Performance problem in reverse algorithm

[Daniel N via digitalmars-d-ldc]

On Monday, 25 August 2014 at 18:04:02 UTC, Fool wrote:
> Thanks, there is no time pressure. I compared ASM and LLVM-IR 
> of the fast and slow versions using clang++. Unfortunately, my 
> foo in this area is non-existent. The only thing I noticed is 
> that the fast version has significantly longer ASM and LLVM-I 
> representations.
>
> Kind regards,
> Fool

I tried changing 'immutable' to 'enum' and adding 'pure nothrow' 
annotations on the functions, got a decent speedup for ldc, I 
currently don't have gdc installed, so I can't check if these 
changes make gdc even faster or if it's part of the difference.

Regards,
Daniel N