Merge pull request #820 from AFLplusplus/dev

push to stable

.custom-format.py

@@ -33,13 +33,13 @@ if CLANG_FORMAT_BIN is None:
         o, _ = p.communicate()
         o = str(o, "utf-8")
         o = re.sub(r".*ersion ", "", o)
-        #o = o[len("clang-format version "):].strip()
-        o = o[:o.find(".")]
+        # o = o[len("clang-format version "):].strip()
+        o = o[: o.find(".")]
         o = int(o)
     except:
-        print ("clang-format-11 is needed. Aborted.")
+        print("clang-format-11 is needed. Aborted.")
         exit(1)
-    #if o < 7:
+    # if o < 7:
     #    if subprocess.call(['which', 'clang-format-7'], stdout=subprocess.PIPE) == 0:
     #        CLANG_FORMAT_BIN = 'clang-format-7'
     #    elif subprocess.call(['which', 'clang-format-8'], stdout=subprocess.PIPE) == 0:
@@ -52,8 +52,8 @@ if CLANG_FORMAT_BIN is None:
     #        print ("clang-format 7 or above is needed. Aborted.")
     #        exit(1)
     else:
-        CLANG_FORMAT_BIN = 'clang-format-11'
-            
+        CLANG_FORMAT_BIN = "clang-format-11"
+
 COLUMN_LIMIT = 80
 for line in fmt.split("\n"):
     line = line.split(":")
@@ -69,26 +69,47 @@ def custom_format(filename):
     in_define = False
     last_line = None
     out = ""
-    
+
     for line in src.split("\n"):
         if line.lstrip().startswith("#"):
-            if line[line.find("#")+1:].lstrip().startswith("define"):
+            if line[line.find("#") + 1 :].lstrip().startswith("define"):
                 in_define = True
-        
-        if "/*" in line and not line.strip().startswith("/*") and line.endswith("*/") and len(line) < (COLUMN_LIMIT-2):
+
+        if (
+            "/*" in line
+            and not line.strip().startswith("/*")
+            and line.endswith("*/")
+            and len(line) < (COLUMN_LIMIT - 2)
+        ):
             cmt_start = line.rfind("/*")
-            line = line[:cmt_start] + " " * (COLUMN_LIMIT-2 - len(line)) + line[cmt_start:]
+            line = (
+                line[:cmt_start]
+                + " " * (COLUMN_LIMIT - 2 - len(line))
+                + line[cmt_start:]
+            )
 
         define_padding = 0
         if last_line is not None and in_define and last_line.endswith("\\"):
             last_line = last_line[:-1]
-            define_padding = max(0, len(last_line[last_line.rfind("\n")+1:]))
+            define_padding = max(0, len(last_line[last_line.rfind("\n") + 1 :]))
 
-        if last_line is not None and last_line.strip().endswith("{") and line.strip() != "":
+        if (
+            last_line is not None
+            and last_line.strip().endswith("{")
+            and line.strip() != ""
+        ):
             line = (" " * define_padding + "\\" if in_define else "") + "\n" + line
-        elif last_line is not None and last_line.strip().startswith("}") and line.strip() != "":
+        elif (
+            last_line is not None
+            and last_line.strip().startswith("}")
+            and line.strip() != ""
+        ):
             line = (" " * define_padding + "\\" if in_define else "") + "\n" + line
-        elif line.strip().startswith("}") and last_line is not None and last_line.strip() != "":
+        elif (
+            line.strip().startswith("}")
+            and last_line is not None
+            and last_line.strip() != ""
+        ):
             line = (" " * define_padding + "\\" if in_define else "") + "\n" + line
 
         if not line.endswith("\\"):
@@ -97,14 +118,15 @@ def custom_format(filename):
         out += line + "\n"
         last_line = line
 
-    return (out)
+    return out
+
 
 args = sys.argv[1:]
 if len(args) == 0:
-    print ("Usage: ./format.py [-i] <filename>")
-    print ()
-    print (" The -i option, if specified, let the script to modify in-place")
-    print (" the source files. By default the results are written to stdout.")
+    print("Usage: ./format.py [-i] <filename>")
+    print()
+    print(" The -i option, if specified, let the script to modify in-place")
+    print(" the source files. By default the results are written to stdout.")
     print()
     exit(1)
 
@@ -120,4 +142,3 @@ for filename in args:
             f.write(code)
     else:
         print(code)
-

.github/workflows/build_aflplusplus_docker.yaml

@@ -0,0 +1,27 @@
+name: Publish Docker Images
+on:
+  push:
+    branches: [ stable ]
+    paths:
+    - Dockerfile
+  pull_request:
+    branches: [ stable ]
+    paths:
+    - Dockerfile
+jobs:
+  push_to_registry:
+    name: Push Docker images to Dockerhub
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@master
+    - name: Login to Dockerhub
+      uses: docker/login-action@v1
+      with:
+        username: ${{ secrets.DOCKER_USERNAME }}
+        password: ${{ secrets.DOCKER_TOKEN }}
+    - name: Publish aflpp to Registry
+      uses: docker/build-push-action@v2
+      with:
+        context: .
+        push: true
+        tags: aflplusplus/aflplusplus:latest

.github/workflows/ci.yml

@@ -3,8 +3,8 @@ name: CI
 on:
   push:
     branches: [ stable, dev ]
-  pull_request:
-    branches: [ stable, dev ]
+#  pull_request:
+#    branches: [ stable, dev ]
 
 jobs:
   build:

.github/workflows/codeql-analysis.yml

@@ -3,8 +3,8 @@ name: "CodeQL"
 on:
   push:
     branches: [ stable, dev ]
-  pull_request:
-    branches: [ stable, dev ]
+#  pull_request:
+#    branches: [ stable, dev ]
 
 jobs:
   analyze:

.github/workflows/rust_custom_mutator.yml

@@ -0,0 +1,30 @@
+name: Rust Custom Mutators
+
+on:
+  push:
+    branches: [ stable, dev ]
+  pull_request:
+    branches: [ stable, dev ]
+
+jobs:
+  test:
+    name: Test Rust Custom Mutator Support
+    runs-on: '${{ matrix.os }}'
+    defaults:
+      run:
+        working-directory: custom_mutators/rust
+    strategy:
+      matrix:
+        os: [ubuntu-20.04]
+    steps:
+      - uses: actions/checkout@v2
+      - name: Install Rust Toolchain
+        uses: actions-rs/toolchain@v1
+        with:
+          toolchain: stable
+      - name: Check Code Compiles
+        run: cargo check
+      - name: Run General Tests
+        run: cargo test
+      - name: Run Tests for afl_internals feature flag
+        run: cd custom_mutator && cargo test --features=afl_internals

Dockerfile

@@ -11,6 +11,8 @@ LABEL "about"="AFLplusplus docker image"
 
 ARG DEBIAN_FRONTEND=noninteractive
 
+env NO_ARCH_OPT 1
+
 RUN apt-get update && \
     apt-get -y install --no-install-suggests --no-install-recommends \
     automake \
@@ -48,16 +50,16 @@ RUN update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-10 0
 
 ENV LLVM_CONFIG=llvm-config-12
 ENV AFL_SKIP_CPUFREQ=1
+ENV AFL_I_DONT_CARE_ABOUT_MISSING_CRASHES=1
 
-RUN git clone https://github.com/vanhauser-thc/afl-cov /afl-cov
+RUN git clone --depth=1 https://github.com/vanhauser-thc/afl-cov /afl-cov
 RUN cd /afl-cov && make install && cd ..
 
 COPY . /AFLplusplus
 WORKDIR /AFLplusplus
 
-RUN export REAL_CXX=g++-10 && export CC=gcc-10 && \
-    export CXX=g++-10 && make clean && \
-    make distrib CFLAGS="-O3 -funroll-loops -D_FORTIFY_SOURCE=2" && make install && make clean
+RUN export CC=gcc-10 && export CXX=g++-10 && make clean && \
+    make distrib && make install && make clean
 
 RUN echo 'alias joe="jupp --wordwrap"' >> ~/.bashrc
 RUN echo 'export PS1="[afl++]$PS1"' >> ~/.bashrc

GNUmakefile

@@ -24,7 +24,7 @@ BIN_PATH    = $(PREFIX)/bin
 HELPER_PATH = $(PREFIX)/lib/afl
 DOC_PATH    = $(PREFIX)/share/doc/afl
 MISC_PATH   = $(PREFIX)/share/afl
-MAN_PATH    = $(PREFIX)/man/man8
+MAN_PATH    = $(PREFIX)/share/man/man8
 
 PROGNAME    = afl
 VERSION     = $(shell grep '^$(HASH)define VERSION ' ../config.h | cut -d '"' -f2)
@@ -57,8 +57,6 @@ ifdef MSAN_BUILD
   override LDFLAGS += -fsanitize=memory
 endif
 
-
-
 ifeq "$(findstring android, $(shell $(CC) --version 2>/dev/null))" ""
 ifeq "$(shell echo 'int main() {return 0; }' | $(CC) $(CFLAGS) -Werror -x c - -flto=full -o .test 2>/dev/null && echo 1 || echo 0 ; rm -f .test )" "1"
 	CFLAGS_FLTO ?= -flto=full
@@ -77,17 +75,17 @@ ifeq "$(shell echo 'int main() {return 0; }' | $(CC) -fno-move-loop-invariants -
 	SPECIAL_PERFORMANCE += -fno-move-loop-invariants -fdisable-tree-cunrolli
 endif
 
-ifeq "$(shell echo 'int main() {return 0; }' | $(CC) $(CFLAGS) -Werror -x c - -march=native -o .test 2>/dev/null && echo 1 || echo 0 ; rm -f .test )" "1"
-  ifndef SOURCE_DATE_EPOCH
-    HAVE_MARCHNATIVE = 1
-    CFLAGS_OPT += -march=native
-  endif
-endif
+#ifeq "$(shell echo 'int main() {return 0; }' | $(CC) $(CFLAGS) -Werror -x c - -march=native -o .test 2>/dev/null && echo 1 || echo 0 ; rm -f .test )" "1"
+#  ifndef SOURCE_DATE_EPOCH
+#    HAVE_MARCHNATIVE = 1
+#    CFLAGS_OPT += -march=native
+#  endif
+#endif
 
 ifneq "$(shell uname)" "Darwin"
-  ifeq "$(HAVE_MARCHNATIVE)" "1"
-    SPECIAL_PERFORMANCE += -march=native
-  endif
+  #ifeq "$(HAVE_MARCHNATIVE)" "1"
+  #  SPECIAL_PERFORMANCE += -march=native
+  #endif
  # OS X does not like _FORTIFY_SOURCE=2
   ifndef DEBUG
     CFLAGS_OPT += -D_FORTIFY_SOURCE=2
@@ -519,7 +517,7 @@ code-format:
 ifndef AFL_NO_X86
 test_build: afl-cc afl-gcc afl-as afl-showmap
 	@echo "[*] Testing the CC wrapper afl-cc and its instrumentation output..."
-	@unset AFL_MAP_SIZE AFL_USE_UBSAN AFL_USE_CFISAN AFL_USE_ASAN AFL_USE_MSAN AFL_CC; ASAN_OPTIONS=detect_leaks=0 AFL_INST_RATIO=100 AFL_PATH=. ./afl-cc test-instr.c -o test-instr 2>&1 || (echo "Oops, afl-cc failed"; exit 1 )
+	@unset AFL_MAP_SIZE AFL_USE_UBSAN AFL_USE_CFISAN AFL_USE_ASAN AFL_USE_MSAN; ASAN_OPTIONS=detect_leaks=0 AFL_INST_RATIO=100 AFL_PATH=. ./afl-cc test-instr.c -o test-instr 2>&1 || (echo "Oops, afl-cc failed"; exit 1 )
 	ASAN_OPTIONS=detect_leaks=0 ./afl-showmap -m none -q -o .test-instr0 ./test-instr < /dev/null
 	echo 1 | ASAN_OPTIONS=detect_leaks=0 ./afl-showmap -m none -q -o .test-instr1 ./test-instr
 	@rm -f test-instr

GNUmakefile.llvm

@@ -43,7 +43,8 @@ endif
 LLVMVER  = $(shell $(LLVM_CONFIG) --version 2>/dev/null | sed 's/git//' | sed 's/svn//' )
 LLVM_MAJOR = $(shell $(LLVM_CONFIG) --version 2>/dev/null | sed 's/\..*//' )
 LLVM_MINOR = $(shell $(LLVM_CONFIG) --version 2>/dev/null | sed 's/.*\.//' | sed 's/git//' | sed 's/svn//' | sed 's/ .*//' )
-LLVM_UNSUPPORTED = $(shell $(LLVM_CONFIG) --version 2>/dev/null | egrep -q '^3\.[0-3]|^19' && echo 1 || echo 0 )
+LLVM_UNSUPPORTED = $(shell $(LLVM_CONFIG) --version 2>/dev/null | egrep -q '^3\.[0-3]|^[0-2]\.' && echo 1 || echo 0 )
+LLVM_TOO_NEW = $(shell $(LLVM_CONFIG) --version 2>/dev/null | egrep -q '^1[3-9]' && echo 1 || echo 0 )
 LLVM_NEW_API = $(shell $(LLVM_CONFIG) --version 2>/dev/null | egrep -q '^1[0-9]' && echo 1 || echo 0 )
 LLVM_10_OK = $(shell $(LLVM_CONFIG) --version 2>/dev/null | egrep -q '^1[1-9]|^10\.[1-9]|^10\.0.[1-9]' && echo 1 || echo 0 )
 LLVM_HAVE_LTO = $(shell $(LLVM_CONFIG) --version 2>/dev/null | egrep -q '^1[1-9]' && echo 1 || echo 0 )
@@ -58,7 +59,11 @@ ifeq "$(LLVMVER)" ""
 endif
 
 ifeq "$(LLVM_UNSUPPORTED)" "1"
-  $(warning llvm_mode only supports llvm versions 3.4 up to 12)
+  $(error llvm_mode only supports llvm from version 3.4 onwards)
+endif
+
+ifeq "$(LLVM_TOO_NEW)" "1"
+  $(warning you are using an in-development llvm version - this might break llvm_mode!)
 endif
 
 LLVM_TOO_OLD=1

README.md

@@ -2,11 +2,9 @@
 
   <img align="right" src="https://raw.githubusercontent.com/andreafioraldi/AFLplusplus-website/master/static/logo_256x256.png" alt="AFL++ Logo">
 
-  ![Travis State](https://api.travis-ci.com/AFLplusplus/AFLplusplus.svg?branch=stable)
+  Release Version: [3.11c](https://github.com/AFLplusplus/AFLplusplus/releases)
 
-  Release Version: [3.00c](https://github.com/AFLplusplus/AFLplusplus/releases)
-
-  Github Version: 3.01a
+  Github Version: 3.12a
 
   Repository: [https://github.com/AFLplusplus/AFLplusplus](https://github.com/AFLplusplus/AFLplusplus)
 
@@ -23,11 +21,18 @@
   mutations, more and better instrumentation, custom module support, etc.
 
   If you want to use afl++ for your academic work, check the [papers page](https://aflplus.plus/papers/)
-  on the website.
+  on the website. To cite our work, look at the [Cite](#cite) section.
+  For comparisons use the fuzzbench `aflplusplus` setup, or use `afl-clang-fast`
+  with `AFL_LLVM_CMPLOG=1`.
 
-## Major changes in afl++ 3.0
+## Major changes in afl++ 3.00 + 3.10
 
-With afl++ 3.0 we introduced changes that break some previous afl and afl++
+With afl++ 3.10 we introduced the following changes from previous behaviours:
+  * The '+' feature of the '-t' option now means to  auto-calculate the timeout
+    with the value given being the maximum timeout. The original meaning of
+    "skipping timeouts instead of abort" is now inherent to the -t option.
+
+With afl++ 3.00 we introduced changes that break some previous afl and afl++
 behaviours and defaults:
 
   * There are no llvm_mode and gcc_plugin subdirectories anymore and there is
@@ -170,7 +175,7 @@ If you want to build afl++ yourself you have many options.
 The easiest choice is to build and install everything:
 
 ```shell
-sudo apt install build-essential python3-dev automake flex bison libglib2.0-dev libpixman-1-dev clang python3-setuptools clang llvm llvm-dev libstdc++-dev
+sudo apt install build-essential python3-dev automake flex bison libglib2.0-dev libpixman-1-dev python3-setuptools clang lld llvm llvm-dev libstdc++-dev
 make distrib
 sudo make install
 ```
@@ -221,7 +226,7 @@ These build options exist:
 * AFL_NO_X86 - if compiling on non-intel/amd platforms
 * LLVM_CONFIG - if your distro doesn't use the standard name for llvm-config (e.g. Debian)
 
-e.g.: make ASAN_BUILD=1
+e.g.: `make ASAN_BUILD=1`
 
 ## Good examples and writeups
 
@@ -233,6 +238,7 @@ Here are some good writeups to show how to effectively use AFL++:
  * [https://securitylab.github.com/research/fuzzing-software-2](https://securitylab.github.com/research/fuzzing-software-2)
  * [https://securitylab.github.com/research/fuzzing-sockets-FTP](https://securitylab.github.com/research/fuzzing-sockets-FTP)
  * [https://securitylab.github.com/research/fuzzing-sockets-FreeRDP](https://securitylab.github.com/research/fuzzing-sockets-FreeRDP)
+ * [https://securitylab.github.com/research/fuzzing-apache-1](https://securitylab.github.com/research/fuzzing-apache-1)
 
 If you are interested in fuzzing structured data (where you define what the
 structure is), these links have you covered:
@@ -298,7 +304,7 @@ Clickable README links for the chosen compiler:
   * [LTO mode - afl-clang-lto](instrumentation/README.lto.md)
   * [LLVM mode - afl-clang-fast](instrumentation/README.llvm.md)
   * [GCC_PLUGIN mode - afl-gcc-fast](instrumentation/README.gcc_plugin.md)
-  * GCC/CLANG mode (afl-gcc/afl-clang) have no README as they have no own features
+  * GCC/CLANG modes (afl-gcc/afl-clang) have no README as they have no own features
 
 You can select the mode for the afl-cc compiler by:
   1. use a symlink to afl-cc: afl-gcc, afl-g++, afl-clang, afl-clang++,
@@ -393,10 +399,19 @@ How to do this is described below.
 
 Then build the target. (Usually with `make`)
 
-**NOTE**: sometimes configure and build systems are fickle and do not like
-stderr output (and think this means a test failure) - which is something
-afl++ like to do to show statistics. It is recommended to disable them via
-`export AFL_QUIET=1`.
+**NOTES**
+
+1. sometimes configure and build systems are fickle and do not like
+   stderr output (and think this means a test failure) - which is something
+   afl++ likes to do to show statistics. It is recommended to disable them via
+   `export AFL_QUIET=1`.
+
+2. sometimes configure and build systems error on warnings - these should be
+   disabled (e.g. `--disable-werror` for some configure scripts).
+
+3. in case the configure/build system complains about afl++'s compiler and
+   aborts then set `export AFL_NOOPT=1` which will then just behave like the
+   real compiler. This option has to be unset again before building the target!
 
