aboutsummaryrefslogtreecommitdiff
path: root/fuzzpuzz.c (follow)
Commit message (Collapse)AuthorAge
* Fix use-after-free in fuzzpuzzBen Harris2023-01-16
| | | | | | | | | | | | When reporting that the game name in a save file isn't recognised, don't include the name from the save file in the error message, partly to avoid the complexity of freeing it properly on two different code paths and partly because including unsanitized data from a fuzzer-supplied save file in the error message just seems dangerous. And properly sanitising it would waste the fuzzer's time exploring the sanitising code. Thanks to Ben Hutchings for reporting the bug.
* Remember to free the game name in fuzzpuzzBen Harris2023-01-12
|
* Don't leak midends in fuzzpuzzBen Harris2023-01-12
| | | | If deserialising a save file fails, the midend still needs to be freed.
* Merge the two versions of fuzzpuzz back togetherBen Harris2023-01-12
| | | | | | | Now there's a single version of the main loop that runs once in normal mode and repeatedly in AFL++ persistent mode. In persistent mode, fmemopen() allows the loop to read the shared-memory buffer as though it were a stdio stream. fmemopen() is POSIX-only, but so is AFL++.
* AFL-specific shared-memory fuzzing modeBen Harris2023-01-12
| | | | | | | Rather than a save file from standard input and then exiting, this reads it from a shared memory buffer and then loops. This makes fuzzing _much_ faster: one core on my laptop can now load about 30,000 save files per second.
* Add a fuzzing harness for PuzzlesBen Harris2023-01-12
This just feeds save files into the loading code, but because of how Puzzles is structured that actually exercises most of its parsers.