Closed bendrissou closed 2 months ago
Are not using the GitHub state? -D does nothing for quite some time now
I am using a slightly older commit. The fuzzer freezes for extremely long intervals during deterministic steps. Not sure how to debug this issue.
Each call to post-processing function consume around 10 milliseconds. This is not a problem in the non-deterministic mode, but once I enable the deterministic mode, the fuzzer starts to freeze. Could it be that deterministic mutations invoke post-processing multiple times at each cycle?
@vanhauser-thc I want to use the old deterministic mutations, including auto extras
, so I have resorted back to a previous commit 1ffb1b6
. I think it's useful to keep the deterministic strategy as I find it useful. I can't find documentation for the new enhanced deterministic strategy, plus the dashboard does not reflect the new changes!
you are using an outdated version with a feature that does not exist anymore. please understand that I have zero time to support that... you are on your own.
Hi,
I am running AFL with a post-processing function. The fuzzer works well. However when I try to run the fuzzer with
-D
(the deterministic stage), the fuzzer executes very slowly then freezes after some time. On a local machine it eventually crashes with code 137. On a large machine, it freezes:Any idea why the deterministic stage does not work well with custom mutators?