feat(fuzz): prioritize favored corpus entries - #15818
Conversation
This comment was marked as outdated.
This comment was marked as outdated.
Sorry, something went wrong.
|
derek bench fuzz compare-ref=master timeout=600 |
|
cc @mablr bench (fuzz) event published. Results will be reported separately. View publisher run Config: subcommand: |
|
derek bench invariant timeout=10800 |
|
cc @grandizzy bench (invariant) event published. Results will be reported separately. View publisher run Config: subcommand: |
|
derek bench invariant timeout=7200 |
|
cc @mablr bench (invariant) event published. Results will be reported separately. View publisher run Config: subcommand: |
06d1647 to
f667074
Compare
✅ Changelog foundThe deterministic check will validate the changed entry. |
|
derek bench invariant timeout=7200 |
|
cc @mablr bench (invariant) event published. Results will be reported separately. View publisher run Config: subcommand: |
|
derek bench invariant timeout=600 trials=6 paired-seeds=true |
|
cc @mablr bench (invariant) event published. Results will be reported separately. View publisher run Config: subcommand: |
6819746 to
a71075d
Compare
|
derek bench invariant timeout=600 trials=6 paired-seeds=true |
|
cc @grandizzy bench (invariant) event published. Results will be reported separately. View publisher run Config: subcommand: |
a71075d to
af5de41
Compare
af5de41 to
7adb9af
Compare
| .collect() | ||
| } | ||
|
|
||
| fn mutation_energy( |
There was a problem hiding this comment.
maybe worth looking at https://github.com/AFLplusplus/LibAFL/blob/de0b34d2c561256e8d241af1a42851ea8803ab22/crates/libafl/src/schedulers/testcase_score.rs#L53-L267
and the metadata tracking
There was a problem hiding this comment.
Would it make sense to start narrowly with favored status and per-entry selection/productivity counts? These directly inform donor selection using metadata Foundry can already track, while execution-time calibration and full power scheduling would require broader fuzzing-lifecycle changes.
4 participants
Stacked on #15817.
Makes the coverage minset affect donor selection, not just eviction. The scheduler gives energy to favored entries, rare edges, and recent coverage yield, while retaining bounded uniform and disk exploration.
Adds a focused scheduling regression test.
Prompted by: @0xalpharush