On 27 February 2015 at 19:47, Keean Schupke <[email protected]> wrote:

>
>
> On 27 February 2015 at 19:40, Jonathan S. Shapiro <[email protected]>
> wrote:
>
>>
>> As an abstract model from which to proceed with optimization, I agree
>> with what you say. Conceptually we can stack the extra arguments and move
>> them back into registers later. It brings on some tricky alias analysis in
>> the presence of by-reference arguments, but it can be done. It may even be
>> a good approach to consider from an optimization perspective.
>>
>>
>> shap
>>
>
>
> And new processors like The Mill's belt architecture rely on it. It
> effectively has an infinite supply of new registers, and every register
> store gets a new register, but you cannot access results older than slot
> 'N' where 'N' is a model dependent number something like 1000+
>
> Keean.
>

I guess kernel calls are the main use case for pass-by-register? In 'C' you
would probably have to write a bit of assembly to get data from a specific
register, and probably register names cannot be part of the language
because they are platform dependent.

I can't help feeling that the details of the platform (number and names of
registers, and special uses for them) should be kept out of the language.


Keean.
_______________________________________________
bitc-dev mailing list
[email protected]
http://www.coyotos.org/mailman/listinfo/bitc-dev

Reply via email to