[sv-ec] late-breaking covergroup Mantises

From: David Scott <david_scott_at_.....>
Date: Tue Jan 22 2008 - 17:42:35 PST

Too late, I expect, for SystemVerilog 2008, we've found some serious issues that arose from our consideration of Mantis 1897 (http://eda.org/svdb/view.php?id=1897) -- now completely approved by SV-EC and the Champions.

Here they are:

• The versions of get_coverage with optional "ref" arguments can't (or shouldn't) quite work as currently described: http://eda.org/svdb/view.php?id=2242
• My interpretation of covergroup option.per_instance leads me to think that post-process coverage merging would be crippled by default if tools actually implemented the standard: http://eda.org/svdb/view.php?id=2243
• Coverage database behavior implied by the standard is relatively ill-defined, leaving a bunch of questions unanswered, I think: http://eda.org/svdb/view.php?id=2244

-- David Scott, Mentor Graphics


--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean. Received on Tue Jan 22 17:42:57 2008