* clean clippy warnings from fuzzers/binary_only/*
* handle unused Results in fuzzers/binary_only/*
* format fuzzers/binary_only/qemu_cmin
* use unchecked memory write in qemu fuzzer examples
* create file_null in fuzzbench_fork_qemu
* compile-time edge module compilation trick
* clippy
* possible since rust 1.79
* split edge module in submodules
* Update frida to 0.14.0 (#2596)
* update frida crate to the latest version
* adapt libafl_frida to the latest version of frida
* tracers and generators private modules
* do not use star export.
* same for drcov
* forgot a file...
* first draft of generic-based edge module for ConstantLengthMapObserver.
* integration of OwnedSizedSlice.
replaced OwnedSlice in ConstMapObserver by the new OwnedSizedSlice.
* fix serde stuff
* no std
* import
* fixed qemu_cmin with new constant map abstraction.
* fix const map
* fix clippy from another pr...
* fix non-null usage
* fix ci?
* new feature stuff
* fixes
* minor fixes
* fmt
* non null
* im stupid
* fmt
* fix fuzzer
* fix fuzzers
* sized slice
* fuzzer fixes
* ptr::NonNull -> NonNull
* shorter trait length
* fmt
* libafl_qemu: Add RISCV support
Adds the following targets (as features):
- riscv32
- riscv64
Added `RISCVCPU` and `CPURISCVState` to the bindings allow list.
Added riscv.rs to the arch module, with all necessary functions and
registers implemented and mapped.
The registers are the same as the ones found in qemus gdbstub xml found
after a build.
Additionally we added all syscall numbers for riscv 64 bit (already
supported by the `syscall_numbers` crate) and also added the missing
ones for riscv 32 bit. We compared both lists and their differences /
equalities with a simple python script and generated a list of the
missing ones, to be complete.
We might PR those to the `syscall_numbers` crate later on.
---------
Co-authored-by: Romain Malmain <romain.malmain@pm.me>
* feat(libafl_core): make executors and feedbacks more cleanly usable outside of LibAFLs Fuzzer loop
* cargo +nightly fmt
* updated type constraints
* reformatted and final type constraint fixes
* made unicode extraction stage useful separately
* fix libafl_cc error message
* fix state type constraint to be constrained on the method
* removed unnecessary observer constraint
* renamed unused variables
* fix unnecessary error wrapping in helper functions
* converted unicode conversion stage into associated function and fixed nautilus changes
* more update
* Remove extra I
* more fmt
* bounds?
* less bounds
* more less bounds
* different trait bounds again
* more less generics
* fix unicode
* fix list
* remove unneeded bound
---------
Co-authored-by: Lukas Dresel <Lukas-Dresel@users.noreply.github.com>
Co-authored-by: Toka <tokazerkje@outlook.com>
* add python grammar loader for Nautilus
* fmt
* fmt toml
* add python to macos CI deps
* install python
* fmt
* ci
* clippy
* fix workflow
* fmt
* fix baby nautilus
* fix nautilus sync
* fmt
* fmt
* clippy
* typo
* fix miri
* remove pyo3 from workspace to packages which need it and make it optional
* go back to AsRef<Path> for nautilus grammar loading
* replace hardcoded python flags for macos build
* typo
* taplo fmt
* revert formatting of libafl_qemu_arch
* ci
* typo
* remove expects in NautilusContext::from_file and make them Results
* remove not(miri) clause in test
* try and fix python build fir ios and android
* again
* android
* tmate
* fix android build
* document load_python_grammar
* log if python or json when loading nautilus grammar
* make nautilus optional
* add nautilus as feature to forkserver_simple_nautilus
* Remove emulation_mode env variable and custom cfg
* Using only the feature flag simplifies things a bit and allow the usage of optional dependencies
* Do not use --all-features on libafl_qemu
* Add missing target_os = "linux"