 ##### configure
 
@@ -478,8 +493,9 @@ default.
 #### c) Minimizing all corpus files
 
 The shorter the input files that still traverse the same path
-within the target, the better the fuzzing will be. This is done with `afl-tmin`
-however it is a long process as this has to be done for every file:
+within the target, the better the fuzzing will be. This minimization
+is done with `afl-tmin` however it is a long process as this has to
+be done for every file:
 
 ```
 mkdir input
@@ -530,12 +546,10 @@ If you need to stop and re-start the fuzzing, use the same command line options
 mutation mode!) and switch the input directory with a dash (`-`):
 `afl-fuzz -i - -o output -- bin/target -d @@`
 
-Note that afl-fuzz enforces memory limits to prevent the system to run out
-of memory. By default this is 50MB for a process. If this is too little for
-the target (which you can usually see by afl-fuzz bailing with the message
-that it could not connect to the forkserver), then you can increase this
-with the `-m` option, the value is in MB. To disable any memory limits
-(beware!) set `-m none` - which is usually required for ASAN compiled targets.
+Memory limits are not enforced by afl-fuzz by default and the system may run
+out of memory. You can decrease the memory with the `-m` option, the value is
+in MB. If this is too small for the target, you can usually see this by
+afl-fuzz bailing with the message that it could not connect to the forkserver.
 
 Adding a dictionary is helpful. See the directory [dictionaries/](dictionaries/) if
 something is already included for your data format, and tell afl-fuzz to load
@@ -548,7 +562,9 @@ afl-fuzz has a variety of options that help to workaround target quirks like
 specific locations for the input file (`-f`), not performing deterministic
 fuzzing (`-d`) and many more. Check out `afl-fuzz -h`.
 
-afl-fuzz never stops fuzzing. To terminate afl++ simply press Control-C.
+By default afl-fuzz never stops fuzzing. To terminate afl++ simply press Control-C
+or send a signal SIGINT. You can limit the number of executions or approximate runtime
+in seconds with options also.
 
 When you start afl-fuzz you will see a user interface that shows what the status
 is:
@@ -730,6 +746,9 @@ campaigns as these are much shorter runnings.
 1. Always:
   * LTO has a much longer compile time which is diametrical to short fuzzing - 
     hence use afl-clang-fast instead.
+  * If you compile with CMPLOG then you can save fuzzing time and reuse that
+    compiled target for both the -c option and the main fuzz target.
+    This will impact the speed by ~15% though.
   * `AFL_FAST_CAL` - Enable fast calibration, this halfs the time the saturated
      corpus needs to be loaded.
   * `AFL_CMPLOG_ONLY_NEW` - only perform cmplog on new found paths, not the
@@ -749,6 +768,10 @@ campaigns as these are much shorter runnings.
   * for CMPLOG targets, 60% for `-l 2`, 40% for `-l 3`
 
 4. Do *not* run any `-M` modes, just running `-S` modes is better for CI fuzzing.
+   `-M` enables deterministic fuzzing, old queue handling etc. which is good for
+   a fuzzing campaign but not good for short CI runs.
+
+How this can look like can e.g. be seen at afl++'s setup in Google's [oss-fuzz](https://github.com/google/oss-fuzz/blob/4bb61df7905c6005000f5766e966e6fe30ab4559/infra/base-images/base-builder/compile_afl#L69).
 
 ## Fuzzing binary-only targets
 
@@ -786,8 +809,7 @@ If [afl-dyninst](https://github.com/vanhauser-thc/afl-dyninst) works for
 your binary, then you can use afl-fuzz normally and it will have twice
 the speed compared to qemu_mode (but slower than persistent mode).
 Note that several other binary rewriters exist, all with their advantages and
-caveats. As rewriting a binary is much faster than Qemu this is a highly
-recommended approach!
+caveats.
 
 ### Unicorn
 
@@ -1163,8 +1185,18 @@ Thank you!
 
 ## Cite
 
+If you use AFLpluplus to compare to your work, please use either `afl-clang-lto`
+or `afl-clang-fast` with `AFL_LLVM_CMPLOG=1` for building targets and
+`afl-fuzz` with the command line option `-l 2` for fuzzing.
+The most effective setup is the `aflplusplus` default configuration on Google's [fuzzbench](https://github.com/google/fuzzbench/tree/master/fuzzers/aflplusplus).
+
 If you use AFLplusplus in scientific work, consider citing [our paper](https://www.usenix.org/conference/woot20/presentation/fioraldi) presented at WOOT'20:
-```
+
++ Andrea Fioraldi, Dominik Maier, Heiko Eißfeldt, and Marc Heuse. “AFL++: Combining incremental steps of fuzzing research”. In 14th USENIX Workshop on Offensive Technologies (WOOT 20). USENIX Association, Aug. 2020.
+
+Bibtex:
+
+```bibtex
 @inproceedings {AFLplusplus-Woot20,
 	author = {Andrea Fioraldi and Dominik Maier and Heiko Ei{\ss}feldt and Marc Heuse},
 	title = {{AFL++}: Combining Incremental Steps of Fuzzing Research},

TODO.md

@@ -2,20 +2,20 @@
 
 ## Roadmap 3.00+
 
- - AFL_MAP_SIZE for qemu_mode and unicorn_mode
  - CPU affinity for many cores? There seems to be an issue > 96 cores
  - afl-plot to support multiple plot_data
  - afl_custom_fuzz_splice_optin()
+ - afl_custom_splice()
  - intel-pt tracer
+ - better autodetection of shifting runtime timeout values
+ - cmplog: use colorization input for havoc?
+ - cmplog: too much tainted bytes, directly add to dict and skip?
+
 
 ## Further down the road
 
 afl-fuzz:
  - setting min_len/max_len/start_offset/end_offset limits for mutation output
- - add __sanitizer_cov_trace_cmp* support via shmem
-
-llvm_mode:
- - add __sanitizer_cov_trace_cmp* support
 
 qemu_mode:
  - non colliding instrumentation
@@ -26,9 +26,12 @@ qemu_mode:
  - add/implement AFL_QEMU_INST_LIBLIST and AFL_QEMU_NOINST_PROGRAM
  - add/implement AFL_QEMU_INST_REGIONS as a list of _START/_END addresses
 
+
 ## Ideas
 
  - LTO/sancov: write current edge to prev_loc and use that information when
    using cmplog or __sanitizer_cov_trace_cmp*. maybe we can deduct by follow
    up edge numbers that both following cmp paths have been found and then
    disable working on this edge id -> cmplog_intelligence branch
+ - use cmplog colorization taint result for havoc locations?
+

afl-cmin

@@ -287,6 +287,10 @@ BEGIN {
     exit 1
   }
 
+  if (0 == system( "test -d "in_dir"/default" )) {
+    in_dir = in_dir "/default"
+  }
+
   if (0 == system( "test -d "in_dir"/queue" )) {
     in_dir = in_dir "/queue"
   }
@@ -342,8 +346,10 @@ BEGIN {
   } else {
     stat_format = "-f '%z %N'" # *BSD, MacOS
   }
-  cmdline = "cd "in_dir" && find . \\( ! -name . -a -type d -prune \\) -o -type f -exec stat "stat_format" \\{\\} \\; | sort -k1n -k2r"
-  cmdline = "ls "in_dir" | (cd "in_dir" && xargs stat "stat_format" 2>/dev/null) | sort -k1n -k2r"
+  cmdline = "(cd "in_dir" && find . \\( ! -name . -a -type d -prune \\) -o -type f -exec stat "stat_format" \\{\\} + | sort -k1n -k2r)"
+  #cmdline = "ls "in_dir" | (cd "in_dir" && xargs stat "stat_format" 2>/dev/null) | sort -k1n -k2r"
+  #cmdline = "(cd "in_dir" && stat "stat_format" *) | sort -k1n -k2r"
+  #cmdline = "(cd "in_dir" && ls | xargs stat "stat_format" ) | sort -k1n -k2r"
   while (cmdline | getline) {
     sub(/^[0-9]+ (\.\/)?/,"",$0)
     infilesSmallToBig[i++] = $0
@@ -354,12 +360,12 @@ BEGIN {
   
   # Make sure that we're not dealing with a directory.
 
-  if (0 == system("test -d "in_dir"/"first_file)) {
+  if (0 == system("test -d ""\""in_dir"/"first_file"\"")) {
     print "[-] Error: The input directory is empty or contains subdirectories - please fix." > "/dev/stderr"
     exit 1
   }
 
-  if (0 == system("ln "in_dir"/"first_file" "trace_dir"/.link_test")) {
+  if (0 == system("ln \""in_dir"/"first_file"\" "trace_dir"/.link_test")) {
     cp_tool = "ln"
   } else {
     cp_tool = "cp"
@@ -374,7 +380,7 @@ BEGIN {
     if (!stdin_file) {
       system( "AFL_CMIN_ALLOW_ANY=1 "AFL_CMIN_CRASHES_ONLY"\""showmap"\" -m "mem_limit" -t "timeout" -o \""trace_dir"/.run_test\" -Z "extra_par" -- \""target_bin"\" "prog_args_string" <\""in_dir"/"first_file"\"")
     } else {
-      system("cp "in_dir"/"first_file" "stdin_file)
+      system("cp \""in_dir"/"first_file"\" "stdin_file)
       system( "AFL_CMIN_ALLOW_ANY=1 "AFL_CMIN_CRASHES_ONLY"\""showmap"\" -m "mem_limit" -t "timeout" -o \""trace_dir"/.run_test\" -Z "extra_par" -A \""stdin_file"\" -- \""target_bin"\" "prog_args_string" </dev/null")
     }
 
@@ -411,8 +417,8 @@ BEGIN {
     retval = system( AFL_CMIN_CRASHES_ONLY"\""showmap"\" -m "mem_limit" -t "timeout" -o \""trace_dir"\" -Z "extra_par" -i \""in_dir"\" -- \""target_bin"\" "prog_args_string)
   } else {
     print "    Processing "in_count" files (forkserver mode)..."
-#    print AFL_CMIN_CRASHES_ONLY"\""showmap"\" -m "mem_limit" -t "timeout" -o \""trace_dir"\" -Z "extra_par" -i \""in_dir"\" -- \""target_bin"\" "prog_args_string" </dev/null"
-    retval = system( AFL_CMIN_CRASHES_ONLY"\""showmap"\" -m "mem_limit" -t "timeout" -o \""trace_dir"\" -Z "extra_par" -i \""in_dir"\" -- \""target_bin"\" "prog_args_string" </dev/null")
+#    print AFL_CMIN_CRASHES_ONLY"\""showmap"\" -m "mem_limit" -t "timeout" -o \""trace_dir"\" -Z "extra_par" -i \""in_dir"\" -A \""stdin_file"\" -- \""target_bin"\" "prog_args_string" </dev/null"
+    retval = system( AFL_CMIN_CRASHES_ONLY"\""showmap"\" -m "mem_limit" -t "timeout" -o \""trace_dir"\" -Z "extra_par" -i \""in_dir"\" -A \""stdin_file"\" -- \""target_bin"\" "prog_args_string" </dev/null")
   }
 
   if (retval && !AFL_CMIN_CRASHES_ONLY) {
@@ -496,7 +502,7 @@ BEGIN {
 
     # copy file unless already done
     if (! (fn in file_already_copied)) {
-      system(cp_tool" "in_dir"/"fn" "out_dir"/"fn)
+      system(cp_tool" \""in_dir"/"fn"\" \""out_dir"/"fn"\"")
       file_already_copied[fn] = ""
       ++out_count
       #printf "tuple nr %d (%d cnt=%d) -> %s\n",tcnt,key,key_count[key],fn > trace_dir"/.log"

afl-cmin.bash

@@ -223,6 +223,7 @@ if [ ! -d "$IN_DIR" ]; then
   exit 1
 fi
 
+test -d "$IN_DIR/default" && IN_DIR="$IN_DIR/default"
 test -d "$IN_DIR/queue" && IN_DIR="$IN_DIR/queue"
 
 find "$OUT_DIR" -name 'id[:_]*' -maxdepth 1 -exec rm -- {} \; 2>/dev/null

afl-plot

@@ -99,7 +99,7 @@ if [ ! -d "$outputdir" ]; then
 
 fi
 
-rm -f "$outputdir/high_freq.png" "$outputdir/low_freq.png" "$outputdir/exec_speed.png"
+rm -f "$outputdir/high_freq.png" "$outputdir/low_freq.png" "$outputdir/exec_speed.png" "$outputdir/edges.png"
 mv -f "$outputdir/index.html" "$outputdir/index.html.orig" 2>/dev/null
 
 echo "[*] Generating plots..."
@@ -152,6 +152,12 @@ set ytics auto
 plot '$inputdir/plot_data' using 1:11 with filledcurve x1 title '' linecolor rgb '#0090ff' fillstyle transparent solid 0.2 noborder, \\
      '$inputdir/plot_data' using 1:11 with lines title '    execs/sec' linecolor rgb '#0090ff' linewidth 3 smooth bezier;
 
+set terminal png truecolor enhanced size 1000,300 butt
+set output '$outputdir/edges.png'
+
+set ytics auto
+plot '$inputdir/plot_data' using 1:13 with lines title '        edges' linecolor rgb '#0090ff' linewidth 3
+
 _EOF_
 
 ) | gnuplot 
@@ -172,6 +178,7 @@ cat >"$outputdir/index.html" <<_EOF_
 <tr><td><b>Generated on:</b></td><td>`date`</td></tr>
 </table>
 <p>
+<img src="edges.png" width=1000 height=300>
 <img src="high_freq.png" width=1000 height=300><p>
 <img src="low_freq.png" width=1000 height=200><p>
 <img src="exec_speed.png" width=1000 height=200>
@@ -183,7 +190,7 @@ _EOF_
 # sensitive, this seems like a reasonable trade-off.
 
 chmod 755 "$outputdir"
-chmod 644 "$outputdir/high_freq.png" "$outputdir/low_freq.png" "$outputdir/exec_speed.png" "$outputdir/index.html"
+chmod 644 "$outputdir/high_freq.png" "$outputdir/low_freq.png" "$outputdir/exec_speed.png" "$outputdir/edges.png" "$outputdir/index.html"
 
 echo "[+] All done - enjoy your charts!"
 

afl-system-config

@@ -39,7 +39,7 @@ if [ "$PLATFORM" = "Linux" ] ; then
   echo Settings applied.
   dmesg | egrep -q 'nospectre_v2|spectre_v2=off' || {
     echo It is recommended to boot the kernel with lots of security off - if you are running a machine that is in a secured network - so set this:
-    echo '  /etc/default/grub:GRUB_CMDLINE_LINUX_DEFAULT="ibpb=off ibrs=off kpti=off l1tf=off mds=off mitigations=off no_stf_barrier noibpb noibrs nopcid nopti nospec_store_bypass_disable nospectre_v1 nospectre_v2 pcid=off pti=off spec_store_bypass_disable=off spectre_v2=off stf_barrier=off"'
+    echo '  /etc/default/grub:GRUB_CMDLINE_LINUX_DEFAULT="ibpb=off ibrs=off kpti=0 l1tf=off mds=off mitigations=off no_stf_barrier noibpb noibrs nopcid nopti nospec_store_bypass_disable nospectre_v1 nospectre_v2 pcid=off pti=off spec_store_bypass_disable=off spectre_v2=off stf_barrier=off srbds=off noexec=off noexec32=off tsx=on tsx_async_abort=off arm64.nopauth audit=0 hardened_usercopy=off ssbd=force-off"'
   }
   DONE=1
 fi
@@ -49,6 +49,12 @@ if [ "$PLATFORM" = "FreeBSD" ] ; then
   sysctl kern.elf64.aslr.enable=0
 } > /dev/null
   echo Settings applied.
+  cat <<EOF
+In order to suppress core file generation during fuzzing it is recommended to set
+me:\\
+	:coredumpsize=0:
+in the ~/.login_conf file for the user used for fuzzing.
+EOF
   echo It is recommended to boot the kernel with lots of security off - if you are running a machine that is in a secured network - so set this:
   echo '  sysctl hw.ibrs_disable=1'
   echo 'Setting kern.pmap.pg_ps_enabled=0 into /boot/loader.conf might be helpful too.'
@@ -60,8 +66,14 @@ if [ "$PLATFORM" = "OpenBSD" ] ; then
   DONE=1
 fi
 if [ "$PLATFORM" = "DragonFly" ] ; then
-  echo
-  echo 'System security features cannot be disabled on DragonFly.'
+  #/sbin/sysctl kern.corefile=/dev/null
+  #echo Settings applied.
+  cat <<EOF
+In order to suppress core file generation during fuzzing it is recommended to set
+me:\\
+	:coredumpsize=0:
+in the ~/.login_conf file for the user used for fuzzing.
+EOF
   DONE=1
 fi
 if [ "$PLATFORM" = "NetBSD" ] ; then
@@ -88,7 +100,7 @@ fi
 if [ "$PLATFORM" = "Haiku" ] ; then
   SETTINGS=~/config/settings/system/debug_server/settings
   [ -r ${SETTINGS} ] && grep -qE "default_action\s+kill" ${SETTINGS} && { echo "Nothing to do"; } || { \
-    echo We change the debug_server default_action from user to silenty kill; \
+    echo We change the debug_server default_action from user to silently kill; \
     [ ! -r ${SETTINGS} ] && echo "default_action kill" >${SETTINGS} || { mv ${SETTINGS} s.tmp; sed -e "s/default_action\s\s*user/default_action kill/" s.tmp > ${SETTINGS}; rm s.tmp; }; \
     echo Settings applied.; \
   }

custom_mutators/honggfuzz/honggfuzz.h

@@ -39,7 +39,7 @@
 #include "libhfcommon/util.h"
 
 #define PROG_NAME    "honggfuzz"
-#define PROG_VERSION "2.3"
+#define PROG_VERSION "2.4"
 
 /* Name of the template which will be replaced with the proper name of the file */
 #define _HF_FILE_PLACEHOLDER "___FILE___"
@@ -208,6 +208,7 @@ typedef struct {
         const char* crashDir;
         const char* covDirNew;
         bool        saveUnique;
+        bool        saveSmaller;
         size_t      dynfileqMaxSz;
         size_t      dynfileqCnt;
         dynfile_t*  dynfileqCurrent;
@@ -279,9 +280,9 @@ typedef struct {
         cmpfeedback_t*  cmpFeedbackMap;
         int             cmpFeedbackFd;
         bool            cmpFeedback;
-        const char*     blacklistFile;
-        uint64_t*       blacklist;
-        size_t          blacklistCnt;
+        const char*     blocklistFile;
+        uint64_t*       blocklist;
+        size_t          blocklistCnt;
         bool            skipFeedbackOnTimeout;
         uint64_t        maxCov[4];
         dynFileMethod_t dynFileMethod;

custom_mutators/honggfuzz/input.h

@@ -77,11 +77,11 @@ static inline uint64_t util_rndGet(uint64_t min, uint64_t max) {
 }
 static inline uint64_t util_rnd64() { return rand_below(afl_struct, 1 << 30); }
 
-static inline size_t input_getRandomInputAsBuf(run_t *run, const uint8_t **buf) {
-  *buf = queue_input;
+static inline const uint8_t* input_getRandomInputAsBuf(run_t* run, size_t* len) {
+  *len = queue_input_size;
   run->dynfile->data = queue_input;
   run->dynfile->size = queue_input_size;
-  return queue_input_size;
+  return queue_input;
 }
 static inline void input_setSize(run_t* run, size_t sz) {
   run->dynfile->size = sz;

custom_mutators/honggfuzz/libhfcommon

@@ -1 +0,0 @@
-.

custom_mutators/honggfuzz/libhfcommon/common.h

@@ -0,0 +1,3 @@
+#ifndef LOG_E
+  #define LOG_E LOG_F
+#endif

custom_mutators/rust/.gitignore

@@ -0,0 +1,10 @@
+# Generated by Cargo
+# will have compiled files and executables
+/target/
+
+# Remove Cargo.lock from gitignore if creating an executable, leave it for libraries
+# More information here https://doc.rust-lang.org/cargo/guide/cargo-toml-vs-cargo-lock.html
+Cargo.lock
+
+# These are backup files generated by rustfmt
+**/*.rs.bk

custom_mutators/rust/Cargo.toml

@@ -0,0 +1,8 @@
+[workspace]
+members = [
+    "custom_mutator-sys",
+    "custom_mutator",
+    "example",
+    # Lain needs a nightly toolchain
+    # "example_lain",
+]

custom_mutators/rust/README.md

@@ -0,0 +1,11 @@
+# Rust Custom Mutators
+
+Bindings to create custom mutators in Rust.
+
+These bindings are documented with rustdoc. To view the documentation run
+```cargo doc -p custom_mutator --open```.
+
+A minimal example can be found in `example`. Build it using `cargo build --example example_mutator`. 
+
+An example using [lain](https://github.com/microsoft/lain) for structured fuzzing can be found in `example_lain`.
+Since lain requires a nightly rust toolchain, you need to set one up before you can play with it.

custom_mutators/rust/custom_mutator-sys/Cargo.toml

@@ -0,0 +1,12 @@
+[package]
+name = "custom_mutator-sys"
+version = "0.1.0"
+authors = ["Julius Hohnerlein <julihoh@users.noreply.github.com>"]
+edition = "2018"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[dependencies]
+
+[build-dependencies]
+bindgen = "0.56"

custom_mutators/rust/custom_mutator-sys/build.rs

@@ -0,0 +1,42 @@
+extern crate bindgen;
+
+use std::env;
+use std::path::PathBuf;
+
+// this code is largely taken straight from the handbook: https://github.com/fitzgen/bindgen-tutorial-bzip2-sys
+fn main() {
+    // Tell cargo to invalidate the built crate whenever the wrapper changes
+    println!("cargo:rerun-if-changed=wrapper.h");
+
+    // The bindgen::Builder is the main entry point
+    // to bindgen, and lets you build up options for
+    // the resulting bindings.
+    let bindings = bindgen::Builder::default()
+        // The input header we would like to generate
+        // bindings for.
+        .header("wrapper.h")
+        .whitelist_type("afl_state_t")
+        .blacklist_type(r"u\d+")
+        .opaque_type(r"_.*")
+        .opaque_type("FILE")
+        .opaque_type("in_addr(_t)?")
+        .opaque_type("in_port(_t)?")
+        .opaque_type("sa_family(_t)?")
+        .opaque_type("sockaddr_in(_t)?")
+        .opaque_type("time_t")
+        .rustfmt_bindings(true)
+        .size_t_is_usize(true)
+        // Tell cargo to invalidate the built crate whenever any of the
+        // included header files changed.
+        .parse_callbacks(Box::new(bindgen::CargoCallbacks))
+        // Finish the builder and generate the bindings.
+        .generate()
+        // Unwrap the Result and panic on failure.
+        .expect("Unable to generate bindings");
+
+    // Write the bindings to the $OUT_DIR/bindings.rs file.
+    let out_path = PathBuf::from(env::var("OUT_DIR").unwrap());
+    bindings
+        .write_to_file(out_path.join("bindings.rs"))
+        .expect("Couldn't write bindings!");
+}

custom_mutators/rust/custom_mutator-sys/src/lib.rs

@@ -0,0 +1,5 @@
+#![allow(non_upper_case_globals)]
+#![allow(non_camel_case_types)]
+#![allow(non_snake_case)]
+
+include!(concat!(env!("OUT_DIR"), "/bindings.rs"));

custom_mutators/rust/custom_mutator-sys/wrapper.h

@@ -0,0 +1,4 @@
+#include "../../../include/afl-fuzz.h"
+#include "../../../include/common.h"
+#include "../../../include/config.h"
+#include "../../../include/debug.h"

custom_mutators/rust/custom_mutator/Cargo.toml

@@ -0,0 +1,13 @@
+[package]
+name = "custom_mutator"
+version = "0.1.0"
+authors = ["Julius Hohnerlein <julihoh@users.noreply.github.com>"]
+edition = "2018"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[features]
+afl_internals = ["custom_mutator-sys"]
+
+[dependencies]
+custom_mutator-sys = { path = "../custom_mutator-sys", optional=true }

custom_mutators/rust/custom_mutator/src/lib.rs

@@ -0,0 +1,740 @@
+#![cfg(unix)]
+//! Somewhat safe and somewhat ergonomic bindings for creating [AFL++](https://github.com/AFLplusplus/AFLplusplus) [custom mutators](https://github.com/AFLplusplus/AFLplusplus/blob/stable/docs/custom_mutators.md) in Rust.
+//!
+//! # Usage
+//! AFL++ custom mutators are expected to be dynamic libraries which expose a set of symbols.
+//! Check out [`CustomMutator`] to see which functions of the API are supported.
+//! Then use [`export_mutator`] to export the correct symbols for your mutator.
+//! In order to use the mutator, your crate needs to be a library crate and have a `crate-type` of `cdylib`.
+//! Putting
+//! ```yaml
+//! [lib]
+//! crate-type = ["cdylib"]
+//! ```
+//! into your `Cargo.toml` should do the trick.
+//! The final executable can be found in `target/(debug|release)/your_crate_name.so`.
+//! # Example
+//! See [`export_mutator`] for an example.
+//!
+//! # On `panic`s
+//! This binding is panic-safe in that it will prevent panics from unwinding into AFL++. Any panic will `abort` at the boundary between the custom mutator and AFL++.
+//!
+//! # Access to AFL++ internals
+//! This crate has an optional feature "afl_internals", which gives access to AFL++'s internal state.
+//! The state is passed to [`CustomMutator::init`], when the feature is activated.
+//!
+//! _This is completely unsafe and uses automatically generated types extracted from the AFL++ source._
+use std::{fmt::Debug, path::Path};
+
+#[cfg(feature = "afl_internals")]
+#[doc(hidden)]
+pub use custom_mutator_sys::afl_state;
+
+#[allow(unused_variables)]
+#[doc(hidden)]
+pub trait RawCustomMutator {
+    #[cfg(feature = "afl_internals")]
+    fn init(afl: &'static afl_state, seed: u32) -> Self
+    where
+        Self: Sized;
+    #[cfg(not(feature = "afl_internals"))]
+    fn init(seed: u32) -> Self
+    where
+        Self: Sized;
+
+    fn fuzz<'b, 's: 'b>(
+        &'s mut self,
+        buffer: &'b mut [u8],
+        add_buff: Option<&[u8]>,
+        max_size: usize,
+    ) -> Option<&'b [u8]>;
+
+    fn fuzz_count(&mut self, buffer: &[u8]) -> u32 {
+        1
+    }
+
+    fn queue_new_entry(&mut self, filename_new_queue: &Path, _filename_orig_queue: Option<&Path>) {}
+
+    fn queue_get(&mut self, filename: &Path) -> bool {
+        true
+    }
+
+    fn describe(&mut self, max_description: usize) -> Option<&str> {
+        Some(default_mutator_describe::<Self>(max_description))
+    }
+
+    fn introspection(&mut self) -> Option<&str> {
+        None
+    }
+
+    /*fn post_process(&self, buffer: &[u8], unsigned char **out_buf)-> usize;
+    int afl_custom_init_trim(&self, buffer: &[u8]);
+    size_t afl_custom_trim(&self, unsigned char **out_buf);
+    int afl_custom_post_trim(&self, unsigned char success);
+    size_t afl_custom_havoc_mutation(&self, buffer: &[u8], unsigned char **out_buf, size_t max_size);
+    unsigned char afl_custom_havoc_mutation_probability(&self);*/
+}
+
+/// Wrappers for the custom mutator which provide the bridging between the C API and CustomMutator.
+/// These wrappers are not intended to be used directly, rather export_mutator will use them to publish the custom mutator C API.
+#[doc(hidden)]
+pub mod wrappers {
+    #[cfg(feature = "afl_internals")]
+    use custom_mutator_sys::afl_state;
+
+    use std::{
+        any::Any,
+        convert::TryInto,
+        ffi::{c_void, CStr, OsStr},
+        mem::ManuallyDrop,
+        os::{raw::c_char, unix::ffi::OsStrExt},
+        panic::catch_unwind,
+        path::Path,
+        process::abort,
+        ptr::null,
+        slice,
+    };
+
+    use crate::RawCustomMutator;
+
+    /// A structure to be used as the data pointer for our custom mutator. This was used as additional storage and is kept for now in case its needed later.
+    /// Also has some convenience functions for FFI conversions (from and to ptr) and tries to make misuse hard (see [`FFIContext::from`]).
+    struct FFIContext<M: RawCustomMutator> {
+        mutator: M,
+        /// buffer for storing the description returned by [`RawCustomMutator::describe`] as a CString
+        description_buffer: Vec<u8>,
+        /// buffer for storing the introspection returned by [`RawCustomMutator::introspect`] as a CString
+        introspection_buffer: Vec<u8>,
+    }
+
+    impl<M: RawCustomMutator> FFIContext<M> {
+        fn from(ptr: *mut c_void) -> ManuallyDrop<Box<Self>> {
+            assert!(!ptr.is_null());
+            ManuallyDrop::new(unsafe { Box::from_raw(ptr as *mut Self) })
+        }
+
+        fn into_ptr(self: Box<Self>) -> *const c_void {
+            Box::into_raw(self) as *const c_void
+        }
+
+        #[cfg(feature = "afl_internals")]
+        fn new(afl: &'static afl_state, seed: u32) -> Box<Self> {
+            Box::new(Self {
+                mutator: M::init(afl, seed),
+                description_buffer: Vec::new(),
+                introspection_buffer: Vec::new(),
+            })
+        }
+        #[cfg(not(feature = "afl_internals"))]
+        fn new(seed: u32) -> Box<Self> {
+            Box::new(Self {
+                mutator: M::init(seed),
+                description_buffer: Vec::new(),
+                introspection_buffer: Vec::new(),
+            })
+        }
+    }
+
+    /// panic handler called for every panic
+    fn panic_handler(method: &str, panic_info: Box<dyn Any + Send + 'static>) -> ! {
+        use std::ops::Deref;
+        let cause = panic_info
+            .downcast_ref::<String>()
+            .map(String::deref)
+            .unwrap_or_else(|| {
+                panic_info
+                    .downcast_ref::<&str>()
+                    .copied()
+                    .unwrap_or("<cause unknown>")
+            });
+        eprintln!("A panic occurred at {}: {}", method, cause);
+        abort()
+    }
+
+    /// Internal function used in the macro
+    #[cfg(not(feature = "afl_internals"))]
+    pub fn afl_custom_init_<M: RawCustomMutator>(seed: u32) -> *const c_void {
+        match catch_unwind(|| FFIContext::<M>::new(seed).into_ptr()) {
+            Ok(ret) => ret,
+            Err(err) => panic_handler("afl_custom_init", err),
+        }
+    }
+
+    /// Internal function used in the macro
+    #[cfg(feature = "afl_internals")]
+    pub fn afl_custom_init_<M: RawCustomMutator>(
+        afl: Option<&'static afl_state>,
+        seed: u32,
+    ) -> *const c_void {
+        match catch_unwind(|| {
+            let afl = afl.expect("mutator func called with NULL afl");
+            FFIContext::<M>::new(afl, seed).into_ptr()
+        }) {
+            Ok(ret) => ret,
+            Err(err) => panic_handler("afl_custom_init", err),
+        }
+    }
+
+    /// Internal function used in the macro
+    pub unsafe fn afl_custom_fuzz_<M: RawCustomMutator>(
+        data: *mut c_void,
+        buf: *mut u8,
+        buf_size: usize,
+        out_buf: *mut *const u8,
+        add_buf: *mut u8,
+        add_buf_size: usize,
+        max_size: usize,
+    ) -> usize {
+        match catch_unwind(|| {
+            let mut context = FFIContext::<M>::from(data);
+            if buf.is_null() {
+                panic!("null buf passed to afl_custom_fuzz")
+            }
+            if out_buf.is_null() {
+                panic!("null out_buf passed to afl_custom_fuzz")
+            }
+            let buff_slice = slice::from_raw_parts_mut(buf, buf_size);
+            let add_buff_slice = if add_buf.is_null() {
+                None
+            } else {
+                Some(slice::from_raw_parts(add_buf, add_buf_size))
+            };
+            match context
+                .mutator
+                .fuzz(buff_slice, add_buff_slice, max_size.try_into().unwrap())
+            {
+                Some(buffer) => {
+                    *out_buf = buffer.as_ptr();
+                    buffer.len().try_into().unwrap()
+                }
+                None => {
+                    // return the input buffer with 0-length to let AFL skip this mutation attempt
+                    *out_buf = buf;
+                    0
+                }
+            }
+        }) {
+            Ok(ret) => ret,
+            Err(err) => panic_handler("afl_custom_fuzz", err),
+        }
+    }
+
+    /// Internal function used in the macro
+    pub unsafe fn afl_custom_fuzz_count_<M: RawCustomMutator>(
+        data: *mut c_void,
+        buf: *const u8,
+        buf_size: usize,
+    ) -> u32 {
+        match catch_unwind(|| {
+            let mut context = FFIContext::<M>::from(data);
+            if buf.is_null() {
+                panic!("null buf passed to afl_custom_fuzz")
+            }
+            let buf_slice = slice::from_raw_parts(buf, buf_size);
+            // see https://doc.rust-lang.org/nomicon/borrow-splitting.html
+            let ctx = &mut **context;
+            let mutator = &mut ctx.mutator;
+            mutator.fuzz_count(buf_slice)
+        }) {
+            Ok(ret) => ret,
+            Err(err) => panic_handler("afl_custom_fuzz_count", err),
+        }
+    }
+
+    /// Internal function used in the macro
+    pub fn afl_custom_queue_new_entry_<M: RawCustomMutator>(
+        data: *mut c_void,
+        filename_new_queue: *const c_char,
+        filename_orig_queue: *const c_char,
+    ) {
+        match catch_unwind(|| {
+            let mut context = FFIContext::<M>::from(data);
+            if filename_new_queue.is_null() {
+                panic!("received null filename_new_queue in afl_custom_queue_new_entry");
+            }
+            let filename_new_queue = Path::new(OsStr::from_bytes(
+                unsafe { CStr::from_ptr(filename_new_queue) }.to_bytes(),
+            ));
+            let filename_orig_queue = if !filename_orig_queue.is_null() {
+                Some(Path::new(OsStr::from_bytes(
+                    unsafe { CStr::from_ptr(filename_orig_queue) }.to_bytes(),
+                )))
+            } else {
+                None
+            };
+            context
+                .mutator
+                .queue_new_entry(filename_new_queue, filename_orig_queue);
+        }) {
+            Ok(ret) => ret,
+            Err(err) => panic_handler("afl_custom_queue_new_entry", err),
+        }
+    }
+
+    /// Internal function used in the macro
+    pub unsafe fn afl_custom_deinit_<M: RawCustomMutator>(data: *mut c_void) {
+        match catch_unwind(|| {
+            // drop the context
+            ManuallyDrop::into_inner(FFIContext::<M>::from(data));
+        }) {
+            Ok(ret) => ret,
+            Err(err) => panic_handler("afl_custom_deinit", err),
+        }
+    }
+
+    /// Internal function used in the macro
+    pub fn afl_custom_introspection_<M: RawCustomMutator>(data: *mut c_void) -> *const c_char {
+        match catch_unwind(|| {
+            let context = &mut *FFIContext::<M>::from(data);
+            if let Some(res) = context.mutator.introspection() {
+                let buf = &mut context.introspection_buffer;
+                buf.clear();
+                buf.extend_from_slice(res.as_bytes());
+                buf.push(0);
+                // unwrapping here, as the error case should be extremely rare
+                CStr::from_bytes_with_nul(&buf).unwrap().as_ptr()
+            } else {
+                null()
+            }
+        }) {
+            Ok(ret) => ret,
+            Err(err) => panic_handler("afl_custom_introspection", err),
+        }
+    }
+
+    /// Internal function used in the macro
+    pub fn afl_custom_describe_<M: RawCustomMutator>(
+        data: *mut c_void,
+        max_description_len: usize,
+    ) -> *const c_char {
+        match catch_unwind(|| {
+            let context = &mut *FFIContext::<M>::from(data);
+            if let Some(res) = context.mutator.describe(max_description_len) {
+                let buf = &mut context.description_buffer;
+                buf.clear();
+                buf.extend_from_slice(res.as_bytes());
+                buf.push(0);
+                // unwrapping here, as the error case should be extremely rare
+                CStr::from_bytes_with_nul(&buf).unwrap().as_ptr()
+            } else {
+                null()
+            }
+        }) {
+            Ok(ret) => ret,
+            Err(err) => panic_handler("afl_custom_describe", err),
+        }
+    }
+
+    /// Internal function used in the macro
+    pub fn afl_custom_queue_get_<M: RawCustomMutator>(
+        data: *mut c_void,
+        filename: *const c_char,
+    ) -> u8 {
+        match catch_unwind(|| {
+            let mut context = FFIContext::<M>::from(data);
+            assert!(!filename.is_null());
+
+            context.mutator.queue_get(Path::new(OsStr::from_bytes(
+                unsafe { CStr::from_ptr(filename) }.to_bytes(),
+            ))) as u8
+        }) {
+            Ok(ret) => ret,
+            Err(err) => panic_handler("afl_custom_queue_get", err),
+        }
+    }
+}
+
+/// exports the given Mutator as a custom mutator as the C interface that AFL++ expects.
+/// It is not possible to call this macro multiple times, because it would define the custom mutator symbols multiple times.
+/// # Example
+/// ```
+/// # #[macro_use] extern crate custom_mutator;
+/// # #[cfg(feature = "afl_internals")]
+/// # use custom_mutator::afl_state;
+/// # use custom_mutator::CustomMutator;
+/// struct MyMutator;
+/// impl CustomMutator for MyMutator {
+///     /// ...
+/// #  type Error = ();
+/// #  #[cfg(feature = "afl_internals")]
+/// #  fn init(_afl_state: &afl_state, _seed: u32) -> Result<Self,()> {unimplemented!()}
+/// #  #[cfg(not(feature = "afl_internals"))]
+/// #  fn init(_seed: u32) -> Result<Self, Self::Error> {unimplemented!()}
