13:51, David Nadlinger via Digitalmars-d wrote:
> On Friday, 20 June 2014 at 00:15:36 UTC, Walter Bright wrote:
>> On 6/19/2014 12:59 PM, Joakim wrote:
>>> I don't know enough about these copyright tainting concerns to say if it's a
>>> good idea, just pointing out that he was talking about the backend, not the
>>> frontend.
>>
>> Ok, but I'll also point out that LDC and GDC are fully free & open source.
> 
> Which wouldn't really help Artur (whether his concerns are justified or not), as we usually tell people to contribute their frontend patches directly to the upstream DMD repository.

Yes. Also, like I've already said, working on top of a downstream tree
would eventually either result in a fork, or fail, with the latter
being the much more likely result.

I'll just add that I now think I've overstated the gains that splitting
out the free frontend would bring. That's because I've since realized
how hard it would still be to deal with development on top of git head,
when one can not immediately test the result /on real code/ (ie using a
non-dmd backend).
Without a truly shared frontend, there's no good solution, I guess. :(

artur