Hi,
This is my first post to the verif academy forum. Do let me know if I’m in using the forum incorrectly and/or am unwittingly violating any guidelines.
Feel free to correct me if I am wrong, but it appears to me that SV does not allow
a) extending covergroups
b) crossing coverage between diff cover groups
c) does not support some stuff that’s useful for AOP like “is also”, when inheritance
Is that correct ? How do we request for these things be added to the language ?
HAH
Yes, this is the right forum to be asking SystemVerilog questions.
You are correct about those features not in SystemVerilog. They have all been requested in the IEEE committee’s bug-tracking system, 2117, 2993, and 3002.
I see a lot of requests for the first issue about extending covergroups. Have that feature would alleviate much of the need for crossing different covergroups, which would be difficult to do in the general case.
AOP is a much bigger topic. You can start many flame wars discussing AOP versus OOP programming paradigms. The only thing I will say here is that SystemVerilog has for the most part taken the OOP path and you may need to think about solving your problems in a very different way than you would in an AOP language.
There are two ways to get your input to the IEEE SystemVerilog committee - either join it directly, or pass your input to your tool vendor indirectly.
Hi HAH,
Not sure how Mentor would perceive this - but Synopsys’s VCS has already extended SystemVerilog to support all the 3 you have asked in Beta/LCA form since few years. But sure they are not standard yet. We need strong users like yourself to demand this to language committee.
Dave - when you say “this is right forum” - may I know if you channelize such requests to SV-EC like committees? Since SV is a “controlled” IEEE standard, individuals/small companies can’t really contribute - unfortunate, but reality!
Ajeetha, CVC
In reply to Ajeetha Kumari CVC:
Hi Ajeetha,
Most of my career has been to channel the requests of users to the companies and committees I work on.
These features are not part of the standard yet. It would certainly help if the users of these non-standard features implemented in other tools would put pressure on their vendor to put them in the standard.