+/// #  fn fuzz<'b,'s:'b>(&'s mut self, _buffer: &'b mut [u8], _add_buff: Option<&[u8]>, _max_size: usize) -> Result<Option<&'b [u8]>, Self::Error> {unimplemented!()}
+/// }
+/// export_mutator!(MyMutator);
+/// ```
+#[macro_export]
+macro_rules! export_mutator {
+    ($mutator_type:ty) => {
+        #[cfg(feature = "afl_internals")]
+        #[no_mangle]
+        pub extern "C" fn afl_custom_init(
+            afl: ::std::option::Option<&'static $crate::afl_state>,
+            seed: ::std::os::raw::c_uint,
+        ) -> *const ::std::os::raw::c_void {
+            $crate::wrappers::afl_custom_init_::<$mutator_type>(afl, seed as u32)
+        }
+
+        #[cfg(not(feature = "afl_internals"))]
+        #[no_mangle]
+        pub extern "C" fn afl_custom_init(
+            _afl: *const ::std::os::raw::c_void,
+            seed: ::std::os::raw::c_uint,
+        ) -> *const ::std::os::raw::c_void {
+            $crate::wrappers::afl_custom_init_::<$mutator_type>(seed as u32)
+        }
+
+        #[no_mangle]
+        pub extern "C" fn afl_custom_fuzz_count(
+            data: *mut ::std::os::raw::c_void,
+            buf: *const u8,
+            buf_size: usize,
+        ) -> u32 {
+            unsafe {
+                $crate::wrappers::afl_custom_fuzz_count_::<$mutator_type>(data, buf, buf_size)
+            }
+        }
+
+        #[no_mangle]
+        pub extern "C" fn afl_custom_fuzz(
+            data: *mut ::std::os::raw::c_void,
+            buf: *mut u8,
+            buf_size: usize,
+            out_buf: *mut *const u8,
+            add_buf: *mut u8,
+            add_buf_size: usize,
+            max_size: usize,
+        ) -> usize {
+            unsafe {
+                $crate::wrappers::afl_custom_fuzz_::<$mutator_type>(
+                    data,
+                    buf,
+                    buf_size,
+                    out_buf,
+                    add_buf,
+                    add_buf_size,
+                    max_size,
+                )
+            }
+        }
+
+        #[no_mangle]
+        pub extern "C" fn afl_custom_queue_new_entry(
+            data: *mut ::std::os::raw::c_void,
+            filename_new_queue: *const ::std::os::raw::c_char,
+            filename_orig_queue: *const ::std::os::raw::c_char,
+        ) {
+            $crate::wrappers::afl_custom_queue_new_entry_::<$mutator_type>(
+                data,
+                filename_new_queue,
+                filename_orig_queue,
+            )
+        }
+
+        #[no_mangle]
+        pub extern "C" fn afl_custom_queue_get(
+            data: *mut ::std::os::raw::c_void,
+            filename: *const ::std::os::raw::c_char,
+        ) -> u8 {
+            $crate::wrappers::afl_custom_queue_get_::<$mutator_type>(data, filename)
+        }
+
+        #[no_mangle]
+        pub extern "C" fn afl_custom_introspection(
+            data: *mut ::std::os::raw::c_void,
+        ) -> *const ::std::os::raw::c_char {
+            $crate::wrappers::afl_custom_introspection_::<$mutator_type>(data)
+        }
+
+        #[no_mangle]
+        pub extern "C" fn afl_custom_describe(
+            data: *mut ::std::os::raw::c_void,
+            max_description_len: usize,
+        ) -> *const ::std::os::raw::c_char {
+            $crate::wrappers::afl_custom_describe_::<$mutator_type>(data, max_description_len)
+        }
+
+        #[no_mangle]
+        pub extern "C" fn afl_custom_deinit(data: *mut ::std::os::raw::c_void) {
+            unsafe { $crate::wrappers::afl_custom_deinit_::<$mutator_type>(data) }
+        }
+    };
+}
+
+#[cfg(test)]
+/// this sanity test is supposed to just find out whether an empty mutator being exported by the macro compiles
+mod sanity_test {
+    #[cfg(feature = "afl_internals")]
+    use super::afl_state;
+
+    use super::{export_mutator, RawCustomMutator};
+
+    struct ExampleMutator;
+
+    impl RawCustomMutator for ExampleMutator {
+        #[cfg(feature = "afl_internals")]
+        fn init(_afl: &afl_state, _seed: u32) -> Self {
+            unimplemented!()
+        }
+
+        #[cfg(not(feature = "afl_internals"))]
+        fn init(_seed: u32) -> Self {
+            unimplemented!()
+        }
+
+        fn fuzz<'b, 's: 'b>(
+            &'s mut self,
+            _buffer: &'b mut [u8],
+            _add_buff: Option<&[u8]>,
+            _max_size: usize,
+        ) -> Option<&'b [u8]> {
+            unimplemented!()
+        }
+    }
+
+    export_mutator!(ExampleMutator);
+}
+
+#[allow(unused_variables)]
+/// A custom mutator.
+/// [`CustomMutator::handle_error`] will be called in case any method returns an [`Result::Err`].
+pub trait CustomMutator {
+    /// The error type. All methods must return the same error type.
+    type Error: Debug;
+
+    /// The method which handles errors.
+    /// By default, this method will log the error to stderr if the environment variable "`AFL_CUSTOM_MUTATOR_DEBUG`" is set and non-empty.
+    /// After logging the error, execution will continue on a best-effort basis.
+    ///
+    /// This default behaviour can be customized by implementing this method.
+    fn handle_error(err: Self::Error) {
+        if std::env::var("AFL_CUSTOM_MUTATOR_DEBUG")
+            .map(|v| !v.is_empty())
+            .unwrap_or(false)
+        {
+            eprintln!("Error in custom mutator: {:?}", err)
+        }
+    }
+
+    #[cfg(feature = "afl_internals")]
+    fn init(afl: &'static afl_state, seed: u32) -> Result<Self, Self::Error>
+    where
+        Self: Sized;
+
+    #[cfg(not(feature = "afl_internals"))]
+    fn init(seed: u32) -> Result<Self, Self::Error>
+    where
+        Self: Sized;
+
+    fn fuzz_count(&mut self, buffer: &[u8]) -> Result<u32, Self::Error> {
+        Ok(1)
+    }
+
+    fn fuzz<'b, 's: 'b>(
+        &'s mut self,
+        buffer: &'b mut [u8],
+        add_buff: Option<&[u8]>,
+        max_size: usize,
+    ) -> Result<Option<&'b [u8]>, Self::Error>;
+
+    fn queue_new_entry(
+        &mut self,
+        filename_new_queue: &Path,
+        filename_orig_queue: Option<&Path>,
+    ) -> Result<(), Self::Error> {
+        Ok(())
+    }
+
+    fn queue_get(&mut self, filename: &Path) -> Result<bool, Self::Error> {
+        Ok(true)
+    }
+
+    fn describe(&mut self, max_description: usize) -> Result<Option<&str>, Self::Error> {
+        Ok(Some(default_mutator_describe::<Self>(max_description)))
+    }
+
+    fn introspection(&mut self) -> Result<Option<&str>, Self::Error> {
+        Ok(None)
+    }
+}
+
+impl<M> RawCustomMutator for M
+where
+    M: CustomMutator,
+    M::Error: Debug,
+{
+    #[cfg(feature = "afl_internals")]
+    fn init(afl: &'static afl_state, seed: u32) -> Self
+    where
+        Self: Sized,
+    {
+        match Self::init(afl, seed) {
+            Ok(r) => r,
+            Err(e) => {
+                Self::handle_error(e);
+                panic!("Error in afl_custom_init")
+            }
+        }
+    }
+
+    #[cfg(not(feature = "afl_internals"))]
+    fn init(seed: u32) -> Self
+    where
+        Self: Sized,
+    {
+        match Self::init(seed) {
+            Ok(r) => r,
+            Err(e) => {
+                Self::handle_error(e);
+                panic!("Error in afl_custom_init")
+            }
+        }
+    }
+
+    fn fuzz_count(&mut self, buffer: &[u8]) -> u32 {
+        match self.fuzz_count(buffer) {
+            Ok(r) => r,
+            Err(e) => {
+                Self::handle_error(e);
+                0
+            }
+        }
+    }
+
+    fn fuzz<'b, 's: 'b>(
+        &'s mut self,
+        buffer: &'b mut [u8],
+        add_buff: Option<&[u8]>,
+        max_size: usize,
+    ) -> Option<&'b [u8]> {
+        match self.fuzz(buffer, add_buff, max_size) {
+            Ok(r) => r,
+            Err(e) => {
+                Self::handle_error(e);
+                None
+            }
+        }
+    }
+
+    fn queue_new_entry(&mut self, filename_new_queue: &Path, filename_orig_queue: Option<&Path>) {
+        match self.queue_new_entry(filename_new_queue, filename_orig_queue) {
+            Ok(r) => r,
+            Err(e) => {
+                Self::handle_error(e);
+            }
+        }
+    }
+
+    fn queue_get(&mut self, filename: &Path) -> bool {
+        match self.queue_get(filename) {
+            Ok(r) => r,
+            Err(e) => {
+                Self::handle_error(e);
+                false
+            }
+        }
+    }
+
+    fn describe(&mut self, max_description: usize) -> Option<&str> {
+        match self.describe(max_description) {
+            Ok(r) => r,
+            Err(e) => {
+                Self::handle_error(e);
+                None
+            }
+        }
+    }
+
+    fn introspection(&mut self) -> Option<&str> {
+        match self.introspection() {
+            Ok(r) => r,
+            Err(e) => {
+                Self::handle_error(e);
+                None
+            }
+        }
+    }
+}
+
+/// the default value to return from [`CustomMutator::describe`].
+fn default_mutator_describe<T: ?Sized>(max_len: usize) -> &'static str {
+    truncate_str_unicode_safe(std::any::type_name::<T>(), max_len)
+}
+
+#[cfg(all(test, not(feature = "afl_internals")))]
+mod default_mutator_describe {
+    struct MyMutator;
+    use super::CustomMutator;
+    impl CustomMutator for MyMutator {
+        type Error = ();
+
+        fn init(_: u32) -> Result<Self, Self::Error> {
+            Ok(Self)
+        }
+
+        fn fuzz<'b, 's: 'b>(
+            &'s mut self,
+            _: &'b mut [u8],
+            _: Option<&[u8]>,
+            _: usize,
+        ) -> Result<Option<&'b [u8]>, Self::Error> {
+            unimplemented!()
+        }
+    }
+
+    #[test]
+    fn test_default_describe() {
+        assert_eq!(
+            MyMutator::init(0).unwrap().describe(64).unwrap().unwrap(),
+            "custom_mutator::default_mutator_describe::MyMutator"
+        );
+    }
+}
+
+/// little helper function to truncate a `str` to a maximum of bytes while retaining unicode safety
+fn truncate_str_unicode_safe(s: &str, max_len: usize) -> &str {
+    if s.len() <= max_len {
+        s
+    } else {
+        if let Some((last_index, _)) = s
+            .char_indices()
+            .take_while(|(index, _)| *index <= max_len)
+            .last()
+        {
+            &s[..last_index]
+        } else {
+            ""
+        }
+    }
+}
+
+#[cfg(test)]
+mod truncate_test {
+    use super::truncate_str_unicode_safe;
+
+    #[test]
+    fn test_truncate() {
+        for (max_len, input, expected_output) in &[
+            (0usize, "a", ""),
+            (1, "a", "a"),
+            (1, "ä", ""),
+            (2, "ä", "ä"),
+            (3, "äa", "äa"),
+            (4, "äa", "äa"),
+            (1, "👎", ""),
+            (2, "👎", ""),
+            (3, "👎", ""),
+            (4, "👎", "👎"),
+            (1, "abc", "a"),
+            (2, "abc", "ab"),
+        ] {
+            let actual_output = truncate_str_unicode_safe(input, *max_len);
+            assert_eq!(
+                &actual_output, expected_output,
+                "{:#?} truncated to {} bytes should be {:#?}, but is {:#?}",
+                input, max_len, expected_output, actual_output
+            );
+        }
+    }
+}

custom_mutators/rust/example/Cargo.toml

@@ -0,0 +1,15 @@
+[package]
+name = "example_mutator"
+version = "0.1.0"
+authors = ["Julius Hohnerlein <julihoh@users.noreply.github.com>"]
+edition = "2018"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[dependencies]
+custom_mutator = { path = "../custom_mutator" }
+
+[[example]]
+name = "example_mutator"
+path = "./src/example_mutator.rs"
+crate-type = ["cdylib"]

custom_mutators/rust/example/src/example_mutator.rs

@@ -0,0 +1,50 @@
+#![cfg(unix)]
+#![allow(unused_variables)]
+
+use custom_mutator::{export_mutator, CustomMutator};
+
+struct ExampleMutator;
+
+impl CustomMutator for ExampleMutator {
+    type Error = ();
+
+    fn init(seed: u32) -> Result<Self, Self::Error> {
+        Ok(Self)
+    }
+
+    fn fuzz<'b, 's: 'b>(
+        &'s mut self,
+        buffer: &'b mut [u8],
+        add_buff: Option<&[u8]>,
+        max_size: usize,
+    ) -> Result<Option<&'b [u8]>, Self::Error> {
+        buffer.reverse();
+        Ok(Some(buffer))
+    }
+}
+
+struct OwnBufferExampleMutator {
+    own_buffer: Vec<u8>,
+}
+
+impl CustomMutator for OwnBufferExampleMutator {
+    type Error = ();
+
+    fn init(seed: u32) -> Result<Self, Self::Error> {
+        Ok(Self {
+            own_buffer: Vec::new(),
+        })
+    }
+
+    fn fuzz<'b, 's: 'b>(
+        &'s mut self,
+        buffer: &'b mut [u8],
+        add_buff: Option<&[u8]>,
+        max_size: usize,
+    ) -> Result<Option<&'b [u8]>, ()> {
+        self.own_buffer.reverse();
+        Ok(Some(self.own_buffer.as_slice()))
+    }
+}
+
+export_mutator!(ExampleMutator);

custom_mutators/rust/example_lain/Cargo.toml

@@ -0,0 +1,16 @@
+[package]
+name = "example_lain"
+version = "0.1.0"
+authors = ["Julius Hohnerlein <julihoh@users.noreply.github.com>"]
+edition = "2018"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[dependencies]
+custom_mutator = { path = "../custom_mutator" }
+lain="0.5"
+
+[[example]]
+name = "example_lain"
+path = "./src/lain_mutator.rs"
+crate-type = ["cdylib"]

custom_mutators/rust/example_lain/rust-toolchain

@@ -0,0 +1 @@
+nightly

custom_mutators/rust/example_lain/src/lain_mutator.rs

@@ -0,0 +1,61 @@
+#![cfg(unix)]
+
+use custom_mutator::{export_mutator, CustomMutator};
+use lain::{
+    mutator::Mutator,
+    prelude::*,
+    rand::{rngs::StdRng, SeedableRng},
+};
+
+#[derive(Debug, Mutatable, NewFuzzed, BinarySerialize)]
+struct MyStruct {
+    field_1: u8,
+
+    #[lain(bits = 3)]
+    field_2: u8,
+
+    #[lain(bits = 5)]
+    field_3: u8,
+
+    #[lain(min = 5, max = 10000)]
+    field_4: u32,
+
+    #[lain(ignore)]
+    ignored_field: u64,
+}
+
+struct LainMutator {
+    mutator: Mutator<StdRng>,
+    buffer: Vec<u8>,
+}
+
+impl CustomMutator for LainMutator {
+    type Error = ();
+
+    fn init(seed: u32) -> Result<Self, ()> {
+        Ok(Self {
+            mutator: Mutator::new(StdRng::seed_from_u64(seed as u64)),
+            buffer: Vec::new(),
+        })
+    }
+
+    fn fuzz<'b, 's: 'b>(
+        &'s mut self,
+        _buffer: &'b mut [u8],
+        _add_buff: Option<&[u8]>,
+        max_size: usize,
+    ) -> Result<Option<&'b [u8]>, ()> {
+        // we just sample an instance of MyStruct, ignoring the current input
+        let instance = MyStruct::new_fuzzed(&mut self.mutator, None);
+        let size = instance.serialized_size();
+        if size > max_size {
+            return Err(());
+        }
+        self.buffer.clear();
+        self.buffer.reserve(size);
+        instance.binary_serialize::<_, BigEndian>(&mut self.buffer);
+        Ok(Some(self.buffer.as_slice()))
+    }
+}
+
+export_mutator!(LainMutator);

docs/Changelog.md

@@ -8,61 +8,105 @@
 Want to stay in the loop on major new features? Join our mailing list by
 sending a mail to <afl-users+subscribe@googlegroups.com>.
 
+### Version ++3.11c (release)
+  - afl-fuzz:
+    - better auto detection of map size
+    - fix sanitizer settings (bug since 3.10c)
+    - fix an off-by-one overwrite in cmplog
+    - add non-unicode variants from unicode-looking dictionary entries
+    - Rust custom mutator API improvements
+    - Imported crash stats painted yellow on resume (only new ones are red)
+  - afl-cc:
+    - added AFL_NOOPT that will just pass everything to the normal
+      gcc/clang compiler without any changes - to pass weird configure
+      scripts
+    - fixed a crash that can occur with ASAN + CMPLOG together plus
+      better support for unicode (thanks to @stbergmann for reporting!)
+    - fixed a crash in LAF transform for empty strings
+    - handle erroneous setups in which multiple afl-compiler-rt are
+      compiled into the target. This now also supports dlopen()
+      instrumented libs loaded before the forkserver and even after the
+      forkserver is started (then with collisions though)
+    - the compiler rt was added also in object building (-c) which
+      should have been fixed years ago but somewhere got lost :(
+    - Renamed CTX to CALLER, added correct/real CTX implementation to
+      CLASSIC
+  - qemu_mode:
+    - added AFL_QEMU_EXCLUDE_RANGES env by @realmadsci, thanks!
+    - if no new/updated checkout is wanted, build with:
+      NO_CHECKOUT=1 ./build_qemu_support.sh
+    - we no longer perform a "git drop"
+  - afl-cmin: support filenames with spaces
 
-### Version ++3.01a (dev)
+
+### Version ++3.10c (release)
   - Mac OS ARM64 support
   - Android support fixed and updated by Joey Jiaojg - thanks!
   - New selective instrumentation option with __AFL_COVERAGE_* commands
     to be placed in the source code.
     Check out instrumentation/README.instrument_list.md
   - afl-fuzz
-    - Making AFL_MAP_SIZE obsolete - afl-fuzz now learns on start the
-      target map size
+    - Making AFL_MAP_SIZE (mostly) obsolete - afl-fuzz now learns on
+      start the target map size
     - upgraded cmplog/redqueen: solving for floating point, solving
       transformations (e.g. toupper, tolower, to/from hex, xor,
-      arithmetics, etc.). this is costly hence new command line option
-      -l that sets the intensity (values 1 to 3). recommended is 1 or 2.
-    - added `AFL_CMPLOG_ONLY_NEW` to not use cmplog on initial testcases from
-      `-i` or resumes (as these have most likely already been done)
+      arithmetics, etc.). This is costly hence new command line option
+      `-l` that sets the intensity (values 1 to 3). Recommended is 2.
+    - added `AFL_CMPLOG_ONLY_NEW` to not use cmplog on initial seeds
+      from `-i` or resumes (these have most likely already been done)
     - fix crash for very, very fast targets+systems (thanks to mhlakhani
       for reporting)
-    - on restarts (-i)/autoresume (AFL_AUTORESUME) the stats are now
-      reloaded and used, thanks to Vimal Joseph for this PR! 
-    - if determinstic mode is active (-D, or -M without -d) then we sync
-      after every queue entry as this can take very long time otherwise
+    - on restarts (`-i`)/autoresume (AFL_AUTORESUME) the stats are now
+      reloaded and used, thanks to Vimal Joseph for this patch! 
+    - changed the meaning of '+' of the '-t' option, it now means to
+      auto-calculate the timeout with the value given being the max
+      timeout. The original meaning of skipping timeouts instead of
+      abort is now inherent to the -t option.
+    - if deterministic mode is active (`-D`, or `-M` without `-d`) then
+      we sync after every queue entry as this can take very long time
+      otherwise
+    - added minimum SYNC_TIME to include/config.h (30 minutes default)
     - better detection if a target needs a large shared map
-    - fix for -Z
+    - fix for `-Z`
+    - fixed a few crashes
     - switched to an even faster RNG
     - added hghwng's patch for faster trace map analysis
+    - printing suggestions for mistyped `AFL_` env variables
+    - added Rust bindings for custom mutators (thanks @julihoh)
   - afl-cc
     - allow instrumenting LLVMFuzzerTestOneInput
     - fixed endless loop for allow/blocklist lines starting with a
       comment (thanks to Zherya for reporting)
     - cmplog/redqueen now also tracks floating point, _ExtInt() + 128bit
     - cmplog/redqueen can now process basic libc++ and libstdc++
-      std::string comparisons (though no position or length type variants)
-    - added support for __afl_coverage_interesting() for LTO and
-      and our own PCGUARD (llvm 10.0.1+), read more about this function
-      and selective coverage in instrumentation/README.instrument_list.md
+      std::string comparisons (no position or length type variants)
+    - added support for __afl_coverage_interesting() for LTO and our
+      own PCGUARD (llvm 10.0.1+), read more about this function and
+      selective coverage in instrumentation/README.instrument_list.md
     - added AFL_LLVM_INSTRUMENT option NATIVE for native clang pc-guard
       support (less performant than our own), GCC for old afl-gcc and
       CLANG for old afl-clang
+    - fixed a potential crash in the LAF feature
+    - workaround for llvm bitcast lto bug
+    - workaround for llvm 13
   - qemuafl
-    - ported QASan to qemuafl! see qemu_mode/libqasan/README.md
+    - QASan (address sanitizer for Qemu) ported to qemuafl!
+      See qemu_mode/libqasan/README.md
     - solved some persistent mode bugs (thanks Dil4rd)
     - solved an issue when dumping the memory maps (thanks wizche)
     - Android support for QASan
   - unicornafl
-    - Substential speed gains in python bindings for certain use cases
+    - Substantial speed gains in python bindings for certain use cases
     - Improved rust bindings
-    - Added a new example harness to compare python, c, and rust bindings
+    - Added a new example harness to compare python, c and rust bindings
+  - afl-cmin and afl-showmap now support the -f option
+  - afl_plot now also generates a graph on the discovered edges
   - changed default: no memory limit for afl-cmin and afl-cmin.bash
-  - warn on any _AFL and __AFL env vars
-  - LLVM mode is now compiled with -j4, unicorn with all cores. qemu was
-    already building with all cores, the gcc plugin needs only one.
+  - warn on any _AFL and __AFL env vars.
+  - set AFL_IGNORE_UNKNOWN_ENVS to not warn on unknown AFL_... env vars
   - added dummy Makefile to instrumentation/
   - Updated utils/afl_frida to be 5% faster, 7% on x86_x64
-  - Added AFL_KILL_SIGNAL env variable (thanks @v-p-b)
+  - Added `AFL_KILL_SIGNAL` env variable (thanks @v-p-b)
   - @Edznux added a nice documentation on how to use rpc.statsd with
     afl++ in docs/rpc_statsd.md, thanks!
 

docs/custom_mutators.md

@@ -4,6 +4,11 @@ This file describes how you can implement custom mutations to be used in AFL.
 For now, we support C/C++ library and Python module, collectivelly named as the
 custom mutator.
 
+There is also experimental support for Rust in `custom_mutators/rust`.
+Please refer to that directory for documentation.
+Run ```cargo doc -p custom_mutator --open``` in that directory to view the
+documentation in your web browser.
+
 Implemented by
 - C/C++ library (`*.so`): Khaled Yakdan from Code Intelligence (<yakdan@code-intelligence.de>)
 - Python module: Christian Holler from Mozilla (<choller@mozilla.com>)

docs/docs.md

@@ -0,0 +1,122 @@
+# Restructure afl++'s documentation
+
+## About us
+
+We are dedicated to everything around fuzzing, our main and most well known
+contribution is the fuzzer `afl++` which is part of all major Unix
+distributions (e.g. Debian, Arch, FreeBSD, etc.) and is deployed on Google's
+oss-fuzz and clusterfuzz. It is rated the top fuzzer on Google's fuzzbench.
+
+We are four individuals from Europe supported by a large community.
+
+All our tools are open source.
+
+## About the afl++ fuzzer project
+
+afl++ inherited it's documentation from the original Google afl project.
+Since then it has been massively improved - feature and performance wise -
+and although the documenation has likewise been continued it has grown out
+of proportion.
+The documentation is done by non-natives to the English language, plus
+none of us has a writer background.
+
+We see questions on afl++ usage on mailing lists (e.g. afl-users), discord
+channels, web forums and as issues in our repository.
+
+This only increases as afl++ has been on the top of Google's fuzzbench
+statistics (which measures the performance of fuzzers) and is now being
+integrated in Google's oss-fuzz and clusterfuzz - and is in many Unix
+packaging repositories, e.g. Debian, FreeBSD, etc.
+
+afl++ now has 44 (!) documentation files with 13k total lines of content.
+This is way too much.
+
+Hence afl++ needs a complete overhaul of it's documentation, both on a 
+organisation/structural level as well as the content.
+
+Overall the following actions have to be performed:
+  * Create a better structure of documentation so it is easier to find the
+    information that is being looked for, combining and/or splitting up the
+    existing documents as needed.
+  * Rewrite some documentation to remove duplication. Several information is
+    present several times in the documentation. These should be removed to
+    where needed so that we have as little bloat as possible.
+  * The documents have been written and modified by a lot of different people,
+    most of them non-native English speaker. Hence an overall review where
+    parts should be rewritten has to be performed and then the rewrite done.
+  * Create a cheat-sheet for a very short best-setup build and run of afl++
+  * Pictures explain more than 1000 words. We need at least 4 images that
+    explain the workflow with afl++:
+      - the build workflow
+      - the fuzzing workflow
+      - the fuzzing campaign management workflow
+      - the overall workflow that is an overview of the above
+      - maybe more? where the technical writes seems it necessary for
+        understanding.
+
+Requirements:
+  * Documentation has to be in Markdown format
+  * Images have to be either in SVG or PNG format.
+  * All documentation should be (moved) in(to) docs/
+
+The project does not require writing new documentation or tutorials beside the
+cheat sheet. The technical information for the cheat sheet will be provided by
+us.
+
+## Metrics
+
+afl++ is a the highest performant fuzzer publicly available - but is also the
+most feature rich and complex. With the publicity of afl++' success and
+deployment in Google projects internally and externally and availability as
+a package on most Linux distributions we see more and more issues being
+created and help requests on our Discord channel that would not be
+necessary if people would have read through all our documentation - which
+is unrealistic.
+
+We expect the the new documenation after this project to be cleaner, easier
+accessible and lighter to digest by our users, resulting in much less
+help requests. On the other hand the amount of users using afl++ should
+increase as well as it will be more accessible which would also increase
+questions again - but overall resulting in a reduction of help requests.
+
+In numbers: we currently have per week on average 5 issues on Github,
+10 questions on discord and 1 on mailing lists that would not be necessary
+with perfect documentation and perfect people.
+
+We would consider this project a success if afterwards we only have
+2 issues on Github and 3 questions on discord anymore that would be answered
+by reading the documentation. The mailing list is usually used by the most
+novice users and we don't expect any less questions there.
+
+## Project Budget
+
+We have zero experience with technical writers, so this is very hard for us
+to calculate. We expect it to be a lot of work though because of the amount
+of documentation we have that needs to be restructured and partially rewritten
+(44 documents with 13k total lines of content).
+
+We assume the daily rate of a very good and experienced technical writer in
+times of a pandemic to be ~500$ (according to web research), and calculate
+the overall amout of work to be around 20 days for everything incl. the
+graphics (but again - this is basically just guessing).
+
+Technical Writer                                              10000$
+Volunteer stipends                                                0$ (waved)
+T-Shirts for the top 10 contributors and helpers to this documentation project:
+	10 afl++ logo t-shirts 		20$ each		200$
+	10 shipping cost of t-shirts    10$ each		100$
+
+Total: 10.300$
+(in the submission form 10.280$ was entered)
+
+## Additional Information
+
+We have participated in Google Summer of Code in 2020 and hope to be selected
+again in 2021.
+
+We have no experience with a technical writer, but we will support that person
+with video calls, chats, emails and messaging, provide all necessary information
+and write technical contents that is required for the success of this project.
+It is clear to us that a technical writer knows how to write, but cannot know
+the technical details in a complex tooling like in afl++. This guidance, input,
+etc. has to come from us.

docs/env_variables.md

@@ -5,6 +5,10 @@
   users or for some types of custom fuzzing setups. See [README.md](README.md) for the general
   instruction manual.
 
+  Note that most tools will warn on any unknown AFL environment variables.
+  This is for warning on typos that can happen. If you want to disable this
+  check then set the `AFL_IGNORE_UNKNOWN_ENVS` environment variable.
+
 ## 1) Settings for all compilers
 
 Starting with afl++ 3.0 there is only one compiler: afl-cc
@@ -18,11 +22,21 @@ To select the different instrumentation modes this can be done by
 `MODE` can be one of `LTO` (afl-clang-lto*), `LLVM` (afl-clang-fast*), `GCC_PLUGIN`
 (afl-g*-fast) or `GCC` (afl-gcc/afl-g++).
 
-
 Because (with the exception of the --afl-MODE command line option) the
 compile-time tools do not accept afl specific command-line options, they
 make fairly broad use of environmental variables instead:
 
+  - Some build/configure scripts break with afl++ compilers. To be able to
+    pass them, do:
+```
+       export CC=afl-cc
+       export CXX=afl-c++
+       export AFL_NOOPT=1
+       ./configure --disable-shared --disabler-werror
+       unset AFL_NOOPT
+       make
+```
+
   - Most afl tools do not print any output if stdout/stderr are redirected.
     If you want to get the output into a file then set the `AFL_DEBUG`
     environment variable.
@@ -156,7 +170,6 @@ Then there are a few specific features that are only available in instrumentatio
      This defaults to 1
    - `AFL_LLVM_LTO_DONTWRITEID` prevents that the highest location ID written
      into the instrumentation is set in a global variable
-   - `AFL_LLVM_LTO_CALLGRAPH` will not generate a binary but a callgraph instead
 
   See [instrumentation/README.lto.md](../instrumentation/README.lto.md) for more information.
 
@@ -449,6 +462,7 @@ checks or alter some of the more exotic semantics of the tool:
     `banner` corresponds to the name of the fuzzer provided through `-M/-S`.
     `afl_version` corresponds to the currently running afl version (e.g `++3.0c`).
     Default (empty/non present) will add no tags to the metrics.
+    See [rpc_statsd.md](rpc_statsd.md) for more information.
 
   - Setting `AFL_CRASH_EXITCODE` sets the exit code afl treats as crash.
     For example, if `AFL_CRASH_EXITCODE='-1'` is set, each input resulting
@@ -515,6 +529,12 @@ The QEMU wrapper used to instrument binary-only code supports several settings:
     stack pointer in which QEMU can find the return address when `start addr` is
     hit.
 
+  - With `AFL_USE_QASAN` you can enable QEMU AddressSanitizer for dynamically
+    linked binaries.
+
+  - With `AFL_QEMU_FORCE_DFL` you force QEMU to ignore the registered signal
+    handlers of the target.
+
 ## 6) Settings for afl-cmin
 
 The corpus minimization script offers very little customization:

docs/ideas.md

@@ -3,6 +3,42 @@
 In the following, we describe a variety of ideas that could be implemented
 for future AFL++ versions.
 
+# GSoC 2021
+
+All GSoC 2021 projects will be in the Rust development language!
+
+## UI for libaflrs
+
+Write a user interface to libaflrs, the upcoming backend of afl++.
+This might look like the afl-fuzz UI, but you can improve on it - and should!
+
+## Schedulers for libaflrs
+
+Schedulers is a mechanism that selects items from the fuzzing corpus based
+on strategy and randomness. One scheduler might focus on long paths,
+another on rarity of edges disocvered, still another on a combination on
+things. Some of the schedulers in afl++ have to be ported, but you are free
+to come up with your own if you want to - and see how it performs.
+
+## Forkserver support for libaflrs
+
+The current libaflrs implementation fuzzes in-memory, however obviously we
+want to support afl instrumented binaries as well.
+Hence a forkserver support needs to be implemented - forking off the target
+and talking to the target via a socketpair and the communication protocol
+within.
+
+## More Observers for libaflrs
+
+An observer is measuring functionality that looks at the target being fuzzed
+and documents something about it. In traditional fuzzing this is the coverage
+in the target, however we want to add various more observers, e.g. stack depth,
+heap usage, etc. - this is a topic for an experienced Rust developer.
+
+# Generic ideas and wishlist - NOT PART OF GSoC 2021 !
+
+The below list is not part of GSoC 2021.
+
 ## Analysis software
 
 Currently analysis is done by using afl-plot, which is rather outdated.

docs/rpc_statsd.md

@@ -1,6 +1,6 @@
 # Remote monitoring with StatsD
 
-StatsD allows you to receive and aggregate metrics from a wide range of application and retransmit them to the backend of your choice.
+StatsD allows you to receive and aggregate metrics from a wide range of applications and retransmit them to the backend of your choice.
 This enables you to create nice and readable dashboards containing all the information you need on your fuzzer instances.
 No need to write your own statistics parsing system, deploy and maintain it to all your instances, sync with your graph rendering system...
 
@@ -45,7 +45,7 @@ For more information on these env vars, check out `docs/env_variables.md`.
 
 The simplest way of using this feature is to use any metric provider and change the host/port of your StatsD daemon,
 with `AFL_STATSD_HOST` and `AFL_STATSD_PORT`, if required (defaults are `localhost` and port `8125`).
-To get started, here are some instruction with free and open source tools.
+To get started, here are some instructions with free and open source tools.
 The following setup is based on Prometheus, statsd_exporter and Grafana.
 Grafana here is not mandatory, but gives you some nice graphs and features.
 
@@ -131,7 +131,7 @@ mappings:
 
 Run `docker-compose up -d`.
 
-Everything should be now setup, you are now able to run your fuzzers with
+Everything should now be setup, you are now able to run your fuzzers with
 
