Open yoava333 opened 6 years ago
I think it's a great idea :-)
+1
Le lun. 23 juil. 2018 à 09:38, ivanfratric notifications@github.com a écrit :
I think it's a great idea :-)
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/ivanfratric/winafl/issues/121#issuecomment-407121868, or mute the thread https://github.com/notifications/unsubscribe-auth/ABaHRal7d6p5ZoPcivKRvDU5dGI-tqztks5uJfwQgaJpZM4VbEZI .
We have a target that has a very slow startup time (~30 seconds) when but the target function is quite fast (500 iter/s). When we want to cmin a corups of ~10k files on a 8 core machine it would take
30 * 10,000 / 8 = ~10.5
hours. I suggest extending the in memory fuzzing feature of winafl to cmin, I created a basic prototype that runs through the same corpus on the same machine in ~7 minutes.What do you think?