AFLGo is an extension of American Fuzzy Lop (AFL).
Given a set of target locations (e.g., folder/file.c:582), AFLGo generates inputs specifically with the objective to exercise these target locations.
Unlike AFL, AFLGo spends most of its time budget on reaching specific target locations without wasting resources stressing unrelated program components. This is particularly interesting in the context of
- patch testing by setting changed statements as targets. When a critical component is changed, we would like to check whether this introduced any vulnerabilities. AFLGo, a fuzzer that can focus on those changes, has a higher chance of exposing the regression.
- static analysis report verification by setting statements as targets that a static analysis reports as potentially dangerous or vulnerability-inducing. When assessing the security of a program, static analysis tools might identify dangerous locations, such as critical system calls. AFLGo can generate inputs that actually show that this is indeed no false positive.
- information flow detection by setting sensitive sources and sinks as targets. To expose data leakage vulnerabilities, a security researcher would like to generate executions that exercise sensitive sources containing private information and sensitive sinks where data becomes visible to the outside world. A directed fuzzer can be used to generate such executions efficiently.
- crash reproduction by setting method calls in the stack-trace as targets. When in-field crashes are reported, only the stack-trace and some environmental parameters are sent to the in-house development team. To preserve the user's privacy, the specific crashing input is often not available. AFLGo could help the in-house team to swiftly reproduce these crashes.
AFLGo is based on AFL from Michał Zaleski <lcamtuf@coredump.cx>. Checkout the project awesome-directed-fuzzing for related work on directed greybox/whitebox fuzzing.
The easiest way to use AFLGo is as patch testing tool in OSS-Fuzz. Here is our integration:
- AFLGO_INST_RATIO -- The proportion of basic blocks instrumented with distance values (default: 100).
- AFLGO_SELECTIVE -- Add AFL-trampoline only to basic blocks with distance values? (default: off).
- AFLGO_PROFILING_FILE -- When CFG-tracing is enabled, the data will be stored here.
- Install LLVM 11.0.0 with Gold-plugin. You can also follow these instructions or run AFLGo building script.
- Install other prerequisite
sudo apt-get update
sudo apt-get install python3
sudo apt-get install python3-dev
sudo apt-get install python3-pip
sudo apt-get install libboost-all-dev # boost is not required if you use genDistance.sh in step 7
sudo pip3 install --upgrade pip
sudo pip3 install networkx
sudo pip3 install pydot
sudo pip3 install pydotplus- Compile AFLGo fuzzer, LLVM-instrumentation pass and the distance calculator
# Checkout source code
git clone https://github.com/aflgo/aflgo.git
export AFLGO=$PWD/aflgo
# Compile source code
pushd $AFLGO
make clean all
cd llvm_mode
make clean all
cd ..
cd distance_calculator/
cmake -G Ninja ./
cmake --build ./
popd- Download subject (e.g., libxml2) or just run libxml2 fuzzing script.
# Clone subject repository
git clone https://gitlab.gnome.org/GNOME/libxml2
export SUBJECT=$PWD/libxml2- Set targets (e.g., changed statements in commit ef709ce2). Writes BBtargets.txt.
# Setup directory containing all temporary files
mkdir temp
export TMP_DIR=$PWD/temp
# Download commit-analysis tool
wget https://raw.githubusercontent.com/jay/showlinenum/develop/showlinenum.awk
chmod +x showlinenum.awk
mv showlinenum.awk $TMP_DIR
# Generate BBtargets from commit ef709ce2
pushd $SUBJECT
git checkout ef709ce2
git diff -U0 HEAD^ HEAD > $TMP_DIR/commit.diff
popd
cat $TMP_DIR/commit.diff | $TMP_DIR/showlinenum.awk show_header=0 path=1 | grep -e "\.[ch]:[0-9]*:+" -e "\.cpp:[0-9]*:+" -e "\.cc:[0-9]*:+" | cut -d+ -f1 | rev | cut -c2- | rev > $TMP_DIR/BBtargets.txt
# Print extracted targets.
echo "Targets:"
cat $TMP_DIR/BBtargets.txt- Note: If there are no targets, there is nothing to instrument!
