mnemonikr
commented
8 months ago 104 improves coverage a bit . Once that's merged, coverage should be -- based on cargo llvm-cov --workspace:
Filename Regions Missed Regions Cover Functions Missed Functions Executed Lines Missed Lines Cover Branches Missed Branches Cover
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
sla/src/ffi/rust.rs 8 0 100.00% 3 0 100.00% 24 0 100.00% 0 0 -
sla/src/ffi/sys.rs 55 16 70.91% 49 13 73.47% 54 18 66.67% 0 0 -
sla/src/opcodes.rs 226 21 90.71% 46 18 60.87% 221 19 91.40% 0 0 -
sla/src/sleigh.rs 175 54 69.14% 62 22 64.52% 436 55 87.39% 0 0 -
src/emulator.rs 1613 493 69.44% 169 10 94.08% 2604 96 96.31% 0 0 -
src/mem.rs 176 58 67.05% 34 9 73.53% 379 82 78.36% 0 0 -
src/processor.rs 78 24 69.23% 19 7 63.16% 136 21 84.56% 0 0 -
sym/src/buf.rs 127 20 84.25% 64 10 84.38% 283 53 81.27% 0 0 -
sym/src/convert.rs 155 9 94.19% 68 0 100.00% 328 3 99.09% 0 0 -
sym/src/sym.rs 497 67 86.52% 156 11 92.95% 1015 81 92.02% 0 0 -
-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
TOTAL 3110 762 75.50% 670 100 85.07% 5480 428 92.19% 0 0 -
In #87 a bug was discovered during PR since the memory module didn't cover branches that were covered by integration tests. It's good that the integration test caught the issue, but it was actually a simple issue that should have been caught by a unit test (reading from constant address space). Added code to handle the case but not a unit test. There are likely other branches not well covered by unit tests.
Should document existing coverage for all modules in this issue and then improve as necessary. Once improved issue may be resolved.