[antlr-interest] Complaints about BaseAST implementation

Paul Johnson gt54-antlr at cyconix.com
Thu Oct 13 00:41:35 PDT 2005


[Completely OT]

Akhilesh Mritunjai wrote:

> Well. One revealation. I work in an EDA tools starup
> and am currently lead for developing a compiler and
> analyzer for a leading HDL. The trees generated can
> have 1-50 million nodes depending on type of input,
> which of course I can't keep in memory because memory
> requirements for tree alone goes into several
> gigabytes. 

50M nodes? I'm intrigued. If this is RTL, then surely you should be 
doing incremental compilation, and should never reach a tree that size. 
If it's a netlist, then surely it's so dumb and flat that you could do 
just about anything and it would work. And, if you don't know what the 
input is, then you could have some sort of adaptive approach.

:)



More information about the antlr-interest mailing list