[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: gEDA: A Plea for Programming Help (Icarus Verilog)



Actualy, one could use formal to validate the compiler. Run the test
circuit through XST and through Icarus, then formally compare the two
results. Also, I think some formal tools permit RTL<->Gates comparison.

The problem here is the cost of the tools and the non-trivial nature of
writing a formal verification tool which uses synthesis, ATPG, fault
simulation, expert systems and other complex technologies to do what
they do.

The problem with the proposed method is estabilishing correctness. That
is, if you use simulation to validate work product, how do you know when
you've simulated enough. Coverage tools are available, but they also
cost $$$. Trivial test-cases for which coverage is easy to determine may
not stress the compiler. Exhaustive testing of sequential circuits can
be, well, exhausting.

I am also willing to contribute to this project. Perl hacking, I can do
and, given a set of test goals, I can readily provide verilog code to
stress the compiler. I already do, actually. Unfortunatly they are of
the dump core variety.

Best regards,
Jason



-----Original Message-----
From: Stephen Williams [mailto:steve@icarus.com] 
Sent: Monday, March 24, 2003 9:07 AM
To: geda-dev@seul.org
Subject: Re: gEDA: A Plea for Programming Help (Icarus Verilog) 



jrsheahan@optushome.com.au said:
> I'll volunteer to help with this. regressions in perl + verilog is 
> something I've done before.

> just thinking though - this is formal verification land, rather than 
> simulation.  could reducing the code work?

Not quite. Formal verification is for verifying the Verilog, but we are
trying to verify the compiler. There are, I believe, subtle differences.
And I really want to go end to and and back again in the tests of -tfpga
so I really want something that compares the Verilog with the mapped
netlist fed through Xilinx tools. The ngd2ver program seems the most
likely tool here.

Anyhow, look at the ivtest project on sourceforge. I use the vvp_reg.pl
test rig almost daily, for example, and the vpi_reg.pl program as well
if I am working on stuff that might affect VPI.


-- 
Steve Williams                "The woods are lovely, dark and deep.
steve at icarus.com           But I have promises to keep,
steve at picturel.com         and lines to code before I sleep,
http://www.picturel.com       And lines to code before I sleep."