On Mon, Jan 24, 2011 at 1:27 PM, Elias Gabriel Amaral da Silva <
[email protected]> wrote:

> But I confess that most BitC issues (such
> as possibility of running with no GC, and the whole discussion about
> boxing) are a bit foreign to me. Properly separating pure code from
> impure code draw my attention...


For the overwhelming majority of applications, avoiding GC is not
interesting.

For a few very narrowly specific codes, avoiding *allocation* during normal
execution is imperative. It is a happy consequence that we can disable GC,
but the goal is to avoid allocation. The rationale is either predictability
of execution time or predictability of resource non-starvation. This is
mainly interesting for things like microkernels. It is secondarily
interesting for things like CODECs. It is probably not very interesting for
anything else.

The goal for BitC from the beginning was to allow such codes to be written
in a safe low-level language. For such codes, it is acceptable to restrict
idioms. It has never been my expectation or intent that this requirement
would be helpful or relevant to general-purpose applications.


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

Reply via email to