I would like to collate a list of what people are measuring for verification.
Since there are numerous things which can be measured, many things that should be measured, and a whole bunch of things that shouldn't never be measured, I would like to poll the community to see which things you (or your managers) are currently measuring, or measured on your previous project.
Once we compile a list I think it would be interesting to sort them by stage in the project and then do some informal voting to decide what the top 3-5 things to measure in each stage of the project.
So as a first step I'll jot down some of the things that I have seen measured without mentioning the stage of the project and if they are useful or not. Here they are:
Coverage Definition completed (for each feature & block) (Ex. 3 of 16 are complete)
Coverage review completed (for each feature & block)
% Functional Coverage achieved (including failed tests)
% Functional Coverage achieved (excluding failed tests)
% Line coverage
% Toggle Coverage
% Condition Coverage
Code review complete for each block
Checker code review complete
# of lines of RTL changed week to week
# of lines of TB changed week to week
# of rtl bugs
# of rtl showstopper/medium/low bugs
# of rtl bugs in each block
# of bugs in the testbench
# of days turnaround time for rtl bugs
# of assertions per lines of code
Please add your lists so we can start doing some analysis…
Originally posted in cdnusers.org by Akiva