Patches, bottlenecks, OpenSource

[Justin Spahr-Summers]

On Thu, 15 Apr 2010 15:41:20 -0400, bearophile 
<bearophileHUGS at lycos.com> wrote:
> 
> Don:
> 
> >And really, D doesn't need many people working on the DMD compiler.<
> 
> I agree, it's like having many people working around a dead corpse trying to revive it. Better use the time to adopt gcc and llvm back-ends at their best, keeping in mind, while designing D, that there are features that those back-ends have and the dmd back-end doesn't have that it will be good to add to the language. Simple example: refusing computed gotos because they are a lot of work to implement is not a justification that holds if both gcc and llvm already 
implement them and allow the front-end to just use them in a simple enough way.
> 
> Bye,
> bearophile

I like DMD, personally, and I've liked the benchmarks that I've seen of 
it (although lack of PowerPC support saddens me). GCC is a behemoth 
according to anyone who's ever done any work with it, and I like DMD's 
official status... even if LDC (for instance) were to become the 
"official" D compiler, the D compilation is so dependent on the LLVM 
system that it'd probably eventually run into issues like what Apple had 
with GCC not doing what they wanted; in Apple's case, they started a new 
project (clang, for those not familiar) so they could have solid control 
over the codebase.

As far as computed gotos, I think Walter explains somewhere on the D 
website that D should be an easy language for compiler writers to 
implement. I don't know the relative implementation difficulty of 
computed gotos, but I think the rationale for minimizing features 
actually in the compiler itself is sound. Reduces feature creep and all 
that good stuff.