python-afl icon indicating copy to clipboard operation
python-afl copied to clipboard

American Fuzzy Lop fork server and instrumentation for pure-Python code

Results 14 python-afl issues
Sort by recently updated
recently updated
newest added

As indicated in README, the instrumentation is slow at the moment. Here are some rough ideas how to speed it up: * Replace `sys.settrace()` with lower-level `PyEval_SetTrace`. * Rewrite bytecode...

help wanted

The errors that afl-fuzz prints are tailored to compiled programs; they are not very helpful in the context of python-afl, especially for newcomers. Related issues: #13, #21, #22,

enhancement

I get the same error when I fuzz all of tests except for target.py and target_persistent.py . But I cant solve it. I hope you can help me, Thanks!! ![Q5BU80K1O7U~23O~...

more info needed

I'm trying to use `python-afl`. When I run `python-afl`, it print this error. `The current memory limit (50.0 MB) is too restrictive, causing an OOM fault in the dynamic linker`...

more info needed

When I use the tool to test a function which has complex input parameters, such as list, dictionary or etc, the afl fuzz would generate new inputs, but the input...

Cc: @maage The afl-cmin test fails with recent AFL++ versions: ``` ====================================================================== ERROR: tests.test_cmin.test ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python3/dist-packages/nose/case.py", line 197, in runTest self.test(*self.arg) File ".../python-afl/tests/test_cmin.py", line...

bug
help wanted

Hello everyone! Why this error does not allow me to run fuzzing? I tried to manually set the `AFL_SKIP_BIN_CHECK=1` environment variables, but it doesn't help.

more info needed

This allows Python code to manually poke at the trace map, which might be useful for blackbox testing. Possible alternative to #9. Before this gets merged: - [ ] Document...

Implemented opcode tracing. This version updates hash until we do something else. So my tests/target.py generates different maps. Without this feature same ops generate just increased count and I felt...

Hey, I was wondering if the opcode tracing added with Python 3.7 might be useful: ```python def wrap(frame, event, arg): frame.f_trace_lines = False frame.f_trace_opcodes = True code = frame.f_code if...

help wanted