Results 59 comments of Marcel Böhme

Sorry about that. We'll investigate. Worst case, we'll trade memory for disk storage and write the trial specific coverage data to the (trial specific) subject-fuzzer/trial folder to handle concurrency in...

Great idea! I suppose this would show the coverage overlap (difference) between fuzzers? Might help with debugging a fuzzer.

Fuzzbench distinguishes runner and measurer and only the seed corpus ever reaches the measurer. So, getting coverage statistics right from the fuzzing process might be a bit more difficult. If...

The following is based on data from [Stephan Lipp](https://www.in.tum.de/i04/lipp/) (@sphl) of TU Munich. I used that data to play around with ways to visualize coverage progress. The graphs can be...

Thanks Danushka! Maybe I can add an interpretation. We use the log10(hits) just because the scale is substantial. The X axis shows the basic block IDs ordered by log10(AFL_hits) -...

That's from the coverage profile generated by executing the entire queue at the end of the campaign.

> Is there a way to access this data? PR #778 is to generate that data in Fuzzbench. > I wonder if it would be helpful to display how often...

That's log10(hits) sorted by Honggfuzz hits. ![image](https://user-images.githubusercontent.com/5443180/94750904-884da880-03ca-11eb-91e0-f75c72b32024.png)

Still committed to do the analysis 😄 Let me know once the run is done.