 ```
 AFL_STATSD_TAGS_FLAVOR=dogstatsd AFL_STATSD=1 afl-fuzz -M test-fuzzer-1 -i i -o o ./bin/my-application @@
@@ -139,5 +139,5 @@ AFL_STATSD_TAGS_FLAVOR=dogstatsd AFL_STATSD=1 afl-fuzz -S test-fuzzer-2 -i i -o
 ...
 ```
 
-This setup may be modified before use in production environment. Depending on your needs: addind passwords, creating volumes for storage,
+This setup may be modified before use in a production environment. Depending on your needs: adding passwords, creating volumes for storage,
 tweaking the metrics gathering to get host metrics (CPU, RAM ...).

include/afl-fuzz.h

@@ -570,6 +570,7 @@ typedef struct afl_state {
       blocks_eff_total,                 /* Blocks subject to effector maps  */
       blocks_eff_select,                /* Blocks selected as fuzzable      */
       start_time,                       /* Unix start time (ms)             */
+      last_sync_time,                   /* Time of last sync                */
       last_path_time,                   /* Time for most recent path (ms)   */
       last_crash_time,                  /* Time for most recent crash (ms)  */
       last_hang_time;                   /* Time for most recent hang (ms)   */
@@ -649,6 +650,7 @@ typedef struct afl_state {
   u32 cmplog_max_filesize;
   u32 cmplog_lvl;
   u32 colorize_success;
+  u8  cmplog_enable_arith, cmplog_enable_transform;
 
   struct afl_pass_stat *pass_stats;
   struct cmp_map *      orig_cmp_map;
@@ -1060,6 +1062,7 @@ u8 has_new_bits_unclassified(afl_state_t *, u8 *);
 void load_extras_file(afl_state_t *, u8 *, u32 *, u32 *, u32);
 void load_extras(afl_state_t *, u8 *);
 void dedup_extras(afl_state_t *);
+void deunicode_extras(afl_state_t *);
 void add_extra(afl_state_t *afl, u8 *mem, u32 len);
 void maybe_add_auto(afl_state_t *, u8 *, u32);
 void save_auto(afl_state_t *);
@@ -1070,8 +1073,8 @@ void destroy_extras(afl_state_t *);
 
 void load_stats_file(afl_state_t *);
 void write_setup_file(afl_state_t *, u32, char **);
-void write_stats_file(afl_state_t *, double, double, double);
-void maybe_update_plot_file(afl_state_t *, double, double);
+void write_stats_file(afl_state_t *, u32, double, double, double);
+void maybe_update_plot_file(afl_state_t *, u32, double, double);
 void show_stats(afl_state_t *);
 void show_init_stats(afl_state_t *);
 

include/common.h

@@ -39,6 +39,7 @@
 #define STRINGIFY_VAL_SIZE_MAX (16)
 
 void detect_file_args(char **argv, u8 *prog_in, bool *use_stdin);
+void print_suggested_envs(char *mispelled_env);
 void check_environment_vars(char **env);
 
 char **argv_cpy_dup(int argc, char **argv);

include/config.h

@@ -10,7 +10,7 @@
                      Dominik Maier <mail@dmnk.co>
 
    Copyright 2016, 2017 Google Inc. All rights reserved.
-   Copyright 2019-2020 AFLplusplus Project. All rights reserved.
+   Copyright 2019-2021 AFLplusplus Project. All rights reserved.
 
    Licensed under the Apache License, Version 2.0 (the "License");
    you may not use this file except in compliance with the License.
@@ -25,8 +25,8 @@
 
 /* Version string: */
 
-// c = release, d = volatile github dev, e = experimental branch
-#define VERSION "++3.01a"
+// c = release, a = volatile github dev, e = experimental branch
+#define VERSION "++3.11c"
 
 /******************************************************
  *                                                    *
@@ -36,26 +36,27 @@
 
 /* CMPLOG/REDQUEEN TUNING
  *
- * Here you can tuning and solving options for cmplog.
+ * Here you can modify tuning and solving options for CMPLOG.
  * Note that these are run-time options for afl-fuzz, no target
  * recompilation required.
  *
  */
 
-/* Enable transform following (XOR/ADD/SUB manipulations, hex en/decoding) */
-// #define CMPLOG_TRANSFORM
+/* if TRANSFORM is enabled with '-l T', this additionally enables base64
+   encoding/decoding */
+// #define CMPLOG_SOLVE_TRANSFORM_BASE64
 
-/* if TRANSFORM is enabled, this additionally enables base64 en/decoding */
-// #define CMPLOG_TRANSFORM_BASE64
+/* If a redqueen pass finds more than one solution, try to combine them? */
+#define CMPLOG_COMBINE
 
-/* Minimum % of the corpus to perform cmplog on. Default: 20% */
-#define CMPLOG_CORPUS_PERCENT 20U
+/* Minimum % of the corpus to perform cmplog on. Default: 10% */
+#define CMPLOG_CORPUS_PERCENT 5U
 
-/* Number of potential posititions from which we decide the cmplog becomes
-   useless, default 16384 */
-#define CMPLOG_POSITIONS_MAX 16384U
+/* Number of potential positions from which we decide if cmplog becomes
+   useless, default 8096 */
+#define CMPLOG_POSITIONS_MAX (12 * 1024)
 
-/* Maximum allowed fails per CMP value. Default: 32 * 3 */
+/* Maximum allowed fails per CMP value. Default: 128 */
 #define CMPLOG_FAIL_MAX 96
 
 /* Now non-cmplog configuration options */
@@ -279,6 +280,11 @@
 
 #define SYNC_INTERVAL 8
 
+/* Sync time (minimum time between syncing in ms, time is halfed for -M main
+   nodes) - default is 30 minutes: */
+
+#define SYNC_TIME (30 * 60 * 1000)
+
 /* Output directory reuse grace period (minutes): */
 
 #define OUTPUT_GRACE 25

include/envs.h

@@ -61,6 +61,7 @@ static char *afl_environment_variables[] = {
     "AFL_FORKSRV_INIT_TMOUT",
     "AFL_HARDEN",
     "AFL_I_DONT_CARE_ABOUT_MISSING_CRASHES",
+    "AFL_IGNORE_UNKNOWN_ENVS",
     "AFL_IMPORT_FIRST",
     "AFL_INST_LIBS",
     "AFL_INST_RATIO",
@@ -79,7 +80,9 @@ static char *afl_environment_variables[] = {
     "AFL_LLVM_BLOCKLIST",
     "AFL_LLVM_CMPLOG",
     "AFL_LLVM_INSTRIM",
+    "AFL_LLVM_CALLER",
     "AFL_LLVM_CTX",
+    "AFL_LLVM_CTX_K",
     "AFL_LLVM_DICT2FILE",
     "AFL_LLVM_DOCUMENT_IDS",
     "AFL_LLVM_INSTRIM_LOOPHEAD",
@@ -102,9 +105,8 @@ static char *afl_environment_variables[] = {
     "AFL_LLVM_INSTRUMENT_FILE",
     "AFL_LLVM_SKIP_NEVERZERO",
     "AFL_NO_AFFINITY",
-    "AFL_LLVM_LTO_CALLGRAPH",
-    "AFL_LLVM_LTO_DONTWRITEID",
     "AFL_LLVM_LTO_STARTID",
+    "AFL_LLVM_LTO_DONTWRITEID",
     "AFL_NO_ARITH",
     "AFL_NO_AUTODICT",
     "AFL_NO_BUILTIN",
@@ -118,10 +120,12 @@ static char *afl_environment_variables[] = {
     "AFL_NO_PYTHON",
     "AFL_UNTRACER_FILE",
     "AFL_LLVM_USE_TRACE_PC",
-    "AFL_NO_X86",  // not really an env but we dont want to warn on it
     "AFL_MAP_SIZE",
     "AFL_MAPSIZE",
     "AFL_MAX_DET_EXTRAS",
+    "AFL_NO_X86",  // not really an env but we dont want to warn on it
+    "AFL_NOOPT",
+    "AFL_PASSTHROUGH",
     "AFL_PATH",
     "AFL_PERFORMANCE_FILE",
     "AFL_PRELOAD",
@@ -131,6 +135,7 @@ static char *afl_environment_variables[] = {
     "AFL_QEMU_DEBUG_MAPS",
     "AFL_QEMU_DISABLE_CACHE",
     "AFL_QEMU_DRIVER_NO_HOOK",
+    "AFL_QEMU_FORCE_DFL",
     "AFL_QEMU_PERSISTENT_ADDR",
     "AFL_QEMU_PERSISTENT_CNT",
     "AFL_QEMU_PERSISTENT_GPR",
@@ -140,6 +145,7 @@ static char *afl_environment_variables[] = {
     "AFL_QEMU_PERSISTENT_RETADDR_OFFSET",
     "AFL_QEMU_PERSISTENT_EXITS",
     "AFL_QEMU_INST_RANGES",
+    "AFL_QEMU_EXCLUDE_RANGES",
     "AFL_QEMU_SNAPSHOT",
     "AFL_QUIET",
     "AFL_RANDOM_ALLOC_CANARY",

instrumentation/LLVMInsTrim.so.cc

@@ -38,7 +38,7 @@ typedef long double max_align_t;
 
 #include "MarkNodes.h"
 #include "afl-llvm-common.h"
-#include "llvm-ngram-coverage.h"
+#include "llvm-alternative-coverage.h"
 
 #include "config.h"
 #include "debug.h"
@@ -135,7 +135,7 @@ struct InsTrim : public ModulePass {
     unsigned int PrevLocSize = 0;
     char *       ngram_size_str = getenv("AFL_LLVM_NGRAM_SIZE");
     if (!ngram_size_str) ngram_size_str = getenv("AFL_NGRAM_SIZE");
-    char *ctx_str = getenv("AFL_LLVM_CTX");
+    char *caller_str = getenv("AFL_LLVM_CALLER");
 
 #ifdef AFL_HAVE_VECTOR_INTRINSICS
     unsigned int ngram_size = 0;
@@ -197,9 +197,9 @@ struct InsTrim : public ModulePass {
                            GlobalValue::ExternalLinkage, 0, "__afl_area_ptr");
     GlobalVariable *AFLPrevLoc;
     GlobalVariable *AFLContext = NULL;
-    LoadInst *      PrevCtx = NULL;  // for CTX sensitive coverage
+    LoadInst *      PrevCaller = NULL;  // for CALLER sensitive coverage
 
-    if (ctx_str)
+    if (caller_str)
 #if defined(__ANDROID__) || defined(__HAIKU__)
       AFLContext = new GlobalVariable(
           M, Int32Ty, false, GlobalValue::ExternalLinkage, 0, "__afl_prev_ctx");
@@ -398,11 +398,11 @@ struct InsTrim : public ModulePass {
         unsigned int cur_loc;
 
         // Context sensitive coverage
-        if (ctx_str && &BB == &F.getEntryBlock()) {
+        if (caller_str && &BB == &F.getEntryBlock()) {
 
-          PrevCtx = IRB.CreateLoad(AFLContext);
-          PrevCtx->setMetadata(M.getMDKindID("nosanitize"),
-                               MDNode::get(C, None));
+          PrevCaller = IRB.CreateLoad(AFLContext);
+          PrevCaller->setMetadata(M.getMDKindID("nosanitize"),
+                                  MDNode::get(C, None));
 
           // does the function have calls? and is any of the calls larger than
           // one basic block?
@@ -441,7 +441,7 @@ struct InsTrim : public ModulePass {
 
           }
 
-        }  // END of ctx_str
+        }  // END of caller_str
 
         if (MarkSetOpt && MS.find(&BB) == MS.end()) { continue; }
 
@@ -485,9 +485,9 @@ struct InsTrim : public ModulePass {
 #endif
           PrevLocTrans = IRB.CreateZExt(PrevLoc, IRB.getInt32Ty());
 
-        if (ctx_str)
+        if (caller_str)
           PrevLocTrans =
-              IRB.CreateZExt(IRB.CreateXor(PrevLocTrans, PrevCtx), Int32Ty);
+              IRB.CreateZExt(IRB.CreateXor(PrevLocTrans, PrevCaller), Int32Ty);
 
         /* Load SHM pointer */
         LoadInst *MapPtr = IRB.CreateLoad(AFLMapPtr);
@@ -535,16 +535,17 @@ struct InsTrim : public ModulePass {
         IRB.CreateStore(Incr, MapPtrIdx)
             ->setMetadata(M.getMDKindID("nosanitize"), MDNode::get(C, None));
 
-        if (ctx_str && has_calls) {
+        if (caller_str && has_calls) {
 
-          // in CTX mode we have to restore the original context for the
+          // in CALLER mode we have to restore the original context for the
           // caller - she might be calling other functions which need the
-          // correct CTX
+          // correct CALLER
           Instruction *Inst = BB.getTerminator();
           if (isa<ReturnInst>(Inst) || isa<ResumeInst>(Inst)) {
 
             IRBuilder<> Post_IRB(Inst);
-            StoreInst * RestoreCtx = Post_IRB.CreateStore(PrevCtx, AFLContext);
+            StoreInst * RestoreCtx =
+                Post_IRB.CreateStore(PrevCaller, AFLContext);
             RestoreCtx->setMetadata(M.getMDKindID("nosanitize"),
                                     MDNode::get(C, None));
 

instrumentation/README.cmplog.md

@@ -1,10 +1,11 @@
 # CmpLog instrumentation
 
-The CmpLog instrumentation enables the logging of the comparisons operands in a
+The CmpLog instrumentation enables logging of comparison operands in a
 shared memory.
 
 These values can be used by various mutators built on top of it.
-At the moment we support the RedQueen mutator (input-2-state instructions only).
+At the moment we support the RedQueen mutator (input-2-state instructions only), 
+for details see [the RedQueen paper](https://www.syssec.ruhr-uni-bochum.de/media/emma/veroeffentlichungen/2018/12/17/NDSS19-Redqueen.pdf).
 
 ## Build
 
@@ -13,7 +14,7 @@ program.
 
 The first version is built using the regular AFL++ instrumentation.
 
-The second one, the CmpLog binary, with setting AFL_LLVM_CMPLOG during the compilation.
+The second one, the CmpLog binary, is built with setting AFL_LLVM_CMPLOG during the compilation.
 
 For example:
 
@@ -26,11 +27,12 @@ export AFL_LLVM_CMPLOG=1
 ./configure --cc=~/path/to/afl-clang-fast
 make
 cp ./program ./program.cmplog
+unset AFL_LLVM_CMPLOG
 ```
 
 ## Use
 
-AFL++ has the new -c option that needs to be used to specify the CmpLog binary (the second
+AFL++ has the new `-c` option that needs to be used to specify the CmpLog binary (the second
 build).
 
