Fix an opcode covergroup that can never pass 75%, by excluding reserved codes and marking the forbidden one illegal.
Hardware context: An opcode decoder with reserved codes and one forbidden code. Exclude what the plan does not need so that a complete regression can reach 100%.
This is a Fix problem: the covergroup in the editor has one realistic defect. Find it and fix it.
A controller accepts 4-bit opcodes. The plan groups them in fours, grp[0] to grp[3]. Codes 12 to 14 are reserved and never sent. Code 15 is KILL, which the bench must never send.
The best regression sends every code from 0 to 11, and the covergroup still stops at 75%.
Signals in the Timeline
| Signal | Width | Role |
|---|---|---|
samples | 8 | input of each check |
Full specification
| Codes | Meaning | In the plan? |
|---|---|---|
| 0 to 11 | commands | yes, in groups of four |
| 12 to 14 | reserved, never sent | no |
| 15 | KILL, must never be sent | no, and its appearance is an error |
Each case samples a fresh instance and reads get_inst_coverage() times 100, rounded. The cases may send reserved codes; they never send KILL, which would stop the run on the coverage build. The full check samples more lists than the examples show.
Examples
Example 1
Every Command but One- Input
opcodes: 0 1 2 3 4 5 6 7 8 9 10
- Output
coverage_percent_x100 = 10000
All three command groups are hit. With grp[3] out of the number, that is 100.00%.
Example 2
A Reserved Code Changes Nothing- Input
opcodes: 2 6 13
- Output
coverage_percent_x100 = 6667
grp[0] and grp[1]: 2 of 3. The 13 is excluded.
Example 3
Only Reserved Codes- Input
opcodes: 12 14
- Output
coverage_percent_x100 = 0
Nothing counted is hit: 0.00%.
Constraints and Contract
- Keep
grp[4] = {[0:15]}and the covergroup and coverpoint names. - Leave codes 12 to 15 out of the coverage number, and make KILL an
illegal_bins.
Hints
Hint 1
With every code from 0 to 11 sent, which group can never be hit?
Hint 2
Is a reserved code an error, or simply not part of the plan? Is KILL?
Hint 3
Exclusions are applied after the codes are dealt into the groups. What happens to a group whose codes are all excluded?
What the grader checks: Your covergroup is sampled with 18 lists of opcodes (never KILL). Its instance coverage is compared with the value IEEE 1800-2023 clause 19 gives. Three lists are shown. Whether 15 is illegal or ignored gives the same numbers on these lists.
18 checks on Verilator coverage build · server. Run the examples first, then check the full regression. Full checks use the server judge. Downloads contain public samples only. A passing result establishes the tested cases, not a proof for all possible inputs.
See all 2 steps and the final capstone