search

ChatAFLndss / ChatAFL

Large Language Model guided Protocol Fuzzing (NDSS'24)
Apache License 2.0
246 stars 27 forks source link

running speed #1

Closed ruikangPeng closed 10 months ago

ruikangPeng commented 11 months ago

Hi, author； I ran ChatAFL according to your guidance document, but I found that the speed of obtaining experimental results was relatively slow. Is this the same as when you ran the experiment?

ChatAFLndss commented 11 months ago

Hi @ruikangPeng ,

Yes, the execution time can vary for some subjects, like Kamailio, due to subject specific requirements such as state cleanup and proper termination. This is related to how we collect code coverage data. After the fuzzing campaign completes, all seeds are replayed on the coverage instrumented version, and coverage metrics are then obtained using the gcov program.

ruikangPeng commented 11 months ago

@ChatAFLndss Thank you for your reply. I have learned a lot from your research, especially the one click script execution experiment, which is particularly eye-catching. I hope to have more communication with you in the field of protocol fuzzing.

ChatAFLndss commented 11 months ago

@ruikangPeng Thank you. Happy to discuss with you more!

moyiun09 commented 10 months ago

Hi @ruikangPeng This OPENAI_API_KEY is provided in the appendix of the article. I did not find the corresponding key. Did you use the key you applied for?