 For example:

instrumentation/README.ctx.md

@@ -4,14 +4,19 @@
 
 This is an LLVM-based implementation of the context sensitive branch coverage.
 
-Basically every function gets its own ID and that ID is combined with the
-edges of the called functions.
+Basically every function gets its own ID and, every time when an edge is logged,
+all the IDs in the callstack are hashed and combined with the edge transition
+hash to augment the classic edge coverage with the information about the
+calling context.
 
 So if both function A and function B call a function C, the coverage
 collected in C will be different.
 
 In math the coverage is collected as follows:
-`map[current_location_ID ^ previous_location_ID >> 1 ^ previous_callee_ID] += 1`
+`map[current_location_ID ^ previous_location_ID >> 1 ^ hash_callstack_IDs] += 1`
+
+The callstack hash is produced XOR-ing the function IDs to avoid explosion with
+recursive functions.
 
 ## Usage
 
@@ -20,3 +25,14 @@ Set the `AFL_LLVM_INSTRUMENT=CTX` or `AFL_LLVM_CTX=1` environment variable.
 It is highly recommended to increase the MAP_SIZE_POW2 definition in
 config.h to at least 18 and maybe up to 20 for this as otherwise too
 many map collisions occur.
+
+## Caller Branch Coverage
+
+If the context sensitive coverage introduces too may collisions and becoming
+detrimental, the user can choose to augment edge coverage with just the
+called function ID, instead of the entire callstack hash.
+
+In math the coverage is collected as follows:
+`map[current_location_ID ^ previous_location_ID >> 1 ^ previous_callee_ID] += 1`
+
+Set the `AFL_LLVM_INSTRUMENT=CALLER` or `AFL_LLVM_CALLER=1` environment variable.

instrumentation/README.gcc_plugin.md

@@ -3,16 +3,20 @@
 See [../README.md](../README.md) for the general instruction manual.
 See [README.llvm.md](README.llvm.md) for the LLVM-based instrumentation.
 
+This document describes how to build and use `afl-gcc-fast` and `afl-g++-fast`,
+which instrument the target with the help of gcc plugins.
+
 TLDR:
-  * `apt-get install gcc-VERSION-plugin-dev`
-  * `make`
-  * gcc and g++ must point to the gcc-VERSION you you have to set AFL_CC/AFL_CXX
+  * check the version of your gcc compiler: `gcc --version`
+  * `apt-get install gcc-VERSION-plugin-dev` or similar to install headers for gcc plugins
+  * `gcc` and `g++` must match the gcc-VERSION you installed headers for. You can set `AFL_CC`/`AFL_CXX`
     to point to these!
-  * just use afl-gcc-fast/afl-g++-fast normally like you would afl-clang-fast
+  * `make`
+  * just use `afl-gcc-fast`/`afl-g++-fast` normally like you would do with `afl-clang-fast`
 
 ## 1) Introduction
 
-The code in this directory allows you to instrument programs for AFL using
+The code in this directory allows to instrument programs for AFL using
 true compiler-level instrumentation, instead of the more crude
 assembly-level rewriting approach taken by afl-gcc and afl-clang. This has
 several interesting properties:
@@ -27,10 +31,10 @@ several interesting properties:
 
   - The instrumentation is CPU-independent. At least in principle, you should
     be able to rely on it to fuzz programs on non-x86 architectures (after
-    building afl-fuzz with AFL_NOX86=1).
+    building `afl-fuzz` with `AFL_NOX86=1`).
 
   - Because the feature relies on the internals of GCC, it is gcc-specific
-    and will *not* work with LLVM (see ../llvm_mode for an alternative).
+    and will *not* work with LLVM (see [README.llvm.md](README.llvm.md) for an alternative).
 
 Once this implementation is shown to be sufficiently robust and portable, it
 will probably replace afl-gcc. For now, it can be built separately and
@@ -41,29 +45,32 @@ The idea and much of the implementation comes from Laszlo Szekeres.
 ## 2) How to use
 
 In order to leverage this mechanism, you need to have modern enough GCC
-(>= version 4.5.0) and the plugin headers installed on your system. That
+(>= version 4.5.0) and the plugin development headers installed on your system. That
 should be all you need. On Debian machines, these headers can be acquired by
 installing the `gcc-VERSION-plugin-dev` packages.
 
-To build the instrumentation itself, type 'make'. This will generate binaries
-called afl-gcc-fast and afl-g++-fast in the parent directory. 
+To build the instrumentation itself, type `make`. This will generate binaries
+called `afl-gcc-fast` and `afl-g++-fast` in the parent directory. 
 
 The gcc and g++ compiler links have to point to gcc-VERSION - or set these
-by pointing the environment variables AFL_CC/AFL_CXX to them.
-If the CC/CXX have been overridden, those compilers will be used from
-those wrappers without using AFL_CXX/AFL_CC settings.
+by pointing the environment variables `AFL_CC`/`AFL_CXX` to them.
+If the `CC`/`CXX` environment variables have been set, those compilers will be 
+preferred over those from the `AFL_CC`/`AFL_CXX` settings.
 
 Once this is done, you can instrument third-party code in a way similar to the
 standard operating mode of AFL, e.g.:
-
-  CC=/path/to/afl/afl-gcc-fast ./configure [...options...]
+```
+  CC=/path/to/afl/afl-gcc-fast
+  CXX=/path/to/afl/afl-g++-fast
+  export CC CXX
+  ./configure [...options...]
   make
+```
+Note: We also used `CXX` to set the C++ compiler to `afl-g++-fast` for C++ code.
 
-Be sure to also include CXX set to afl-g++-fast for C++ code.
-
-The tool honors roughly the same environmental variables as afl-gcc (see
-[env_variables.md](../docs/env_variables.md). This includes AFL_INST_RATIO,
-AFL_USE_ASAN, AFL_HARDEN, and AFL_DONT_OPTIMIZE.
+The tool honors roughly the same environmental variables as `afl-gcc` (see
+[env_variables.md](../docs/env_variables.md). This includes `AFL_INST_RATIO`,
+`AFL_USE_ASAN`, `AFL_HARDEN`, and `AFL_DONT_OPTIMIZE`.
 
 Note: if you want the GCC plugin to be installed on your system for all
 users, you need to build it before issuing 'make install' in the parent
@@ -72,7 +79,7 @@ directory.
 ## 3) Gotchas, feedback, bugs
 
 This is an early-stage mechanism, so field reports are welcome. You can send bug
-reports to afl@aflplus.plus
+reports to afl@aflplus.plus.
 
 ## 4) Bonus feature #1: deferred initialization
 
@@ -88,7 +95,7 @@ file before getting to the fuzzed data.
 In such cases, it's beneficial to initialize the forkserver a bit later, once
 most of the initialization work is already done, but before the binary attempts
 to read the fuzzed input and parse it; in some cases, this can offer a 10x+
-performance gain. You can implement delayed initialization in LLVM mode in a
+performance gain. You can implement delayed initialization in GCC mode in a
 fairly simple way.
 
 First, locate a suitable location in the code where the delayed cloning can
@@ -117,7 +124,7 @@ With the location selected, add this code in the appropriate spot:
 ```
 
 You don't need the #ifdef guards, but they will make the program still work as
-usual when compiled with a tool other than afl-gcc-fast/afl-clang-fast.
+usual when compiled with a compiler other than afl-gcc-fast/afl-clang-fast.
 
 Finally, recompile the program with afl-gcc-fast (afl-gcc or afl-clang will
 *not* generate a deferred-initialization binary) - and you should be all set!
@@ -127,7 +134,7 @@ Finally, recompile the program with afl-gcc-fast (afl-gcc or afl-clang will
 Some libraries provide APIs that are stateless, or whose state can be reset in
 between processing different input files. When such a reset is performed, a
 single long-lived process can be reused to try out multiple test cases,
-eliminating the need for repeated fork() calls and the associated OS overhead.
+eliminating the need for repeated `fork()` calls and the associated OS overhead.
 
 The basic structure of the program that does this would be:
 
@@ -160,5 +167,9 @@ wary of memory leaks and the state of file descriptors.
 When running in this mode, the execution paths will inherently vary a bit
 depending on whether the input loop is being entered for the first time or
 executed again. To avoid spurious warnings, the feature implies
-AFL_NO_VAR_CHECK and hides the "variable path" warnings in the UI.
+`AFL_NO_VAR_CHECK` and hides the "variable path" warnings in the UI.
 
+## 6) Bonus feature #3: selective instrumentation
+
+It can be more effective to fuzzing to only instrument parts of the code.
+For details see [README.instrument_list.md](README.instrument_list.md).

instrumentation/README.instrument_list.md

@@ -47,10 +47,10 @@ A special function is `__afl_coverage_interesting`.
 To use this, you must define `void __afl_coverage_interesting(u8 val, u32 id);`.
 Then you can use this function globally, where the `val` parameter can be set
 by you, the `id` parameter is for afl-fuzz and will be overwritten.
-Note that useful parameters are for `val` are: 1, 2, 3, 4, 8, 16, 32, 64, 128.
+Note that useful parameters for `val` are: 1, 2, 3, 4, 8, 16, 32, 64, 128.
 A value of e.g. 33 will be seen as 32 for coverage purposes.
 
-## 3) Selective instrumenation with AFL_LLVM_ALLOWLIST/AFL_LLVM_DENYLIST
+## 3) Selective instrumentation with AFL_LLVM_ALLOWLIST/AFL_LLVM_DENYLIST
 
 This feature is equivalent to llvm 12 sancov feature and allows to specify
 on a filename and/or function name level to instrument these or skip them.

instrumentation/README.lto.md

@@ -88,16 +88,35 @@ apt-get install -y clang-12 clang-tools-12 libc++1-12 libc++-12-dev \
 ### Building llvm yourself (version 12)
 
 Building llvm from github takes quite some long time and is not painless:
-```
+```sh
 sudo apt install binutils-dev  # this is *essential*!
-git clone https://github.com/llvm/llvm-project
+git clone --depth=1 https://github.com/llvm/llvm-project
 cd llvm-project
 mkdir build
 cd build
-cmake -DLLVM_ENABLE_PROJECTS='clang;clang-tools-extra;compiler-rt;libclc;libcxx;libcxxabi;libunwind;lld' -DCMAKE_BUILD_TYPE=Release -DLLVM_BINUTILS_INCDIR=/usr/include/ ../llvm/
-make -j $(nproc)
-export PATH=`pwd`/bin:$PATH
-export LLVM_CONFIG=`pwd`/bin/llvm-config
+
+# Add -G Ninja if ninja-build installed
+# "Building with ninja significantly improves your build time, especially with
+# incremental builds, and improves your memory usage."
+cmake \
+    -DCLANG_INCLUDE_DOCS="OFF" \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DLLVM_BINUTILS_INCDIR=/usr/include/ \
+    -DLLVM_BUILD_LLVM_DYLIB="ON" \
+    -DLLVM_ENABLE_BINDINGS="OFF" \
+    -DLLVM_ENABLE_PROJECTS='clang;compiler-rt;libcxx;libcxxabi;libunwind;lld' \
+    -DLLVM_ENABLE_WARNINGS="OFF" \
+    -DLLVM_INCLUDE_BENCHMARKS="OFF" \
+    -DLLVM_INCLUDE_DOCS="OFF" \
+    -DLLVM_INCLUDE_EXAMPLES="OFF" \
+    -DLLVM_INCLUDE_TESTS="OFF" \
+    -DLLVM_LINK_LLVM_DYLIB="ON" \
+    -DLLVM_TARGETS_TO_BUILD="host" \
+    ../llvm/
+cmake --build . --parallel
+export PATH="$(pwd)/bin:$PATH"
+export LLVM_CONFIG="$(pwd)/bin/llvm-config"
+export LD_LIBRARY_PATH="$(llvm-config --libdir)${LD_LIBRARY_PATH:+:$LD_LIBRARY_PATH}"
 cd /path/to/AFLplusplus/
 make
 sudo make install

instrumentation/README.neverzero.md

@@ -16,7 +16,7 @@ at a very little cost (one instruction per edge).
 (The alternative of saturated counters has been tested also and proved to be
 inferior in terms of path discovery.)
 
-This is implemented in afl-gcc, however for llvm_mode this is optional if
+This is implemented in afl-gcc and afl-gcc-fast, however for llvm_mode this is optional if
 the llvm version is below 9 - as there is a perfomance bug that is only fixed
 in version 9 and onwards.
 

instrumentation/README.ngram.md

@@ -10,8 +10,8 @@ by Jinghan Wang, et. al.
 Note that the original implementation (available
 [here](https://github.com/bitsecurerlab/afl-sensitive))
 is built on top of AFL's QEMU mode.
-This is essentially a port that uses LLVM vectorized instructions to achieve
-the same results when compiling source code.
+This is essentially a port that uses LLVM vectorized instructions (available from
+llvm versions 4.0.1 and higher) to achieve the same results when compiling source code.
 
 In math the branch coverage is performed as follows:
 `map[current_location ^ prev_location[0] >> 1 ^ prev_location[1] >> 1 ^ ... up to n-1`] += 1`

instrumentation/README.out_of_line.md

@@ -1,18 +1,16 @@
-===========================================
-Using afl++ without inlined instrumentation
-===========================================
+## Using afl++ without inlined instrumentation
 
   This file describes how you can disable inlining of instrumentation.
 
 
 By default, the GCC plugin will duplicate the effects of calling
-__afl_trace (see afl-gcc-rt.o.c) in instrumented code, instead of
+`__afl_trace` (see `afl-gcc-rt.o.c`) in instrumented code, instead of
 issuing function calls.
 
 The calls are presumed to be slower, more so because the rt file
 itself is not optimized by the compiler.
 
-Setting AFL_GCC_OUT_OF_LINE=1 in the environment while compiling code
+Setting `AFL_GCC_OUT_OF_LINE=1` in the environment while compiling code
 with the plugin will disable this inlining, issuing calls to the
 unoptimized runtime instead.
 

instrumentation/README.persistent_mode.md

@@ -16,7 +16,7 @@ Examples can be found in [utils/persistent_mode](../utils/persistent_mode).
 ## 2) TLDR;
 
 Example `fuzz_target.c`:
-```
+```c
 #include "what_you_need_for_your_target.h"
 
 __AFL_FUZZ_INIT();
@@ -60,14 +60,14 @@ The speed increase is usually x10 to x20.
 If you want to be able to compile the target without afl-clang-fast/lto then
 add this just after the includes:
 
-```
+```c
 #ifndef __AFL_FUZZ_TESTCASE_LEN
   ssize_t fuzz_len;
   #define __AFL_FUZZ_TESTCASE_LEN fuzz_len
   unsigned char fuzz_buf[1024000];
   #define __AFL_FUZZ_TESTCASE_BUF fuzz_buf
   #define __AFL_FUZZ_INIT() void sync(void);
-  #define __AFL_LOOP(x) ((fuzz_len = read(0, fuzz_buf, sizeof(fuzz_buf))) > 0 ?
+  #define __AFL_LOOP(x) ((fuzz_len = read(0, fuzz_buf, sizeof(fuzz_buf))) > 0 ? 1 : 0)
   #define __AFL_INIT() sync() 
 #endif
 ```
@@ -75,7 +75,7 @@ add this just after the includes:
 ## 3) Deferred initialization
 
 AFL tries to optimize performance by executing the targeted binary just once,
-stopping it just before main(), and then cloning this "main" process to get
+stopping it just before `main()`, and then cloning this "main" process to get
 a steady supply of targets to fuzz.
 
 Although this approach eliminates much of the OS-, linker- and libc-level
@@ -97,7 +97,7 @@ a location after:
   - The creation of any vital threads or child processes - since the forkserver
     can't clone them easily.
 
-  - The initialization of timers via setitimer() or equivalent calls.
+  - The initialization of timers via `setitimer()` or equivalent calls.
 
   - The creation of temporary files, network sockets, offset-sensitive file
     descriptors, and similar shared-state resources - but only provided that
@@ -150,9 +150,9 @@ the impact of memory leaks and similar glitches; 1000 is a good starting point,
 and going much higher increases the likelihood of hiccups without giving you
 any real performance benefits.
 
-A more detailed template is shown in ../utils/persistent_mode/.
-Similarly to the previous mode, the feature works only with afl-clang-fast; #ifdef
-guards can be used to suppress it when using other compilers.
+A more detailed template is shown in `../utils/persistent_mode/.`
+Similarly to the previous mode, the feature works only with afl-clang-fast; 
+`#ifdef` guards can be used to suppress it when using other compilers.
 
 Note that as with the previous mode, the feature is easy to misuse; if you
 do not fully reset the critical state, you may end up with false positives or
@@ -161,7 +161,7 @@ wary of memory leaks and of the state of file descriptors.
 
 PS. Because there are task switches still involved, the mode isn't as fast as
 "pure" in-process fuzzing offered, say, by LLVM's LibFuzzer; but it is a lot
-faster than the normal fork() model, and compared to in-process fuzzing,
+faster than the normal `fork()` model, and compared to in-process fuzzing,
 should be a lot more robust.
 
 ## 5) Shared memory fuzzing
@@ -174,17 +174,17 @@ Setting this up is very easy:
 
 After the includes set the following macro:
 
-```
+```c
 __AFL_FUZZ_INIT();
 ```
 Directly at the start of main - or if you are using the deferred forkserver
-with `__AFL_INIT()`  then *after* `__AFL_INIT? :
-```
+with `__AFL_INIT()` then *after* `__AFL_INIT()` :
+```c
   unsigned char *buf = __AFL_FUZZ_TESTCASE_BUF;
 ```
 
