Introduce HashMutator #2771
Introduce HashMutator #2771
Conversation
* Rules * more * aa
* fixing empty multipart name * fixing clippy * improve flexibility of DumpToDiskStage * adding note to MIGRATION.md
Updates the requirements on [bindgen](https://github.com/rust-lang/rust-bindgen) to permit the latest version. - [Release notes](https://github.com/rust-lang/rust-bindgen/releases) - [Changelog](https://github.com/rust-lang/rust-bindgen/blob/main/CHANGELOG.md) - [Commits](rust-lang/rust-bindgen@v0.70.1...v0.71.1) --- updated-dependencies: - dependency-name: bindgen dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
* no from stage * fixer * doc fix * how was this working???? * more fixes * delete more * rq * cargo-fuzz * m * aa
* go * fixing stuf * hello from windows * more * lolg * lolf * fix * a --------- Co-authored-by: Your Name <[email protected]>
* Maybe fix CI * does this help? * Very dirty 'fix'
* fixing empty multipart name * fixing clippy * New rules for the contributing (AFLplusplus#2752) * Rules * more * aa * Improve Flexibility of DumpToDiskStage (AFLplusplus#2753) * fixing empty multipart name * fixing clippy * improve flexibility of DumpToDiskStage * adding note to MIGRATION.md * Introduce WrappingMutator * introducing mutators for int types * fixing no_std * random fixes * Add hash derivation for WrappingInput * Revert fixes that broke things * Derive Default on WrappingInput * Add unit tests * Fixes according to code review * introduce mappable ValueInputs * remove unnecessary comments * Elide more lifetimes * remove dead code * simplify hashing * improve docs * improve randomization * rename method to align with standard library * add typedefs for int types for ValueMutRefInput * rename test * add safety notice to trait function * improve randomize performance for i128/u128 * rename macro * improve comment * actually check return values in test * make 128 bit int randomize even more efficient * shifting signed values --------- Co-authored-by: Dongjia "toka" Zhang <[email protected]> Co-authored-by: Dominik Maier <[email protected]>
|
Or even |
|
As I stated in the discussion thread, I think a method for rejecting inputs that were already tried would be more useful (but I don't know your use case, so..) |
I'm targeting the TCP/IP stack of an OS, so each execution takes in the order of magnitude of 1s, although most of that is spent in wait states (hence previous work like overcommit). Even still, the added runtime of this would be nothing compared to the execution, so this felt like an easy win.
Something like this would definitely further improve the situation. Do you suggest creating a wrapping executor that returns either Tracing this back it seems most appropriate in the stage? But that seems not that generic. So maybe in I'm also not sure if there's an opportunity here to combine this somehow with |
|
I think it could simply wrap an executor, yeah. And have an extra observation that's "skipped" -if it's true the testcase isn't interesting. Should be easy enough to do. \We can still merge this PR as well, but the feedback should be renamed IMHO. |
See #2759.
Some mutators report
MutationResult::Mutated, even if nothing actually changes about the input.HashMutatoris a wrapper around other mutators that hashes inputs pre- and post-mutation to ensureMutationResult::Mutatedis only reported if something actually changed.This may be worth using on slow targets, where the hashing is quicker than the unnecessary additional executions of the target for previously tried inputs.