- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi, I am wondering which strategy would be the best for code coverage analysis of a full design.
Let's say I have a design with 10+ sub-modules; I have testbenches that focus on testing each individual sub-module. For simplification, we assume there is only 1 testbench associated with a sub-module. I am looking for a solution that would prevent having to run all of the testbenches if a sub-module is modified. I thought of something like as shown below. For each testbench associated with a particular sub-module instance, I save the code coverage statistics for only this particular instance. I'm not 100 sure it would work, but if so, if I modify a sub-module, I could just run again the testbench associated with this particular instance; I would not need to run again the others.
foreach arch $arch_list {
# Compile the test architecture
vcom -work work -2008
# Restart and Run the simulation
restart -f
run -all
# save code coverage statistics
coverage save -code bcefs -instance sim:/testbench/<associated instance>_inst
}
vcover merge
coverage open
coverage report -byinstance -file -detail -code {s b c e f } -noexcludedhits
Any idea if this could work? Or do you have a better strategy? Thank you for your time Damien
Link Copied
0 Replies
Reply
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page