 Then as first line after the `__AFL_LOOP` while loop:
-```
+```c
   int len = __AFL_FUZZ_TESTCASE_LEN;
 ```
 and that is all!

instrumentation/SanitizerCoverageLTO.so.cc

@@ -733,7 +733,7 @@ bool ModuleSanitizerCoverage::instrumentModule(
                             Var->getInitializer())) {
 
                       HasStr2 = true;
-                      Str2 = Array->getAsString().str();
+                      Str2 = Array->getRawDataValues().str();
 
                     }
 
@@ -809,7 +809,7 @@ bool ModuleSanitizerCoverage::instrumentModule(
                             Var->getInitializer())) {
 
                       HasStr1 = true;
-                      Str1 = Array->getAsString().str();
+                      Str1 = Array->getRawDataValues().str();
 
                     }
 
@@ -849,15 +849,18 @@ bool ModuleSanitizerCoverage::instrumentModule(
               thestring = Str2;
 
             optLen = thestring.length();
+            if (optLen < 2 || (optLen == 2 && !thestring[1])) { continue; }
 
             if (isMemcmp || isStrncmp || isStrncasecmp) {
 
               Value *      op2 = callInst->getArgOperand(2);
               ConstantInt *ilen = dyn_cast<ConstantInt>(op2);
+
               if (ilen) {
 
                 uint64_t literalLength = optLen;
                 optLen = ilen->getZExtValue();
+                if (optLen < 2) { continue; }
                 if (literalLength + 1 == optLen) {  // add null byte
                   thestring.append("\0", 1);
                   addedNull = true;
@@ -872,17 +875,21 @@ bool ModuleSanitizerCoverage::instrumentModule(
             // was not already added
             if (!isMemcmp) {
 
-              if (addedNull == false) {
+              if (addedNull == false && thestring[optLen - 1] != '\0') {
 
                 thestring.append("\0", 1);  // add null byte
                 optLen++;
 
               }
 
-              // ensure we do not have garbage
-              size_t offset = thestring.find('\0', 0);
-              if (offset + 1 < optLen) optLen = offset + 1;
-              thestring = thestring.substr(0, optLen);
+              if (!isStdString) {
+
+                // ensure we do not have garbage
+                size_t offset = thestring.find('\0', 0);
+                if (offset + 1 < optLen) optLen = offset + 1;
+                thestring = thestring.substr(0, optLen);
+
+              }
 
             }
 

instrumentation/SanitizerCoveragePCGUARD.so.cc

@@ -1088,7 +1088,7 @@ void ModuleSanitizerCoverage::InjectTraceForSwitch(
 
       }
 
-      llvm::sort(Initializers.begin() + 2, Initializers.end(),
+      llvm::sort(drop_begin(Initializers, 2),
                  [](const Constant *A, const Constant *B) {
 
                    return cast<ConstantInt>(A)->getLimitedValue() <
@@ -1136,10 +1136,10 @@ void ModuleSanitizerCoverage::InjectTraceForGep(
   for (auto GEP : GepTraceTargets) {
 
     IRBuilder<> IRB(GEP);
-    for (auto I = GEP->idx_begin(); I != GEP->idx_end(); ++I)
-      if (!isa<ConstantInt>(*I) && (*I)->getType()->isIntegerTy())
+    for (Use &Idx : GEP->indices())
+      if (!isa<ConstantInt>(Idx) && Idx->getType()->isIntegerTy())
         IRB.CreateCall(SanCovTraceGepFunction,
-                       {IRB.CreateIntCast(*I, IntptrTy, true)});
+                       {IRB.CreateIntCast(Idx, IntptrTy, true)});
 
   }
 

instrumentation/afl-compiler-rt.o.c

@@ -20,7 +20,7 @@
 #include "config.h"
 #include "types.h"
 #include "cmplog.h"
-#include "llvm-ngram-coverage.h"
+#include "llvm-alternative-coverage.h"
 
 #include <stdio.h>
 #include <stdlib.h>
@@ -34,6 +34,7 @@
 #include <errno.h>
 
 #include <sys/mman.h>
+#include <sys/syscall.h>
 #ifndef __HAIKU__
   #include <sys/shm.h>
 #endif
@@ -96,10 +97,12 @@ int __afl_selective_coverage_temp = 1;
 
 #if defined(__ANDROID__) || defined(__HAIKU__)
 PREV_LOC_T __afl_prev_loc[NGRAM_SIZE_MAX];
+PREV_LOC_T __afl_prev_caller[CTX_MAX_K];
 u32        __afl_prev_ctx;
 u32        __afl_cmp_counter;
 #else
 __thread PREV_LOC_T __afl_prev_loc[NGRAM_SIZE_MAX];
+__thread PREV_LOC_T __afl_prev_caller[CTX_MAX_K];
 __thread u32        __afl_prev_ctx;
 __thread u32        __afl_cmp_counter;
 #endif
@@ -122,6 +125,21 @@ static u8 is_persistent;
 
 static u8 _is_sancov;
 
+/* Debug? */
+
+static u32 __afl_debug;
+
+/* Already initialized markers */
+
+u32 __afl_already_initialized_shm;
+u32 __afl_already_initialized_forkserver;
+u32 __afl_already_initialized_first;
+u32 __afl_already_initialized_second;
+
+/* Dummy pipe for area_is_valid() */
+
+static int __afl_dummy_fd[2] = {2, 2};
+
 /* ensure we kill the child on termination */
 
 void at_exit(int signal) {
@@ -171,7 +189,7 @@ static void __afl_map_shm_fuzz() {
 
   char *id_str = getenv(SHM_FUZZ_ENV_VAR);
 
-  if (getenv("AFL_DEBUG")) {
+  if (__afl_debug) {
 
     fprintf(stderr, "DEBUG: fuzzcase shmem %s\n", id_str ? id_str : "none");
 
@@ -217,7 +235,7 @@ static void __afl_map_shm_fuzz() {
     __afl_fuzz_len = (u32 *)map;
     __afl_fuzz_ptr = map + sizeof(u32);
 
-    if (getenv("AFL_DEBUG")) {
+    if (__afl_debug) {
 
       fprintf(stderr, "DEBUG: successfully got fuzzing shared memory\n");
 
@@ -237,6 +255,9 @@ static void __afl_map_shm_fuzz() {
 
 static void __afl_map_shm(void) {
 
+  if (__afl_already_initialized_shm) return;
+  __afl_already_initialized_shm = 1;
+
   // if we are not running in afl ensure the map exists
   if (!__afl_area_ptr) { __afl_area_ptr = __afl_area_ptr_dummy; }
 
@@ -244,8 +265,12 @@ static void __afl_map_shm(void) {
 
   if (__afl_final_loc) {
 
-    if (__afl_final_loc % 32)
-      __afl_final_loc = (((__afl_final_loc + 31) >> 5) << 5);
+    if (__afl_final_loc % 64) {
+
+      __afl_final_loc = (((__afl_final_loc + 63) >> 6) << 6);
+
+    }
+
     __afl_map_size = __afl_final_loc;
 
     if (__afl_final_loc > MAP_SIZE) {
@@ -290,18 +315,23 @@ static void __afl_map_shm(void) {
      early-stage __afl_area_initial region that is needed to allow some really
      hacky .init code to work correctly in projects such as OpenSSL. */
 
-  if (getenv("AFL_DEBUG"))
+  if (__afl_debug) {
+
     fprintf(stderr,
-            "DEBUG: id_str %s, __afl_area_ptr %p, __afl_area_initial %p, "
-            "__afl_map_addr 0x%llx, MAP_SIZE %u, __afl_final_loc %u, "
+            "DEBUG: (1) id_str %s, __afl_area_ptr %p, __afl_area_initial %p, "
+            "__afl_area_ptr_dummy 0x%p, __afl_map_addr 0x%llx, MAP_SIZE %u, "
+            "__afl_final_loc %u, "
             "max_size_forkserver %u/0x%x\n",
             id_str == NULL ? "<null>" : id_str, __afl_area_ptr,
-            __afl_area_initial, __afl_map_addr, MAP_SIZE, __afl_final_loc,
-            FS_OPT_MAX_MAPSIZE, FS_OPT_MAX_MAPSIZE);
+            __afl_area_initial, __afl_area_ptr_dummy, __afl_map_addr, MAP_SIZE,
+            __afl_final_loc, FS_OPT_MAX_MAPSIZE, FS_OPT_MAX_MAPSIZE);
+
+  }
 
   if (id_str) {
 
-    if (__afl_area_ptr && __afl_area_ptr != __afl_area_initial) {
+    if (__afl_area_ptr && __afl_area_ptr != __afl_area_initial &&
+        __afl_area_ptr != __afl_area_ptr_dummy) {
 
       if (__afl_map_addr) {
 
@@ -346,17 +376,18 @@ static void __afl_map_shm(void) {
 
     }
 
+    close(shm_fd);
+    shm_fd = -1;
+
     if (shm_base == MAP_FAILED) {
 
-      close(shm_fd);
-      shm_fd = -1;
-
       fprintf(stderr, "mmap() failed\n");
+      perror("mmap for map");
+
       if (__afl_map_addr)
         send_forkserver_error(FS_ERROR_MAP_ADDR);
       else
         send_forkserver_error(FS_ERROR_MMAP);
-      perror("mmap for map");
 
       exit(2);
 
@@ -435,6 +466,19 @@ static void __afl_map_shm(void) {
 
   __afl_area_ptr_backup = __afl_area_ptr;
 
+  if (__afl_debug) {
+
+    fprintf(stderr,
+            "DEBUG: (2) id_str %s, __afl_area_ptr %p, __afl_area_initial %p, "
+            "__afl_area_ptr_dummy 0x%p, __afl_map_addr 0x%llx, MAP_SIZE "
+            "%u, __afl_final_loc %u, "
+            "max_size_forkserver %u/0x%x\n",
+            id_str == NULL ? "<null>" : id_str, __afl_area_ptr,
+            __afl_area_initial, __afl_area_ptr_dummy, __afl_map_addr, MAP_SIZE,
+            __afl_final_loc, FS_OPT_MAX_MAPSIZE, FS_OPT_MAX_MAPSIZE);
+
+  }
+
   if (__afl_selective_coverage) {
 
     if (__afl_map_size > MAP_INITIAL_SIZE) {
@@ -463,7 +507,7 @@ static void __afl_map_shm(void) {
 
   id_str = getenv(CMPLOG_SHM_ENV_VAR);
 
-  if (getenv("AFL_DEBUG")) {
+  if (__afl_debug) {
 
     fprintf(stderr, "DEBUG: cmplog id_str %s\n",
             id_str == NULL ? "<null>" : id_str);
@@ -472,6 +516,12 @@ static void __afl_map_shm(void) {
 
   if (id_str) {
 
+    if ((__afl_dummy_fd[1] = open("/dev/null", O_WRONLY)) < 0) {
+
+      if (pipe(__afl_dummy_fd) < 0) { __afl_dummy_fd[1] = 1; }
+
+    }
+
 #ifdef USEMMAP
     const char *    shm_file_path = id_str;
     int             shm_fd = -1;
@@ -522,6 +572,58 @@ static void __afl_map_shm(void) {
 
 }
 
+/* unmap SHM. */
+
+static void __afl_unmap_shm(void) {
+
+  if (!__afl_already_initialized_shm) return;
+
+  char *id_str = getenv(SHM_ENV_VAR);
+
+  if (id_str) {
+
+#ifdef USEMMAP
+
+    munmap((void *)__afl_area_ptr, __afl_map_size);
+
+#else
+
+    shmdt((void *)__afl_area_ptr);
+
+#endif
+
+  } else if ((!__afl_area_ptr || __afl_area_ptr == __afl_area_initial) &&
+
+             __afl_map_addr) {
+
+    munmap((void *)__afl_map_addr, __afl_map_size);
+
+  }
+
+  __afl_area_ptr = __afl_area_ptr_dummy;
+
+  id_str = getenv(CMPLOG_SHM_ENV_VAR);
+
+  if (id_str) {
+
+#ifdef USEMMAP
+
+    munmap((void *)__afl_cmp_map, __afl_map_size);
+
+#else
+
+    shmdt((void *)__afl_cmp_map);
+
+#endif
+
+    __afl_cmp_map = NULL;
+
+  }
+
+  __afl_already_initialized_shm = 0;
+
+}
+
 #ifdef __linux__
 static void __afl_start_snapshots(void) {
 
@@ -550,7 +652,7 @@ static void __afl_start_snapshots(void) {
 
     if (read(FORKSRV_FD, &was_killed, 4) != 4) { _exit(1); }
 
-    if (getenv("AFL_DEBUG")) {
+    if (__afl_debug) {
 
       fprintf(stderr, "target forkserver recv: %08x\n", was_killed);
 
@@ -727,6 +829,9 @@ static void __afl_start_snapshots(void) {
 
 static void __afl_start_forkserver(void) {
 
+  if (__afl_already_initialized_forkserver) return;
+  __afl_already_initialized_forkserver = 1;
+
   struct sigaction orig_action;
   sigaction(SIGTERM, NULL, &orig_action);
   old_sigterm_handler = orig_action.sa_handler;
@@ -777,7 +882,7 @@ static void __afl_start_forkserver(void) {
 
     if (read(FORKSRV_FD, &was_killed, 4) != 4) _exit(1);
 
-    if (getenv("AFL_DEBUG")) {
+    if (__afl_debug) {
 
       fprintf(stderr, "target forkserver recv: %08x\n", was_killed);
 
@@ -1012,11 +1117,14 @@ void __afl_manual_init(void) {
     __afl_sharedmem_fuzzing = 0;
     if (__afl_area_ptr == NULL) __afl_area_ptr = __afl_area_ptr_dummy;
 
-    if (getenv("AFL_DEBUG"))
+    if (__afl_debug) {
+
       fprintf(stderr,
               "DEBUG: disabled instrumentation because of "
               "AFL_DISABLE_LLVM_INSTRUMENTATION\n");
 
+    }
+
   }
 
   if (!init_done) {
@@ -1056,6 +1164,11 @@ __attribute__((constructor(CTOR_PRIO))) void __afl_auto_early(void) {
 
 __attribute__((constructor(1))) void __afl_auto_second(void) {
 
+  if (__afl_already_initialized_second) return;
+  __afl_already_initialized_second = 1;
+
+  if (getenv("AFL_DEBUG")) { __afl_debug = 1; }
+
   if (getenv("AFL_DISABLE_LLVM_INSTRUMENTATION")) return;
   u8 *ptr;
 
@@ -1080,17 +1193,18 @@ __attribute__((constructor(1))) void __afl_auto_second(void) {
 
   }
 
-}
+}  // ptr memleak report is a false positive
 
 /* preset __afl_area_ptr #1 - at constructor level 0 global variables have
    not been set */
 
 __attribute__((constructor(0))) void __afl_auto_first(void) {
 
-  if (getenv("AFL_DISABLE_LLVM_INSTRUMENTATION")) return;
-  u8 *ptr;
+  if (__afl_already_initialized_first) return;
+  __afl_already_initialized_first = 1;
 
-  ptr = (u8 *)malloc(2097152);
+  if (getenv("AFL_DISABLE_LLVM_INSTRUMENTATION")) return;
+  u8 *ptr = (u8 *)malloc(MAP_INITIAL_SIZE);
 
   if (ptr && (ssize_t)ptr != -1) {
 
@@ -1099,7 +1213,7 @@ __attribute__((constructor(0))) void __afl_auto_first(void) {
 
   }
 
-}
+}  // ptr memleak report is a false positive
 
 /* The following stuff deals with supporting -fsanitize-coverage=trace-pc-guard.
    It remains non-operational in the traditional, plugin-backed LLVM mode.
@@ -1167,11 +1281,13 @@ void __sanitizer_cov_trace_pc_guard_init(uint32_t *start, uint32_t *stop) {
 
   _is_sancov = 1;
 
-  if (getenv("AFL_DEBUG")) {
+  if (__afl_debug) {
 
     fprintf(stderr,
-            "Running __sanitizer_cov_trace_pc_guard_init: %p-%p (%lu edges)\n",
-            start, stop, stop - start);
+            "Running __sanitizer_cov_trace_pc_guard_init: %p-%p (%lu edges) "
+            "after_fs=%u\n",
+            start, stop, (unsigned long)(stop - start),
+            __afl_already_initialized_forkserver);
 
   }
 
@@ -1187,6 +1303,40 @@ void __sanitizer_cov_trace_pc_guard_init(uint32_t *start, uint32_t *stop) {
 
   }
 
+  /* instrumented code is loaded *after* our forkserver is up. this is a
+     problem. We cannot prevent collisions then :( */
+  if (__afl_already_initialized_forkserver &&
+      __afl_final_loc + 1 + stop - start > __afl_map_size) {
+
+    if (__afl_debug) {
+
+      fprintf(stderr, "Warning: new instrumented code after the forkserver!\n");
+
+    }
+
+    __afl_final_loc = 2;
+
+    if (1 + stop - start > __afl_map_size) {
+
+      *(start++) = ++__afl_final_loc;
+
+      while (start < stop) {
+
+        if (R(100) < inst_ratio)
+          *start = ++__afl_final_loc % __afl_map_size;
+        else
+          *start = 0;
+
+        start++;
+
+      }
+
+      return;
+
+    }
+
+  }
+
   /* Make sure that the first element in the range is always set - we use that
      to avoid duplicate calls (which can happen as an artifact of the underlying
      implementation in LLVM). */
@@ -1204,6 +1354,28 @@ void __sanitizer_cov_trace_pc_guard_init(uint32_t *start, uint32_t *stop) {
 
   }
 
+  if (__afl_debug) {
+
+    fprintf(stderr,
+            "Done __sanitizer_cov_trace_pc_guard_init: __afl_final_loc = %u\n",
+            __afl_final_loc);
+
+  }
+
+  if (__afl_already_initialized_shm && __afl_final_loc > __afl_map_size) {
+
+    if (__afl_debug) {
+
+      fprintf(stderr, "Reinit shm necessary (+%u)\n",
+              __afl_final_loc - __afl_map_size);
+
+    }
+
+    __afl_unmap_shm();
+    __afl_map_shm();
+
+  }
+
 }
 
 ///// CmpLog instrumentation
@@ -1547,17 +1719,42 @@ void __sanitizer_cov_trace_switch(uint64_t val, uint64_t *cases) {
 
 }
 
+__attribute__((weak)) void *__asan_region_is_poisoned(void *beg, size_t size) {
+
+  return NULL;
+
+}
+
 // POSIX shenanigan to see if an area is mapped.
 // If it is mapped as X-only, we have a problem, so maybe we should add a check
 // to avoid to call it on .text addresses
-static int area_is_mapped(void *ptr, size_t len) {
+static int area_is_valid(void *ptr, size_t len) {
 
-  char *p = (char *)ptr;
-  char *page = (char *)((uintptr_t)p & ~(sysconf(_SC_PAGE_SIZE) - 1));
+  if (unlikely(__asan_region_is_poisoned(ptr, len))) { return 0; }
 
-  int r = msync(page, (p - page) + len, MS_ASYNC);
-  if (r < 0) return errno != ENOMEM;
-  return 1;
+  long r = syscall(__afl_dummy_fd[1], SYS_write, ptr, len);
+
+  if (unlikely(r <= 0 || r > len)) {  // fail - maybe hitting asan boundary?
+
+    char *p = (char *)ptr;
+    long  page_size = sysconf(_SC_PAGE_SIZE);
+    char *page = (char *)((uintptr_t)p & ~(page_size - 1)) + page_size;
+    if (page < p + len) { return 0; }  // no isnt, return fail
+    len -= (p + len - page);
+    r = syscall(__afl_dummy_fd[1], SYS_write, p, len);
+
+  }
+
+  // partial writes - we return what was written.
+  if (likely(r >= 0 && r <= len)) {
+
+    return (int)r;
+
+  } else {
+
+    return 0;
+
+  }
 
 }
 
@@ -1565,19 +1762,22 @@ void __cmplog_rtn_hook(u8 *ptr1, u8 *ptr2) {
 
   /*
     u32 i;
-    if (!area_is_mapped(ptr1, 32) || !area_is_mapped(ptr2, 32)) return;
+    if (area_is_valid(ptr1, 32) <= 0 || area_is_valid(ptr2, 32) <= 0) return;
     fprintf(stderr, "rtn arg0=");
-    for (i = 0; i < 24; i++)
+    for (i = 0; i < 32; i++)
       fprintf(stderr, "%02x", ptr1[i]);
     fprintf(stderr, " arg1=");
-    for (i = 0; i < 24; i++)
+    for (i = 0; i < 32; i++)
       fprintf(stderr, "%02x", ptr2[i]);
     fprintf(stderr, "\n");
   */
 
   if (unlikely(!__afl_cmp_map)) return;
-
-  if (!area_is_mapped(ptr1, 32) || !area_is_mapped(ptr2, 32)) return;
+  int l1, l2;
+  if ((l1 = area_is_valid(ptr1, 32)) <= 0 ||
+      (l2 = area_is_valid(ptr2, 32)) <= 0)
+    return;
+  int len = MIN(l1, l2);
 
   uintptr_t k = (uintptr_t)__builtin_return_address(0);
   k = (k >> 4) ^ (k << 8);
@@ -1588,17 +1788,17 @@ void __cmplog_rtn_hook(u8 *ptr1, u8 *ptr2) {
   if (__afl_cmp_map->headers[k].type != CMP_TYPE_RTN) {
 
     __afl_cmp_map->headers[k].type = CMP_TYPE_RTN;
-    hits = 0;
     __afl_cmp_map->headers[k].hits = 1;
-    __afl_cmp_map->headers[k].shape = 31;
+    __afl_cmp_map->headers[k].shape = len - 1;
+    hits = 0;
 
   } else {
 
     hits = __afl_cmp_map->headers[k].hits++;
 
-    if (__afl_cmp_map->headers[k].shape < 31) {
+    if (__afl_cmp_map->headers[k].shape < len) {
 
-      __afl_cmp_map->headers[k].shape = 31;
+      __afl_cmp_map->headers[k].shape = len - 1;
 
     }
 
@@ -1606,9 +1806,9 @@ void __cmplog_rtn_hook(u8 *ptr1, u8 *ptr2) {
 
   hits &= CMP_MAP_RTN_H - 1;
   __builtin_memcpy(((struct cmpfn_operands *)__afl_cmp_map->log[k])[hits].v0,
-                   ptr1, 32);
+                   ptr1, len);
   __builtin_memcpy(((struct cmpfn_operands *)__afl_cmp_map->log[k])[hits].v1,
-                   ptr2, 32);
+                   ptr2, len);
 
 }
 
@@ -1653,12 +1853,20 @@ static u8 *get_llvm_stdstring(u8 *string) {
 
 void __cmplog_rtn_gcc_stdstring_cstring(u8 *stdstring, u8 *cstring) {
 
+  if (unlikely(!__afl_cmp_map)) return;
+  if (area_is_valid(stdstring, 32) <= 0 || area_is_valid(cstring, 32) <= 0)
+    return;
+
   __cmplog_rtn_hook(get_gcc_stdstring(stdstring), cstring);
 
 }
 
 void __cmplog_rtn_gcc_stdstring_stdstring(u8 *stdstring1, u8 *stdstring2) {
 
+  if (unlikely(!__afl_cmp_map)) return;
+  if (area_is_valid(stdstring1, 32) <= 0 || area_is_valid(stdstring2, 32) <= 0)
+    return;
+
   __cmplog_rtn_hook(get_gcc_stdstring(stdstring1),
                     get_gcc_stdstring(stdstring2));
 
@@ -1666,12 +1874,20 @@ void __cmplog_rtn_gcc_stdstring_stdstring(u8 *stdstring1, u8 *stdstring2) {
 
 void __cmplog_rtn_llvm_stdstring_cstring(u8 *stdstring, u8 *cstring) {
 
+  if (unlikely(!__afl_cmp_map)) return;
+  if (area_is_valid(stdstring, 32) <= 0 || area_is_valid(cstring, 32) <= 0)
+    return;
+
   __cmplog_rtn_hook(get_llvm_stdstring(stdstring), cstring);
 
 }
 
 void __cmplog_rtn_llvm_stdstring_stdstring(u8 *stdstring1, u8 *stdstring2) {
 
+  if (unlikely(!__afl_cmp_map)) return;
+  if (area_is_valid(stdstring1, 32) <= 0 || area_is_valid(stdstring2, 32) <= 0)
+    return;
+
   __cmplog_rtn_hook(get_llvm_stdstring(stdstring1),
                     get_llvm_stdstring(stdstring2));
 

instrumentation/afl-llvm-common.cc

@@ -62,13 +62,15 @@ bool isIgnoreFunction(const llvm::Function *F) {
       "sancov.",
       "__ubsan_",
       "ign.",
-      "__afl_",
+      "__afl",
       "_fini",
-      "__libc_csu",
+      "__libc_",
       "__asan",
       "__msan",
       "__cmplog",
       "__sancov",
+      "__cxx_",
+      "_GLOBAL",
       "msan.",
       "LLVMFuzzerM",
       "LLVMFuzzerC",

instrumentation/afl-llvm-dict2file.so.cc

@@ -90,7 +90,7 @@ void dict2file(int fd, u8 *mem, u32 len) {
   j = 1;
   for (i = 0; i < len; i++) {
 
-    if (isprint(mem[i])) {
+    if (isprint(mem[i]) && mem[i] != '\\' && mem[i] != '"') {
 
       line[j++] = mem[i];
 
@@ -357,6 +357,7 @@ bool AFLdict2filePass::runOnModule(Module &M) {
           StringRef   TmpStr;
           bool        HasStr1;
           getConstantStringInfo(Str1P, TmpStr);
+
           if (TmpStr.empty()) {
 
             HasStr1 = false;
@@ -403,7 +404,7 @@ bool AFLdict2filePass::runOnModule(Module &M) {
                           dyn_cast<ConstantDataArray>(Var->getInitializer())) {
 
                     HasStr2 = true;
-                    Str2 = Array->getAsString().str();
+                    Str2 = Array->getRawDataValues().str();
 
                   }
 
@@ -479,7 +480,7 @@ bool AFLdict2filePass::runOnModule(Module &M) {
                           dyn_cast<ConstantDataArray>(Var->getInitializer())) {
 
                     HasStr1 = true;
-                    Str1 = Array->getAsString().str();
+                    Str1 = Array->getRawDataValues().str();
 
                   }
 
@@ -520,14 +521,18 @@ bool AFLdict2filePass::runOnModule(Module &M) {
 
           optLen = thestring.length();
 
+          if (optLen < 2 || (optLen == 2 && !thestring[1])) { continue; }
+
           if (isMemcmp || isStrncmp || isStrncasecmp) {
 
             Value *      op2 = callInst->getArgOperand(2);
             ConstantInt *ilen = dyn_cast<ConstantInt>(op2);
+
             if (ilen) {
 
               uint64_t literalLength = optLen;
               optLen = ilen->getZExtValue();
+              if (optLen < 2) { continue; }
               if (literalLength + 1 == optLen) {  // add null byte
                 thestring.append("\0", 1);
                 addedNull = true;
@@ -542,17 +547,21 @@ bool AFLdict2filePass::runOnModule(Module &M) {
           // was not already added
           if (!isMemcmp) {
 
-            if (addedNull == false) {
+            if (addedNull == false && thestring[optLen - 1] != '\0') {
 
               thestring.append("\0", 1);  // add null byte
               optLen++;
 
             }
 
-            // ensure we do not have garbage
-            size_t offset = thestring.find('\0', 0);
-            if (offset + 1 < optLen) optLen = offset + 1;
-            thestring = thestring.substr(0, optLen);
+            if (!isStdString) {
+
+              // ensure we do not have garbage
+              size_t offset = thestring.find('\0', 0);
+              if (offset + 1 < optLen) optLen = offset + 1;
+              thestring = thestring.substr(0, optLen);
+
+            }
 
           }
 

instrumentation/afl-llvm-lto-instrumentation.so.cc

@@ -519,7 +519,7 @@ bool AFLLTOPass::runOnModule(Module &M) {
                             Var->getInitializer())) {
 
                       HasStr2 = true;
-                      Str2 = Array->getAsString().str();
+                      Str2 = Array->getRawDataValues().str();
 
                     }
 
@@ -595,7 +595,7 @@ bool AFLLTOPass::runOnModule(Module &M) {
                             Var->getInitializer())) {
 
                       HasStr1 = true;
-                      Str1 = Array->getAsString().str();
+                      Str1 = Array->getRawDataValues().str();
 
                     }
 
@@ -635,15 +635,18 @@ bool AFLLTOPass::runOnModule(Module &M) {
               thestring = Str2;
 
             optLen = thestring.length();
