[erlang-questions] Updates, lenses, and why cross-module inlining would be nice

Jesper Louis Andersen jesper.louis.andersen@REDACTED
Wed Feb 17 16:38:06 CET 2016


On Mon, Feb 15, 2016 at 4:38 PM, Pierre Fenoll <pierrefenoll@REDACTED>
wrote:

> This is surely a very naive idea on how to one day get cross-module
> inlining
> or at least enlarge the range of possibilities for the JIT.
>

A JIT should be free to build up optimized code with full inlining as long
as it knows how to deoptimize that code again should a new version of a
module be loaded. One of the alluring thing of a JIT is that it is free to
make dynamic choices in the running code without any priori knowledge. A
traditional compiler which works on static data needs to operate on static
analysis alone. This is usually easier if the program has less wiggle room
of interpretation. And this is the case if you have a static type system...



-- 
J.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20160217/a01bb315/attachment.htm>


More information about the erlang-questions mailing list