Hello,

> Also, the approach describe there has some fundamental flaws.
> According to Gopal V, "It has a huge array which it uses like C uses RAM.
> Not pretty, fast, stable or useful", and according to Rhys Weatherly,
> "He's using a similar approach to egcs-jvm.  See the pnet FAQ for a
> description of why this isn't a good way to port gcc to a stack machine",
> and they are both right.

Still a useful foundation, and good direction to bring plenty of
compilers over to .NET.  

The important bit about this work is the fact that the register-based
RTL representation is used to map to CIL code.  The other nuisances are
technical details.  

In the "solvable" universe, not in the "impossible" universe.

Miguel.
_______________________________________________
Mono-list maillist  -  [EMAIL PROTECTED]
http://lists.ximian.com/mailman/listinfo/mono-list

Reply via email to