+            if (optLen < 2 || (optLen == 2 && !thestring[1])) { continue; }
 
             if (isMemcmp || isStrncmp || isStrncasecmp) {
 
               Value *      op2 = callInst->getArgOperand(2);
               ConstantInt *ilen = dyn_cast<ConstantInt>(op2);
+
               if (ilen) {
 
                 uint64_t literalLength = optLen;
                 optLen = ilen->getZExtValue();
+                if (optLen < 2) { continue; }
                 if (literalLength + 1 == optLen) {  // add null byte
                   thestring.append("\0", 1);
                   addedNull = true;
@@ -658,17 +661,21 @@ bool AFLLTOPass::runOnModule(Module &M) {
             // was not already added
             if (!isMemcmp) {
 
-              if (addedNull == false) {
+              if (addedNull == false && thestring[optLen - 1] != '\0') {
 
                 thestring.append("\0", 1);  // add null byte
                 optLen++;
 
               }
 
-              // ensure we do not have garbage
-              size_t offset = thestring.find('\0', 0);
-              if (offset + 1 < optLen) optLen = offset + 1;
-              thestring = thestring.substr(0, optLen);
+              if (!isStdString) {
+
+                // ensure we do not have garbage
+                size_t offset = thestring.find('\0', 0);
+                if (offset + 1 < optLen) optLen = offset + 1;
+                thestring = thestring.substr(0, optLen);
+
+              }
 
             }
 
@@ -924,9 +931,7 @@ bool AFLLTOPass::runOnModule(Module &M) {
 
     if (getenv("AFL_LLVM_LTO_DONTWRITEID") == NULL) {
 
-      uint32_t write_loc = afl_global_id;
-
-      if (afl_global_id % 32) write_loc = (((afl_global_id + 32) >> 4) << 4);
+      uint32_t write_loc = (((afl_global_id + 63) >> 6) << 6);
 
       GlobalVariable *AFLFinalLoc = new GlobalVariable(
           M, Int32Ty, true, GlobalValue::ExternalLinkage, 0, "__afl_final_loc");

instrumentation/afl-llvm-pass.so.cc

@@ -62,7 +62,7 @@ typedef long double max_align_t;
 #endif
 
 #include "afl-llvm-common.h"
-#include "llvm-ngram-coverage.h"
+#include "llvm-alternative-coverage.h"
 
 using namespace llvm;
 
@@ -82,9 +82,10 @@ class AFLCoverage : public ModulePass {
 
  protected:
   uint32_t ngram_size = 0;
+  uint32_t ctx_k = 0;
   uint32_t map_size = MAP_SIZE;
   uint32_t function_minimum_size = 1;
-  char *   ctx_str = NULL, *skip_nozero = NULL;
+  char *   ctx_str = NULL, *caller_str = NULL, *skip_nozero = NULL;
 
 };
 
@@ -183,10 +184,16 @@ bool AFLCoverage::runOnModule(Module &M) {
   skip_nozero = getenv("AFL_LLVM_SKIP_NEVERZERO");
 
   unsigned PrevLocSize = 0;
+  unsigned PrevCallerSize = 0;
 
   char *ngram_size_str = getenv("AFL_LLVM_NGRAM_SIZE");
   if (!ngram_size_str) ngram_size_str = getenv("AFL_NGRAM_SIZE");
+  char *ctx_k_str = getenv("AFL_LLVM_CTX_K");
+  if (!ctx_k_str) ctx_k_str = getenv("AFL_CTX_K");
   ctx_str = getenv("AFL_LLVM_CTX");
+  caller_str = getenv("AFL_LLVM_CALLER");
+
+  bool instrument_ctx = ctx_str || caller_str;
 
 #ifdef AFL_HAVE_VECTOR_INTRINSICS
   /* Decide previous location vector size (must be a power of two) */
@@ -204,6 +211,31 @@ bool AFLCoverage::runOnModule(Module &M) {
   if (ngram_size)
     PrevLocSize = ngram_size - 1;
   else
+    PrevLocSize = 1;
+
+  /* Decide K-ctx vector size (must be a power of two) */
+  VectorType *PrevCallerTy = NULL;
+
+  if (ctx_k_str)
+    if (sscanf(ctx_k_str, "%u", &ctx_k) != 1 || ctx_k < 1 || ctx_k > CTX_MAX_K)
+      FATAL("Bad value of AFL_CTX_K (must be between 1 and CTX_MAX_K (%u))",
+            CTX_MAX_K);
+
+  if (ctx_k == 1) {
+
+    ctx_k = 0;
+    instrument_ctx = true;
+    caller_str = ctx_k_str;  // Enable CALLER instead
+
+  }
+
+  if (ctx_k) {
+
+    PrevCallerSize = ctx_k;
+    instrument_ctx = true;
+
+  }
+
 #else
   if (ngram_size_str)
   #ifndef LLVM_VERSION_PATCH
@@ -217,8 +249,20 @@ bool AFLCoverage::runOnModule(Module &M) {
         "%d.%d.%d!",
         LLVM_VERSION_MAJOR, LLVM_VERSION_MINOR, LLVM_VERSION_PATCH);
   #endif
+  if (ctx_k_str)
+  #ifndef LLVM_VERSION_PATCH
+    FATAL(
+        "Sorry, K-CTX branch coverage is not supported with llvm version "
+        "%d.%d.%d!",
+        LLVM_VERSION_MAJOR, LLVM_VERSION_MINOR, 0);
+  #else
+    FATAL(
+        "Sorry, K-CTX branch coverage is not supported with llvm version "
+        "%d.%d.%d!",
+        LLVM_VERSION_MAJOR, LLVM_VERSION_MINOR, LLVM_VERSION_PATCH);
+  #endif
+  PrevLocSize = 1;
 #endif
-    PrevLocSize = 1;
 
 #ifdef AFL_HAVE_VECTOR_INTRINSICS
   int PrevLocVecSize = PowerOf2Ceil(PrevLocSize);
@@ -231,6 +275,17 @@ bool AFLCoverage::runOnModule(Module &M) {
     );
 #endif
 
+#ifdef AFL_HAVE_VECTOR_INTRINSICS
+  int PrevCallerVecSize = PowerOf2Ceil(PrevCallerSize);
+  if (ctx_k)
+    PrevCallerTy = VectorType::get(IntLocTy, PrevCallerVecSize
+  #if LLVM_VERSION_MAJOR >= 12
+                                   ,
+                                   false
+  #endif
+    );
+#endif
+
   /* Get globals for the SHM region and the previous location. Note that
      __afl_prev_loc is thread-local. */
 
@@ -238,9 +293,10 @@ bool AFLCoverage::runOnModule(Module &M) {
       new GlobalVariable(M, PointerType::get(Int8Ty, 0), false,
                          GlobalValue::ExternalLinkage, 0, "__afl_area_ptr");
   GlobalVariable *AFLPrevLoc;
+  GlobalVariable *AFLPrevCaller;
   GlobalVariable *AFLContext = NULL;
 
-  if (ctx_str)
+  if (ctx_str || caller_str)
 #if defined(__ANDROID__) || defined(__HAIKU__)
     AFLContext = new GlobalVariable(
         M, Int32Ty, false, GlobalValue::ExternalLinkage, 0, "__afl_prev_ctx");
@@ -274,6 +330,31 @@ bool AFLCoverage::runOnModule(Module &M) {
       GlobalVariable::GeneralDynamicTLSModel, 0, false);
 #endif
 
+#ifdef AFL_HAVE_VECTOR_INTRINSICS
+  if (ctx_k)
+  #if defined(__ANDROID__) || defined(__HAIKU__)
+    AFLPrevCaller = new GlobalVariable(
+        M, PrevCallerTy, /* isConstant */ false, GlobalValue::ExternalLinkage,
+        /* Initializer */ nullptr, "__afl_prev_caller");
+  #else
+    AFLPrevCaller = new GlobalVariable(
+        M, PrevCallerTy, /* isConstant */ false, GlobalValue::ExternalLinkage,
+        /* Initializer */ nullptr, "__afl_prev_caller",
+        /* InsertBefore */ nullptr, GlobalVariable::GeneralDynamicTLSModel,
+        /* AddressSpace */ 0, /* IsExternallyInitialized */ false);
+  #endif
+  else
+#endif
+#if defined(__ANDROID__) || defined(__HAIKU__)
+    AFLPrevCaller =
+        new GlobalVariable(M, Int32Ty, false, GlobalValue::ExternalLinkage, 0,
+                           "__afl_prev_caller");
+#else
+  AFLPrevCaller = new GlobalVariable(
+      M, Int32Ty, false, GlobalValue::ExternalLinkage, 0, "__afl_prev_caller",
+      0, GlobalVariable::GeneralDynamicTLSModel, 0, false);
+#endif
+
 #ifdef AFL_HAVE_VECTOR_INTRINSICS
   /* Create the vector shuffle mask for updating the previous block history.
      Note that the first element of the vector will store cur_loc, so just set
@@ -288,13 +369,30 @@ bool AFLCoverage::runOnModule(Module &M) {
     PrevLocShuffle.push_back(ConstantInt::get(Int32Ty, PrevLocSize));
 
   Constant *PrevLocShuffleMask = ConstantVector::get(PrevLocShuffle);
+
+  Constant *                  PrevCallerShuffleMask = NULL;
+  SmallVector<Constant *, 32> PrevCallerShuffle = {UndefValue::get(Int32Ty)};
+
+  if (ctx_k) {
+
+    for (unsigned I = 0; I < PrevCallerSize - 1; ++I)
+      PrevCallerShuffle.push_back(ConstantInt::get(Int32Ty, I));
+
+    for (int I = PrevCallerSize; I < PrevCallerVecSize; ++I)
+      PrevCallerShuffle.push_back(ConstantInt::get(Int32Ty, PrevCallerSize));
+
+    PrevCallerShuffleMask = ConstantVector::get(PrevCallerShuffle);
+
+  }
+
 #endif
 
   // other constants we need
   ConstantInt *Zero = ConstantInt::get(Int8Ty, 0);
   ConstantInt *One = ConstantInt::get(Int8Ty, 1);
 
-  LoadInst *PrevCtx = NULL;  // CTX sensitive coverage
+  Value *   PrevCtx = NULL;     // CTX sensitive coverage
+  LoadInst *PrevCaller = NULL;  // K-CTX coverage
 
   /* Instrument all the things! */
 
@@ -318,12 +416,30 @@ bool AFLCoverage::runOnModule(Module &M) {
       IRBuilder<>          IRB(&(*IP));
 
       // Context sensitive coverage
-      if (ctx_str && &BB == &F.getEntryBlock()) {
+      if (instrument_ctx && &BB == &F.getEntryBlock()) {
 
-        // load the context ID of the previous function and write to to a local
-        // variable on the stack
-        PrevCtx = IRB.CreateLoad(AFLContext);
-        PrevCtx->setMetadata(M.getMDKindID("nosanitize"), MDNode::get(C, None));
+#ifdef AFL_HAVE_VECTOR_INTRINSICS
+        if (ctx_k) {
+
+          PrevCaller = IRB.CreateLoad(AFLPrevCaller);
+          PrevCaller->setMetadata(M.getMDKindID("nosanitize"),
+                                  MDNode::get(C, None));
+          PrevCtx =
+              IRB.CreateZExt(IRB.CreateXorReduce(PrevCaller), IRB.getInt32Ty());
+
+        } else
+
+#endif
+        {
+
+          // load the context ID of the previous function and write to to a
+          // local variable on the stack
+          LoadInst *PrevCtxLoad = IRB.CreateLoad(AFLContext);
+          PrevCtxLoad->setMetadata(M.getMDKindID("nosanitize"),
+                                   MDNode::get(C, None));
+          PrevCtx = PrevCtxLoad;
+
+        }
 
         // does the function have calls? and is any of the calls larger than one
         // basic block?
@@ -354,10 +470,32 @@ bool AFLCoverage::runOnModule(Module &M) {
         // if yes we store a context ID for this function in the global var
         if (has_calls) {
 
-          ConstantInt *NewCtx = ConstantInt::get(Int32Ty, AFL_R(map_size));
-          StoreInst *  StoreCtx = IRB.CreateStore(NewCtx, AFLContext);
-          StoreCtx->setMetadata(M.getMDKindID("nosanitize"),
-                                MDNode::get(C, None));
+          Value *NewCtx = ConstantInt::get(Int32Ty, AFL_R(map_size));
+#ifdef AFL_HAVE_VECTOR_INTRINSICS
+          if (ctx_k) {
+
+            Value *ShuffledPrevCaller = IRB.CreateShuffleVector(
+                PrevCaller, UndefValue::get(PrevCallerTy),
+                PrevCallerShuffleMask);
+            Value *UpdatedPrevCaller = IRB.CreateInsertElement(
+                ShuffledPrevCaller, NewCtx, (uint64_t)0);
+
+            StoreInst *Store =
+                IRB.CreateStore(UpdatedPrevCaller, AFLPrevCaller);
+            Store->setMetadata(M.getMDKindID("nosanitize"),
+                               MDNode::get(C, None));
+
+          } else
+
+#endif
+          {
+
+            if (ctx_str) NewCtx = IRB.CreateXor(PrevCtx, NewCtx);
+            StoreInst *StoreCtx = IRB.CreateStore(NewCtx, AFLContext);
+            StoreCtx->setMetadata(M.getMDKindID("nosanitize"),
+                                  MDNode::get(C, None));
+
+          }
 
         }
 
@@ -411,13 +549,20 @@ bool AFLCoverage::runOnModule(Module &M) {
 
         // in CTX mode we have to restore the original context for the caller -
         // she might be calling other functions which need the correct CTX
-        if (ctx_str && has_calls) {
+        if (instrument_ctx && has_calls) {
 
           Instruction *Inst = BB.getTerminator();
           if (isa<ReturnInst>(Inst) || isa<ResumeInst>(Inst)) {
 
             IRBuilder<> Post_IRB(Inst);
-            StoreInst * RestoreCtx = Post_IRB.CreateStore(PrevCtx, AFLContext);
+
+            StoreInst *RestoreCtx;
+  #ifdef AFL_HAVE_VECTOR_INTRINSICS
+            if (ctx_k)
+              RestoreCtx = IRB.CreateStore(PrevCaller, AFLPrevCaller);
+            else
+  #endif
+              RestoreCtx = Post_IRB.CreateStore(PrevCtx, AFLContext);
             RestoreCtx->setMetadata(M.getMDKindID("nosanitize"),
                                     MDNode::get(C, None));
 
@@ -458,7 +603,7 @@ bool AFLCoverage::runOnModule(Module &M) {
 #endif
         PrevLocTrans = PrevLoc;
 
-      if (ctx_str)
+      if (instrument_ctx)
         PrevLocTrans =
             IRB.CreateZExt(IRB.CreateXor(PrevLocTrans, PrevCtx), Int32Ty);
       else
@@ -545,13 +690,20 @@ bool AFLCoverage::runOnModule(Module &M) {
       // in CTX mode we have to restore the original context for the caller -
       // she might be calling other functions which need the correct CTX.
       // Currently this is only needed for the Ubuntu clang-6.0 bug
-      if (ctx_str && has_calls) {
+      if (instrument_ctx && has_calls) {
 
         Instruction *Inst = BB.getTerminator();
         if (isa<ReturnInst>(Inst) || isa<ResumeInst>(Inst)) {
 
           IRBuilder<> Post_IRB(Inst);
-          StoreInst * RestoreCtx = Post_IRB.CreateStore(PrevCtx, AFLContext);
+
+          StoreInst *RestoreCtx;
+#ifdef AFL_HAVE_VECTOR_INTRINSICS
+          if (ctx_k)
+            RestoreCtx = IRB.CreateStore(PrevCaller, AFLPrevCaller);
+          else
+#endif
+            RestoreCtx = Post_IRB.CreateStore(PrevCtx, AFLContext);
           RestoreCtx->setMetadata(M.getMDKindID("nosanitize"),
                                   MDNode::get(C, None));
 

instrumentation/cmplog-instructions-pass.cc

@@ -19,12 +19,13 @@
 #include <stdlib.h>
 #include <unistd.h>
 
+#include <iostream>
 #include <list>
 #include <string>
 #include <fstream>
 #include <sys/time.h>
-#include "llvm/Config/llvm-config.h"
 
+#include "llvm/Config/llvm-config.h"
 #include "llvm/ADT/Statistic.h"
 #include "llvm/IR/IRBuilder.h"
 #include "llvm/IR/LegacyPassManager.h"
@@ -265,13 +266,20 @@ bool CmpLogInstructions::hookInstrs(Module &M) {
       unsigned int  max_size = Val->getType()->getIntegerBitWidth(), cast_size;
       unsigned char do_cast = 0;
 
-      if (!SI->getNumCases() || max_size < 16 || max_size % 8) {
+      if (!SI->getNumCases() || max_size < 16) {
 
         // if (!be_quiet) errs() << "skip trivial switch..\n";
         continue;
 
       }
 
+      if (max_size % 8) {
+
+        max_size = (((max_size / 8) + 1) * 8);
+        do_cast = 1;
+
+      }
+
       IRBuilder<> IRB(SI->getParent());
       IRB.SetInsertPoint(SI);
 
@@ -310,36 +318,8 @@ bool CmpLogInstructions::hookInstrs(Module &M) {
 
       if (do_cast) {
 
-        ConstantInt *cint = dyn_cast<ConstantInt>(Val);
-        if (cint) {
-
-          uint64_t val = cint->getZExtValue();
-          // fprintf(stderr, "ConstantInt: %lu\n", val);
-          switch (cast_size) {
-
-            case 8:
-              CompareTo = ConstantInt::get(Int8Ty, val);
-              break;
-            case 16:
-              CompareTo = ConstantInt::get(Int16Ty, val);
-              break;
-            case 32:
-              CompareTo = ConstantInt::get(Int32Ty, val);
-              break;
-            case 64:
-              CompareTo = ConstantInt::get(Int64Ty, val);
-              break;
-            case 128:
-              CompareTo = ConstantInt::get(Int128Ty, val);
-              break;
-
-          }
-
-        } else {
-
-          CompareTo = IRB.CreateBitCast(Val, IntegerType::get(C, cast_size));
-
-        }
+        CompareTo =
+            IRB.CreateIntCast(CompareTo, IntegerType::get(C, cast_size), false);
 
       }
 
@@ -361,27 +341,8 @@ bool CmpLogInstructions::hookInstrs(Module &M) {
 
           if (do_cast) {
 
-            uint64_t val = cint->getZExtValue();
-            // fprintf(stderr, "ConstantInt: %lu\n", val);
-            switch (cast_size) {
-
-              case 8:
-                new_param = ConstantInt::get(Int8Ty, val);
-                break;
-              case 16:
-                new_param = ConstantInt::get(Int16Ty, val);
-                break;
-              case 32:
-                new_param = ConstantInt::get(Int32Ty, val);
-                break;
-              case 64:
-                new_param = ConstantInt::get(Int64Ty, val);
-                break;
-              case 128:
-                new_param = ConstantInt::get(Int128Ty, val);
-                break;
-
-            }
+            new_param =
+                IRB.CreateIntCast(cint, IntegerType::get(C, cast_size), false);
 
           }
 
@@ -457,7 +418,7 @@ bool CmpLogInstructions::hookInstrs(Module &M) {
       IntegerType *        intTyOp0 = NULL;
       IntegerType *        intTyOp1 = NULL;
       unsigned             max_size = 0, cast_size = 0;
-      unsigned char        attr = 0, do_cast = 0;
+      unsigned char        attr = 0;
       std::vector<Value *> args;
 
       CmpInst *cmpInst = dyn_cast<CmpInst>(selectcmpInst);
@@ -523,7 +484,6 @@ bool CmpLogInstructions::hookInstrs(Module &M) {
           max_size = 128;
 
         attr += 8;
-        do_cast = 1;
 
       } else {
 
@@ -540,7 +500,9 @@ bool CmpLogInstructions::hookInstrs(Module &M) {
 
       }
 
-      if (!max_size || max_size % 8 || max_size < 16) { continue; }
+      if (!max_size || max_size < 16) { continue; }
+
+      if (max_size % 8) { max_size = (((max_size / 8) + 1) * 8); }
 
       if (max_size > 128) {
 
@@ -553,7 +515,6 @@ bool CmpLogInstructions::hookInstrs(Module &M) {
         }
 
         max_size = 128;
-        do_cast = 1;
 
       }
 
@@ -569,92 +530,30 @@ bool CmpLogInstructions::hookInstrs(Module &M) {
           break;
         default:
           cast_size = 128;
-          do_cast = 1;
 
       }
 
-      if (do_cast) {
+      // errs() << "[CMPLOG] cmp  " << *cmpInst << "(in function " <<
+      // cmpInst->getFunction()->getName() << ")\n";
 
-        // F*cking LLVM optimized out any kind of bitcasts of ConstantInt values
-        // creating illegal calls. WTF. So we have to work around this.
+      // first bitcast to integer type of the same bitsize as the original
+      // type (this is a nop, if already integer)
+      Value *op0_i = IRB.CreateBitCast(
+          op0, IntegerType::get(C, op0->getType()->getPrimitiveSizeInBits()));
+      // then create a int cast, which does zext, trunc or bitcast. In our case
+      // usually zext to the next larger supported type (this is a nop if
+      // already the right type)
+      Value *V0 =
+          IRB.CreateIntCast(op0_i, IntegerType::get(C, cast_size), false);
+      args.push_back(V0);
+      Value *op1_i = IRB.CreateBitCast(
+          op1, IntegerType::get(C, op1->getType()->getPrimitiveSizeInBits()));
+      Value *V1 =
+          IRB.CreateIntCast(op1_i, IntegerType::get(C, cast_size), false);
+      args.push_back(V1);
 
-        ConstantInt *cint = dyn_cast<ConstantInt>(op0);
-        if (cint) {
-
-          uint64_t val = cint->getZExtValue();
-          // fprintf(stderr, "ConstantInt: %lu\n", val);
-          ConstantInt *new_param = NULL;
-          switch (cast_size) {
-
-            case 8:
-              new_param = ConstantInt::get(Int8Ty, val);
-              break;
-            case 16:
-              new_param = ConstantInt::get(Int16Ty, val);
-              break;
-            case 32:
-              new_param = ConstantInt::get(Int32Ty, val);
-              break;
-            case 64:
-              new_param = ConstantInt::get(Int64Ty, val);
-              break;
-            case 128:
-              new_param = ConstantInt::get(Int128Ty, val);
-              break;
-
-          }
-
-          if (!new_param) { continue; }
-          args.push_back(new_param);
-
-        } else {
-
-          Value *V0 = IRB.CreateBitCast(op0, IntegerType::get(C, cast_size));
-          args.push_back(V0);
-
-        }
-
-        cint = dyn_cast<ConstantInt>(op1);
-        if (cint) {
-
-          uint64_t     val = cint->getZExtValue();
-          ConstantInt *new_param = NULL;
-          switch (cast_size) {
-
-            case 8:
-              new_param = ConstantInt::get(Int8Ty, val);
-              break;
-            case 16:
-              new_param = ConstantInt::get(Int16Ty, val);
-              break;
-            case 32:
-              new_param = ConstantInt::get(Int32Ty, val);
-              break;
-            case 64:
-              new_param = ConstantInt::get(Int64Ty, val);
-              break;
-            case 128:
-              new_param = ConstantInt::get(Int128Ty, val);
-              break;
-
-          }
-
-          if (!new_param) { continue; }
-          args.push_back(new_param);
-
-        } else {
-
-          Value *V1 = IRB.CreateBitCast(op1, IntegerType::get(C, cast_size));
-          args.push_back(V1);
-
-        }
-
-      } else {
-
-        args.push_back(op0);
-        args.push_back(op1);
-
-      }
+      // errs() << "[CMPLOG] casted parameters:\n0: " << *V0 << "\n1: " << *V1
+      // << "\n";
 
       ConstantInt *attribute = ConstantInt::get(Int8Ty, attr);
       args.push_back(attribute);
@@ -667,7 +566,7 @@ bool CmpLogInstructions::hookInstrs(Module &M) {
       }
 
       // fprintf(stderr, "_ExtInt(%u) castTo %u with attr %u didcast %u\n",
-      //         max_size, cast_size, attr, do_cast);
+      //         max_size, cast_size, attr);
 
       switch (cast_size) {
 

instrumentation/compare-transform-pass.so.cc

@@ -229,9 +229,9 @@ bool CompareTransform::transformCmps(Module &M, const bool processStrcmp,
                           dyn_cast<ConstantDataArray>(Var->getInitializer())) {
 
                     HasStr2 = true;
-                    Str2 = Array->getAsString();
+                    Str2 = Array->getRawDataValues();
                     valueMap[Str2P] = new std::string(Str2.str());
-                    fprintf(stderr, "glo2 %s\n", Str2.str().c_str());
+                    // fprintf(stderr, "glo2 %s\n", Str2.str().c_str());
 
                   }
 
@@ -254,7 +254,7 @@ bool CompareTransform::transformCmps(Module &M, const bool processStrcmp,
                             Var->getInitializer())) {
 
                       HasStr1 = true;
-                      Str1 = Array->getAsString();
+                      Str1 = Array->getRawDataValues();
                       valueMap[Str1P] = new std::string(Str1.str());
                       // fprintf(stderr, "glo1 %s\n", Str1.str().c_str());
 
@@ -316,7 +316,7 @@ bool CompareTransform::transformCmps(Module &M, const bool processStrcmp,
               uint64_t len = ilen->getZExtValue();
               // if len is zero this is a pointless call but allow real
               // implementation to worry about that
-              if (!len) continue;
+              if (len < 2) continue;
 
               if (isMemcmp) {
 
@@ -362,19 +362,22 @@ bool CompareTransform::transformCmps(Module &M, const bool processStrcmp,
     bool        HasStr1 = getConstantStringInfo(Str1P, Str1);
     bool        HasStr2 = getConstantStringInfo(Str2P, Str2);
     uint64_t    constStrLen, unrollLen, constSizedLen = 0;
-    bool        isMemcmp =
-        !callInst->getCalledFunction()->getName().compare(StringRef("memcmp"));
-    bool isSizedcmp = isMemcmp ||
-                      !callInst->getCalledFunction()->getName().compare(
-                          StringRef("strncmp")) ||
-                      !callInst->getCalledFunction()->getName().compare(
-                          StringRef("strncasecmp"));
+    bool        isMemcmp = false;
+    bool        isSizedcmp = false;
+    bool        isCaseInsensitive = false;
