|
Well, I'm interested in creating a VPI interface
for my code coverage tool, Covered. Currently, the tool reads in generated
VCD files and then performs a partial "resimulation" of the data to figure out
the coverage metrics that it needs for reporting purposes. The benefit of
reading VCD files is that most, if not all, simulators are capable of generating
one, allowing Covered to remain simulator independent. However, the
process of "resimulation" is quite costly in terms of time (and possibly file
space -- if the VCD file(s) is/are quite large) and I don't currently support
all Verilog constructs, so I'm interested in removing the need to perform this
"resimulation" step.
The VPI is quite interesting to me in that I can
access more than just value changes within the design as it is running.
Thus, I can get the information I need (signal, _expression_ and statement
information) without having to do a lot of post-processing of the
results. Of course, there would be some overhead for getting this
information, but the end result is something that is quite a bit better than
what I currently have. Additionally, the VPI seems to be becoming
standardized and accepted by many simulator vendors -- achieving my original
goal of not becoming a simulator-dependent tool.
I really enjoy using Icarus Verilog (it is my
simulator of choice when I'm not at work :) and would like to see/learn with
this tool, if possible. I currently don't know exactly what I need to be
added to the VVP VPI, but was more interested in knowing if it would be possible
to extend IV's VPI interface in the future (I wasn't sure if the current state
of the VPI in the VVP was more or less fixed for good reasons that I wasn't
aware of).
I appreciate the responses and will make
suggestions for pieces/parts that I would need to be added as I come across
them. I will also take a look at Cver and give that a try in the
meantime. As always, keep up the great work that is being done to
Icarus!
Trevor
|
- gEDA: Icarus Verilog VPI support Trevor Williams
- Re: gEDA: Icarus Verilog VPI support Stephen Williams
- Re: gEDA: Icarus Verilog VPI support Matthew Ballance
- Re: gEDA: Icarus Verilog VPI suppor... Trevor and Jenny Williams
- Re: gEDA: Icarus Verilog VPI su... Stephen Williams
- Re: gEDA: Icarus Verilog V... Trevor and Jenny Williams
- Re: gEDA: Icarus Veril... Stephen Williams
- Re: gEDA: Icarus V... Trevor and Jenny Williams
- gEDA: pcb-20050609 and gtk Daniel J Wisehart
- Re: gEDA: Icarus Verilog VPI su... Paul Johnson
