jhc vs ghc and the surprising result involving ghc generated assembly.

Florian Weimer fw at deneb.enyo.de
Wed Nov 2 13:47:54 EST 2005


>> It seems that the goto-based version leads to different static branch
>> prediction results, which happen to be favorable. 
>
> It has nothing to do with branch prediction. I know 
> it is determined ENTIRELY by stack use.

In both cases, The C compiler emits code which doesn't use the stack.


More information about the Glasgow-haskell-users mailing list