+- Kodi Community Forum (https://forum.kodi.tv)
+-- Forum: Development (https://forum.kodi.tv/forumdisplay.php?fid=32)
+--- Forum: Kodi Application (https://forum.kodi.tv/forumdisplay.php?fid=93)
+---- Forum: GSoC (https://forum.kodi.tv/forumdisplay.php?fid=299)
+----- Forum: GSOC 2019 (https://forum.kodi.tv/forumdisplay.php?fid=293)
+----- Thread: GSoC 2019 Proposal- Fuzzing (/showthread.php?tid=342634)
GSoC 2019 Proposal- Fuzzing - ac-lucas - 2019-04-01
Hi!
My name is Amy and I am a freshman at the University of Michigan. I am interested in project 4.9- fuzz testing. I have put together my proposal and opened it for comments here.
I am open to feedback for improving my proposal (but not so much grammar/punctuation type things). Note: I have temporarily removed the sample photo of my code since it is code from a school project and posting it online can constitute an honor code violation.
Thank you for this opportunity, I look forward to possibly working with Team Kodi in the coming months!
RE: GSoC 2019 Proposal- Fuzzing - Razze - 2019-04-01
Hey there nice to have you.
Liked your proposal, just one question so far. Why did you go for american fuzzy lop? Did you checkout libfuzzer?
RE: GSoC 2019 Proposal- Fuzzing - ac-lucas - 2019-04-02
Hi!
I wrote specifically about AFL because all of the sources I read while researching fuzzing used it, but I wasn't set on it because the project description mentioned that the decision process would be part of the project. I actually hadn't specifically looked up the features of Libfuzzer. But, after reading a bit about the comparison it seems like Libfuzzer has a few more useful features than AFL, especially the range of supported sanitizers. I've also seen recommendations to run both since they have slightly different strategies- it seems like this might be the way to go to catch the most bugs possible. Plus, it could eliminate the need for running sanitizers seperately.
Would you recommend that I edit my proposal and timeline to fuzz with both? Thanks for the feedback!
RE: GSoC 2019 Proposal- Fuzzing - Razze - 2019-04-02
I would recommend to focus on one of them, after all, gsoc feels kinda short
From the things I heard, I would have thought, that libFuzzer would fit better with kodi code, but I haven't played with that.
RE: GSoC 2019 Proposal- Fuzzing - ac-lucas - 2019-04-03
(2019-04-02, 13:42)Razze Wrote: I would recommend to focus on one of them, after all, gsoc feels kinda short
From the things I heard, I would have thought, that libFuzzer would fit better with kodi code, but I haven't played with that.
Sounds good. I agree that it would be a better fit and have modified my proposal! Thanks again for the feedback.