- Generate CG and intra-procedural CFGs from subject (i.e., libxml2).
# Set aflgo-instrumenter
export CC=$AFLGO/afl-clang-fast
export CXX=$AFLGO/afl-clang-fast++
# Set aflgo-instrumentation flags
export COPY_CFLAGS=$CFLAGS
export COPY_CXXFLAGS=$CXXFLAGS
export ADDITIONAL="-targets=$TMP_DIR/BBtargets.txt -outdir=$TMP_DIR -flto -fuse-ld=gold -Wl,-plugin-opt=save-temps"
export CFLAGS="$CFLAGS $ADDITIONAL"
export CXXFLAGS="$CXXFLAGS $ADDITIONAL"
# Build libxml2 (in order to generate CG and CFGs).
# Meanwhile go have a coffee ☕️
export LDFLAGS=-lpthread
pushd $SUBJECT
./autogen.sh
./configure --disable-shared
make clean
make xmllint
popd
# * If the linker (CCLD) complains that you should run ranlib, make
# sure that libLTO.so and LLVMgold.so (from building LLVM with Gold)
# can be found in /usr/lib/bfd-plugins
# * If the compiler crashes, there is some problem with LLVM not
# supporting our instrumentation (afl-llvm-pass.so.cc:540-577).
# LLVM has changed the instrumentation-API very often :(
# -> Check LLVM-version, fix problem, and prepare pull request.
# * You can speed up the compilation with a parallel build. However,
# this may impact which BBs are identified as targets.
# See https://github.com/aflgo/aflgo/issues/41.
# Test whether CG/CFG extraction was successful
$SUBJECT/xmllint --valid --recover $SUBJECT/test/dtd3
ls $TMP_DIR/dot-files
echo "Function targets"
cat $TMP_DIR/Ftargets.txt
# Clean up
cat $TMP_DIR/BBnames.txt | rev | cut -d: -f2- | rev | sort | uniq > $TMP_DIR/BBnames2.txt && mv $TMP_DIR/BBnames2.txt $TMP_DIR/BBnames.txt
cat $TMP_DIR/BBcalls.txt | sort | uniq > $TMP_DIR/BBcalls2.txt && mv $TMP_DIR/BBcalls2.txt $TMP_DIR/BBcalls.txt
# Generate distance ☕️
# $AFLGO/scripts/genDistance.sh is the original, but significantly slower, version
$AFLGO/scripts/gen_distance_fast.py $SUBJECT $TMP_DIR xmllint
# Check distance file
echo "Distance values:"
head -n5 $TMP_DIR/distance.cfg.txt
echo "..."
tail -n5 $TMP_DIR/distance.cfg.txt- Note: If
distance.cfg.txtis empty, there was some problem computing the CG-level and BB-level target distance. See$TMP_DIR/step*. - Instrument subject (i.e., libxml2)
export CFLAGS="$COPY_CFLAGS -distance=$TMP_DIR/distance.cfg.txt"
export CXXFLAGS="$COPY_CXXFLAGS -distance=$TMP_DIR/distance.cfg.txt"
# Clean and build subject with distance instrumentation ☕️
pushd $SUBJECT
make clean
./configure --disable-shared
make xmllint
popdIf your compilation crashes in this step, have a look at Issue #4.
- We set the exponential annealing-based power schedule (-z exp).
- We set the time-to-exploitation to 45min (-c 45m), assuming the fuzzer is run for about an hour.
# Construct seed corpus
mkdir in
cp $SUBJECT/test/dtd* in
cp $SUBJECT/test/dtds/* in
$AFLGO/afl-fuzz -S ef709ce2 -z exp -c 45m -i in -o out $SUBJECT/xmllint --valid --recover @@- Tipp: Concurrently fuzz the most recent version as master with classical AFL :)
$AFL/afl-fuzz -M master -i in -o out $MASTER/xmllint --valid --recover @@- Run more fuzzing scripts of various real programs like Binutils, jasper, lrzip, libming and DARPA CGC.