+    Function *  Callee = callInst->getCalledFunction();
+    if (Callee) {
+
+      isMemcmp = Callee->getName().compare("memcmp") == 0;
+      isSizedcmp = isMemcmp || Callee->getName().compare("strncmp") == 0 ||
+                   Callee->getName().compare("strncasecmp") == 0;
+      isCaseInsensitive = Callee->getName().compare("strcasecmp") == 0 ||
+                          Callee->getName().compare("strncasecmp") == 0;
+
+    }
+
     Value *sizedValue = isSizedcmp ? callInst->getArgOperand(2) : NULL;
     bool   isConstSized = sizedValue && isa<ConstantInt>(sizedValue);
-    bool isCaseInsensitive = !callInst->getCalledFunction()->getName().compare(
-                                 StringRef("strcasecmp")) ||
-                             !callInst->getCalledFunction()->getName().compare(
-                                 StringRef("strncasecmp"));
 
     if (!(HasStr1 || HasStr2)) {
 
@@ -417,15 +420,29 @@ bool CompareTransform::transformCmps(Module &M, const bool processStrcmp,
 
     }
 
+    if (TmpConstStr.length() < 2 ||
+        (TmpConstStr.length() == 2 && !TmpConstStr[1])) {
+
+      continue;
+
+    }
+
     // add null termination character implicit in c strings
-    TmpConstStr.append("\0", 1);
+    if (!isMemcmp && TmpConstStr[TmpConstStr.length() - 1]) {
+
+      TmpConstStr.append("\0", 1);
+
+    }
 
     // in the unusual case the const str has embedded null
     // characters, the string comparison functions should terminate
     // at the first null
-    if (!isMemcmp)
+    if (!isMemcmp) {
+
       TmpConstStr.assign(TmpConstStr, 0, TmpConstStr.find('\0') + 1);
 
+    }
+
     constStrLen = TmpConstStr.length();
     // prefer use of StringRef (in comparison to std::string a StringRef has
     // built-in runtime bounds checking, which makes debugging easier)
@@ -436,15 +453,6 @@ bool CompareTransform::transformCmps(Module &M, const bool processStrcmp,
     else
       unrollLen = constStrLen;
 
-    /*
-        if (!be_quiet)
-          errs() << callInst->getCalledFunction()->getName() << ": unroll len "
-                 << unrollLen
-                 << ((isSizedcmp && !isConstSized) ? ", variable n" : "") << ":
-       "
-                 << ConstStr << "\n";
-    */
-
     /* split before the call instruction */
     BasicBlock *bb = callInst->getParent();
     BasicBlock *end_bb = bb->splitBasicBlock(BasicBlock::iterator(callInst));

instrumentation/llvm-alternative-coverage.h

@@ -14,5 +14,8 @@ typedef u64 PREV_LOC_T;
 /* Maximum ngram size */
 #define NGRAM_SIZE_MAX 16U
 
+/* Maximum K for top-K context sensitivity */
+#define CTX_MAX_K 32U
+
 #endif
 

instrumentation/split-compares-pass.so.cc

@@ -149,8 +149,11 @@ bool SplitComparesTransform::simplifyFPCompares(Module &M) {
     auto op1 = FcmpInst->getOperand(1);
 
     /* find out what the new predicate is going to be */
-    auto               pred = dyn_cast<CmpInst>(FcmpInst)->getPredicate();
+    auto cmp_inst = dyn_cast<CmpInst>(FcmpInst);
+    if (!cmp_inst) { continue; }
+    auto               pred = cmp_inst->getPredicate();
     CmpInst::Predicate new_pred;
+
     switch (pred) {
 
       case CmpInst::FCMP_UGE:
@@ -276,8 +279,11 @@ bool SplitComparesTransform::simplifyCompares(Module &M) {
     auto op1 = IcmpInst->getOperand(1);
 
     /* find out what the new predicate is going to be */
-    auto               pred = dyn_cast<CmpInst>(IcmpInst)->getPredicate();
+    auto cmp_inst = dyn_cast<CmpInst>(IcmpInst);
+    if (!cmp_inst) { continue; }
+    auto               pred = cmp_inst->getPredicate();
     CmpInst::Predicate new_pred;
+
     switch (pred) {
 
       case CmpInst::ICMP_UGE:
@@ -412,8 +418,11 @@ bool SplitComparesTransform::simplifyIntSignedness(Module &M) {
     IntegerType *IntType = IntegerType::get(C, bitw);
 
     /* get the new predicate */
-    auto               pred = dyn_cast<CmpInst>(IcmpInst)->getPredicate();
+    auto cmp_inst = dyn_cast<CmpInst>(IcmpInst);
+    if (!cmp_inst) { continue; }
+    auto               pred = cmp_inst->getPredicate();
     CmpInst::Predicate new_pred;
+
     if (pred == CmpInst::ICMP_SGT) {
 
       new_pred = CmpInst::ICMP_UGT;
@@ -603,6 +612,10 @@ size_t SplitComparesTransform::splitFPCompares(Module &M) {
     if (op_size != op1->getType()->getPrimitiveSizeInBits()) { continue; }
 
     const unsigned int sizeInBits = op0->getType()->getPrimitiveSizeInBits();
+
+    // BUG FIXME TODO: u64 does not work for > 64 bit ... e.g. 80 and 128 bit
+    if (sizeInBits > 64) { continue; }
+
     const unsigned int precision = sizeInBits == 32    ? 24
                                    : sizeInBits == 64  ? 53
                                    : sizeInBits == 128 ? 113
@@ -610,8 +623,7 @@ size_t SplitComparesTransform::splitFPCompares(Module &M) {
                                    : sizeInBits == 80  ? 65
                                                        : sizeInBits - 8;
 
-    const unsigned shiftR_exponent = precision - 1;
-    // BUG FIXME TODO: u64 does not work for > 64 bit ... e.g. 80 and 128 bit
+    const unsigned           shiftR_exponent = precision - 1;
     const unsigned long long mask_fraction =
         (1ULL << (shiftR_exponent - 1)) | ((1ULL << (shiftR_exponent - 1)) - 1);
     const unsigned long long mask_exponent =
@@ -1113,7 +1125,9 @@ size_t SplitComparesTransform::splitIntCompares(Module &M, unsigned bitw) {
     auto op0 = IcmpInst->getOperand(0);
     auto op1 = IcmpInst->getOperand(1);
 
-    auto pred = dyn_cast<CmpInst>(IcmpInst)->getPredicate();
+    auto cmp_inst = dyn_cast<CmpInst>(IcmpInst);
+    if (!cmp_inst) { continue; }
+    auto pred = cmp_inst->getPredicate();
 
     BasicBlock *end_bb = bb->splitBasicBlock(BasicBlock::iterator(IcmpInst));
 

qemu_mode/QEMUAFL_VERSION

@@ -1 +1 @@
-6ab6bf28de
+d1ca56b84e

qemu_mode/README.md

@@ -17,7 +17,7 @@ The idea and much of the initial implementation comes from Andrew Griffiths.
 The actual implementation on current QEMU (shipped as qemuafl) is from
 Andrea Fioraldi. Special thanks to abiondo that re-enabled TCG chaining.
 
-## 2) How to use
+## 2) How to use qemu_mode
 
 The feature is implemented with a patched QEMU. The simplest way
 to build it is to run ./build_qemu_support.sh. The script will download,
@@ -99,6 +99,13 @@ Just set AFL_QEMU_INST_RANGES=A,B,C...
 The format of the items in the list is either a range of addresses like 0x123-0x321
 or a module name like module.so (that is matched in the mapped object filename).
 
+Alternatively you can tell QEMU to ignore part of an address space for instrumentation.
+
+Just set AFL_QEMU_EXCLUDE_RANGES=A,B,C...
+
+The format of the items on the list is the same as for AFL_QEMU_INST_RANGES, and excluding ranges
+takes priority over any included ranges or AFL_INST_LIBS.
+
 ## 7) CompareCoverage
 
 CompareCoverage is a sub-instrumentation with effects similar to laf-intel.
@@ -176,7 +183,12 @@ Comparative measurements of execution speed or instrumentation coverage will be
 fairly meaningless if the optimization levels or instrumentation scopes don't
 match.
 
-## 12) Gotchas, feedback, bugs
+## 12) Other features
+
+With `AFL_QEMU_FORCE_DFL` you force QEMU to ignore the registered signal
+handlers of the target.
+
+## 13) Gotchas, feedback, bugs
 
 If you need to fix up checksums or do other cleanup on mutated test cases, see
 utils/custom_mutators/ for a viable solution.
@@ -197,19 +209,12 @@ with -march=core2, can help.
 Beyond that, this is an early-stage mechanism, so fields reports are welcome.
 You can send them to <afl-users@googlegroups.com>.
 
-## 13) Alternatives: static rewriting
+## 14) Alternatives: static rewriting
 
 Statically rewriting binaries just once, instead of attempting to translate
 them at run time, can be a faster alternative. That said, static rewriting is
 fraught with peril, because it depends on being able to properly and fully model
 program control flow without actually executing each and every code path.
 
-The best implementation is this one:
-
-  https://github.com/vanhauser-thc/afl-dyninst
-
-The issue however is Dyninst which is not rewriting the binaries so that
-they run stable. A lot of crashes happen, especially in C++ programs that
-use throw/catch. Try it first, and if it works for you be happy as it is
-2-3x as fast as qemu_mode, however usually not as fast as QEMU persistent mode.
-
+Checkout the "Fuzzing binary-only targets" section in our main README.md and
+the docs/binaryonly_fuzzing.md document for more information and hints.

qemu_mode/build_qemu_support.sh

@@ -59,51 +59,11 @@ if [ ! -f "../afl-showmap" ]; then
 
 fi
 
-PREREQ_NOTFOUND=
-for i in git wget sha384sum bison flex iconv patch pkg-config; do
-
-  T=`command -v "$i" 2>/dev/null`
-
-  if [ "$T" = "" ]; then
-
-    echo "[-] Error: '$i' not found, please install first."
-    PREREQ_NOTFOUND=1
-
-  fi
-
-done
-
-PYTHONBIN=`command -v python3 || command -v python || command -v python2`
-
-if [ "$PYTHONBIN" = "" ]; then
-  echo "[-] Error: 'python' not found, please install using 'sudo apt install python3'."
-  PREREQ_NOTFOUND=1
-fi
-
-
-if [ ! -d "/usr/include/glib-2.0/" -a ! -d "/usr/local/include/glib-2.0/" ]; then
-
-  echo "[-] Error: devel version of 'glib2' not found, please install first."
-  PREREQ_NOTFOUND=1
-
-fi
-
-if [ ! -d "/usr/include/pixman-1/" -a ! -d "/usr/local/include/pixman-1/" ]; then
-
-  echo "[-] Error: devel version of 'pixman-1' not found, please install first."
-  PREREQ_NOTFOUND=1
-
-fi
-
 if echo "$CC" | grep -qF /afl-; then
 
   echo "[-] Error: do not use afl-gcc or afl-clang to compile this tool."
-  PREREQ_NOTFOUND=1
-
-fi
-
-if [ "$PREREQ_NOTFOUND" = "1" ]; then
   exit 1
+
 fi
 
 echo "[+] All checks passed!"
@@ -131,9 +91,13 @@ test -d qemuafl || { echo "[-] Not checked out, please install git or check your
 echo "[+] Got qemuafl."
 
 cd "qemuafl" || exit 1
-echo "[*] Checking out $QEMUAFL_VERSION"
-sh -c 'git stash && git stash drop' 1>/dev/null 2>/dev/null
-git checkout "$QEMUAFL_VERSION" || echo Warning: could not check out to commit $QEMUAFL_VERSION
+if [ -n "$NO_CHECKOUT" ]; then
+  echo "[*] Skipping checkout to $QEMUAFL_VERSION"
+else
+  echo "[*] Checking out $QEMUAFL_VERSION"
+  sh -c 'git stash' 1>/dev/null 2>/dev/null
+  git checkout "$QEMUAFL_VERSION" || echo Warning: could not check out to commit $QEMUAFL_VERSION
+fi
 
 echo "[*] Making sure imported headers matches"
 cp "../../include/config.h" "./qemuafl/imported/" || exit 1
@@ -233,15 +197,13 @@ QEMU_CONF_FLAGS=" \
   --disable-xen \
   --disable-xen-pci-passthrough \
   --disable-xfsctl \
-  --enable-pie \
-  --python=${PYTHONBIN} \
   --target-list="${CPU_TARGET}-linux-user" \
   --without-default-devices \
   "
 
 if [ -n "${CROSS_PREFIX}" ]; then
 
-  QEMU_CONF_FLAGS="${QEMU_CONF_FLAGS} --cross-prefix=${CROSS_PREFIX}"
+  QEMU_CONF_FLAGS="$QEMU_CONF_FLAGS --cross-prefix=$CROSS_PREFIX"
 
 fi
 
@@ -249,10 +211,15 @@ if [ "$STATIC" = "1" ]; then
 
   echo Building STATIC binary
 
-  QEMU_CONF_FLAGS="${QEMU_CONF_FLAGS} \
+  QEMU_CONF_FLAGS="$QEMU_CONF_FLAGS \
     --static \
     --extra-cflags=-DAFL_QEMU_STATIC_BUILD=1 \
     "
+
+else
+
+  QEMU_CONF_FLAGS="${QEMU_CONF_FLAGS} --enable-pie "
+
 fi
 
 if [ "$DEBUG" = "1" ]; then
@@ -262,7 +229,7 @@ if [ "$DEBUG" = "1" ]; then
   # --enable-gcov might go here but incurs a mesonbuild error on meson
   # versions prior to 0.56:
   # https://github.com/qemu/meson/commit/903d5dd8a7dc1d6f8bef79e66d6ebc07c
-  QEMU_CONF_FLAGS="${QEMU_CONF_FLAGS} \
+  QEMU_CONF_FLAGS="$QEMU_CONF_FLAGS \
     --disable-strip \
     --enable-debug \
     --enable-debug-info \
@@ -275,7 +242,7 @@ if [ "$DEBUG" = "1" ]; then
 
 else
 
-  QEMU_CONF_FLAGS="${QEMU_CONF_FLAGS} \
+  QEMU_CONF_FLAGS="$QEMU_CONF_FLAGS \
     --disable-debug-info \
     --disable-debug-mutex \
     --disable-debug-tcg \
@@ -290,7 +257,7 @@ if [ "$PROFILING" = "1" ]; then
 
   echo Building PROFILED binary
 
-  QEMU_CONF_FLAGS="${QEMU_CONF_FLAGS} \
+  QEMU_CONF_FLAGS="$QEMU_CONF_FLAGS \
     --enable-gprof \
     --enable-profiler \
     "
@@ -298,7 +265,7 @@ if [ "$PROFILING" = "1" ]; then
 fi
 
 # shellcheck disable=SC2086
-./configure ${QEMU_CONF_FLAGS} || exit 1
+./configure $QEMU_CONF_FLAGS || exit 1
 
 echo "[+] Configuration complete."
 
@@ -368,18 +335,42 @@ if [ "$ORIG_CROSS" = "" ]; then
   then # works on Arch Linux
     CROSS=$CPU_TARGET-pc-linux-gnu-gcc
   fi
+  if ! command -v "$CROSS" > /dev/null && [ "$CPU_TARGET" = "i386" ]
+  then
+    CROSS=i686-linux-gnu-gcc
+    if ! command -v "$CROSS" > /dev/null
+    then # works on Arch Linux
+      CROSS=i686-pc-linux-gnu-gcc
+    fi
+    if ! command -v "$CROSS" > /dev/null && [ "`uname -m`" = "x86_64" ]
+    then # set -m32
+      test "$CC" = "" && CC="gcc"
+      CROSS="$CC"
+      CROSS_FLAGS=-m32
+    fi
+  fi
 fi
 
-if ! command -v "$CROSS" > /dev/null
-then
+if ! command -v "$CROSS" > /dev/null ; then
+  if [ "$CPU_TARGET" = "$(uname -m)" ] ; then
+    echo "[+] Building afl++ qemu support libraries with CC=$CC"
+    echo "[+] Building libcompcov ..."
+    make -C libcompcov && echo "[+] libcompcov ready"
+    echo "[+] Building unsigaction ..."
+    make -C unsigaction && echo "[+] unsigaction ready"
+    echo "[+] Building libqasan ..."
+    make -C libqasan && echo "[+] unsigaction ready"
+  else
     echo "[!] Cross compiler $CROSS could not be found, cannot compile libcompcov libqasan and unsigaction"
+  fi
 else
+  echo "[+] Building afl++ qemu support libraries with CC=\"$CROSS $CROSS_FLAGS\""
   echo "[+] Building libcompcov ..."
-  make -C libcompcov CC=$CROSS && echo "[+] libcompcov ready"
+  make -C libcompcov CC="$CROSS $CROSS_FLAGS" && echo "[+] libcompcov ready"
   echo "[+] Building unsigaction ..."
-  make -C unsigaction CC=$CROSS && echo "[+] unsigaction ready"
+  make -C unsigaction CC="$CROSS $CROSS_FLAGS" && echo "[+] unsigaction ready"
   echo "[+] Building libqasan ..."
-  make -C libqasan CC=$CROSS && echo "[+] unsigaction ready"
+  make -C libqasan CC="$CROSS $CROSS_FLAGS" && echo "[+] unsigaction ready"
 fi
 
 echo "[+] All done for qemu_mode, enjoy!"

qemu_mode/libcompcov/libcompcov.so.c

@@ -29,6 +29,8 @@
 #include <sys/types.h>
 #include <sys/shm.h>
 #include <stdbool.h>
+#include <stdint.h>
+#include <inttypes.h>
 
 #include "types.h"
 #include "config.h"
@@ -159,14 +161,15 @@ static void __compcov_load(void) {
 
 }
 
-static void __compcov_trace(u64 cur_loc, const u8 *v0, const u8 *v1, size_t n) {
+static void __compcov_trace(uintptr_t cur_loc, const u8 *v0, const u8 *v1,
+                            size_t n) {
 
   size_t i;
 
   if (debug_fd != 1) {
 
     char debugbuf[4096];
-    snprintf(debugbuf, sizeof(debugbuf), "0x%llx %s %s %zu\n", cur_loc,
+    snprintf(debugbuf, sizeof(debugbuf), "0x%" PRIxPTR " %s %s %zu\n", cur_loc,
              v0 == NULL ? "(null)" : (char *)v0,
              v1 == NULL ? "(null)" : (char *)v1, n);
     write(debug_fd, debugbuf, strlen(debugbuf));
@@ -206,7 +209,7 @@ int strcmp(const char *str1, const char *str2) {
 
     if (n <= MAX_CMP_LENGTH) {
 
-      u64 cur_loc = (u64)retaddr;
+      uintptr_t cur_loc = (uintptr_t)retaddr;
       cur_loc = (cur_loc >> 4) ^ (cur_loc << 8);
       cur_loc &= MAP_SIZE - 1;
 
@@ -235,7 +238,7 @@ int strncmp(const char *str1, const char *str2, size_t len) {
 
     if (n <= MAX_CMP_LENGTH) {
 
-      u64 cur_loc = (u64)retaddr;
+      uintptr_t cur_loc = (uintptr_t)retaddr;
       cur_loc = (cur_loc >> 4) ^ (cur_loc << 8);
       cur_loc &= MAP_SIZE - 1;
 
@@ -265,7 +268,7 @@ int strcasecmp(const char *str1, const char *str2) {
 
     if (n <= MAX_CMP_LENGTH) {
 
-      u64 cur_loc = (u64)retaddr;
+      uintptr_t cur_loc = (uintptr_t)retaddr;
       cur_loc = (cur_loc >> 4) ^ (cur_loc << 8);
       cur_loc &= MAP_SIZE - 1;
 
@@ -296,7 +299,7 @@ int strncasecmp(const char *str1, const char *str2, size_t len) {
 
     if (n <= MAX_CMP_LENGTH) {
 
-      u64 cur_loc = (u64)retaddr;
+      uintptr_t cur_loc = (uintptr_t)retaddr;
       cur_loc = (cur_loc >> 4) ^ (cur_loc << 8);
       cur_loc &= MAP_SIZE - 1;
 
@@ -324,7 +327,7 @@ int memcmp(const void *mem1, const void *mem2, size_t len) {
 
     if (n <= MAX_CMP_LENGTH) {
 
-      u64 cur_loc = (u64)retaddr;
+      uintptr_t cur_loc = (uintptr_t)retaddr;
       cur_loc = (cur_loc >> 4) ^ (cur_loc << 8);
       cur_loc &= MAP_SIZE - 1;
 

qemu_mode/libqasan/README.md

@@ -4,16 +4,25 @@ This library is the injected runtime used by QEMU AddressSanitizer (QASan).
 
 The original repository is [here](https://github.com/andreafioraldi/qasan).
 
-The version embedded in qemuafl is an updated version of just the usermode part and this runtime in injected via LD_PRELOAD (so works just for dynamically linked binaries).
+The version embedded in qemuafl is an updated version of just the usermode part
+and this runtime is injected via LD_PRELOAD (so works just for dynamically
+linked binaries).
 
-The usage is super simple, just set the env var `AFL_USE_QASAN=1` when fuzzing in qemu mode (-Q). afl-fuzz will automatically set AFL_PRELOAD to load this library and enable the QASan instrumentation in afl-qemu-trace.
+The usage is super simple, just set the env var `AFL_USE_QASAN=1` when fuzzing
+in qemu mode (-Q). afl-fuzz will automatically set AFL_PRELOAD to load this
+library and enable the QASan instrumentation in afl-qemu-trace.
 
-For debugging purposes, we still suggest to run the original QASan as the stacktrace support for ARM (just a debug feature, it does not affect the bug finding capabilities during fuzzing) is WIP.
+For debugging purposes, we still suggest to run the original QASan as the
+stacktrace support for ARM (just a debug feature, it does not affect the bug
+finding capabilities during fuzzing) is WIP.
 
-### When I should use QASan?
+### When should I use QASan?
 
-If your target binary is PIC x86_64, you should before give a try to [retrowrite](https://github.com/HexHive/retrowrite) for static rewriting.
+If your target binary is PIC x86_64, you should also give a try to
+[retrowrite](https://github.com/HexHive/retrowrite) for static rewriting.
 
-If it fails, or if your binary is for another architecture, or you want to use persistent and snapshot mdoe, AFL++ QASan mode is what you want/have to use.
+If it fails, or if your binary is for another architecture, or you want to use
+persistent and snapshot mode, AFL++ QASan mode is what you want/have to use.
 
-Note that the overhead of libdislocator when combined with QEMU mode is much lower but it can catch less bugs. This is a short blanket, take your choice.
+Note that the overhead of libdislocator when combined with QEMU mode is much
+lower but it can catch less bugs. This is a short blanket, take your choice.

qemu_mode/libqasan/hooks.c

@@ -26,6 +26,8 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 #include "libqasan.h"
 #include "map_macro.h"
 
+ssize_t (*__lq_libc_write)(int, const void *, size_t);
+ssize_t (*__lq_libc_read)(int, void *, size_t);
 char *(*__lq_libc_fgets)(char *, int, FILE *);
 int (*__lq_libc_atoi)(const char *);
 long (*__lq_libc_atol)(const char *);
@@ -35,6 +37,8 @@ void __libqasan_init_hooks(void) {
 
   __libqasan_init_malloc();
 
+  __lq_libc_write = ASSERT_DLSYM(write);
+  __lq_libc_read = ASSERT_DLSYM(read);
   __lq_libc_fgets = ASSERT_DLSYM(fgets);
   __lq_libc_atoi = ASSERT_DLSYM(atoi);
   __lq_libc_atol = ASSERT_DLSYM(atol);
@@ -42,6 +46,30 @@ void __libqasan_init_hooks(void) {
 
 }
 
+ssize_t write(int fd, const void *buf, size_t count) {
+
+  void *rtv = __builtin_return_address(0);
+
+  QASAN_DEBUG("%14p: write(%d, %p, %zu)\n", rtv, fd, buf, count);
+  ssize_t r = __lq_libc_write(fd, buf, count);
+  QASAN_DEBUG("\t\t = %zd\n", r);
+
+  return r;
+
+}
+
+ssize_t read(int fd, void *buf, size_t count) {
+
+  void *rtv = __builtin_return_address(0);
+
+  QASAN_DEBUG("%14p: read(%d, %p, %zu)\n", rtv, fd, buf, count);
+  ssize_t r = __lq_libc_read(fd, buf, count);
+  QASAN_DEBUG("\t\t = %zd\n", r);
+
+  return r;
+
+}
+
 #ifdef __ANDROID__
 size_t malloc_usable_size(const void *ptr) {
 
@@ -54,7 +82,7 @@ size_t malloc_usable_size(void *ptr) {
 
   QASAN_DEBUG("%14p: malloc_usable_size(%p)\n", rtv, ptr);
   size_t r = __libqasan_malloc_usable_size((void *)ptr);
-  QASAN_DEBUG("\t\t = %ld\n", r);
+  QASAN_DEBUG("\t\t = %zu\n", r);
 
   return r;
 
@@ -64,7 +92,7 @@ void *malloc(size_t size) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: malloc(%ld)\n", rtv, size);
+  QASAN_DEBUG("%14p: malloc(%zu)\n", rtv, size);
   void *r = __libqasan_malloc(size);
   QASAN_DEBUG("\t\t = %p\n", r);
 
@@ -76,7 +104,7 @@ void *calloc(size_t nmemb, size_t size) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: calloc(%ld, %ld)\n", rtv, nmemb, size);
+  QASAN_DEBUG("%14p: calloc(%zu, %zu)\n", rtv, nmemb, size);
   void *r = __libqasan_calloc(nmemb, size);
   QASAN_DEBUG("\t\t = %p\n", r);
 
@@ -88,7 +116,7 @@ void *realloc(void *ptr, size_t size) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: realloc(%p, %ld)\n", rtv, ptr, size);
+  QASAN_DEBUG("%14p: realloc(%p, %zu)\n", rtv, ptr, size);
   void *r = __libqasan_realloc(ptr, size);
   QASAN_DEBUG("\t\t = %p\n", r);
 
@@ -100,7 +128,7 @@ int posix_memalign(void **memptr, size_t alignment, size_t size) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: posix_memalign(%p, %ld, %ld)\n", rtv, memptr, alignment,
+  QASAN_DEBUG("%14p: posix_memalign(%p, %zu, %zu)\n", rtv, memptr, alignment,
               size);
   int r = __libqasan_posix_memalign(memptr, alignment, size);
   QASAN_DEBUG("\t\t = %d [*memptr = %p]\n", r, *memptr);
@@ -113,7 +141,7 @@ void *memalign(size_t alignment, size_t size) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: memalign(%ld, %ld)\n", rtv, alignment, size);
+  QASAN_DEBUG("%14p: memalign(%zu, %zu)\n", rtv, alignment, size);
   void *r = __libqasan_memalign(alignment, size);
   QASAN_DEBUG("\t\t = %p\n", r);
 
@@ -125,7 +153,7 @@ void *aligned_alloc(size_t alignment, size_t size) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: aligned_alloc(%ld, %ld)\n", rtv, alignment, size);
+  QASAN_DEBUG("%14p: aligned_alloc(%zu, %zu)\n", rtv, alignment, size);
   void *r = __libqasan_aligned_alloc(alignment, size);
   QASAN_DEBUG("\t\t = %p\n", r);
 
@@ -137,7 +165,7 @@ void *valloc(size_t size) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: valloc(%ld)\n", rtv, size);
+  QASAN_DEBUG("%14p: valloc(%zu)\n", rtv, size);
   void *r = __libqasan_memalign(sysconf(_SC_PAGESIZE), size);
   QASAN_DEBUG("\t\t = %p\n", r);
 
@@ -149,7 +177,7 @@ void *pvalloc(size_t size) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: pvalloc(%ld)\n", rtv, size);
+  QASAN_DEBUG("%14p: pvalloc(%zu)\n", rtv, size);
   size_t page_size = sysconf(_SC_PAGESIZE);
   size = (size & (page_size - 1)) + page_size;
   void *r = __libqasan_memalign(page_size, size);
@@ -174,7 +202,9 @@ char *fgets(char *s, int size, FILE *stream) {
 
   QASAN_DEBUG("%14p: fgets(%p, %d, %p)\n", rtv, s, size, stream);
   QASAN_STORE(s, size);
+#ifndef __ANDROID__
   QASAN_LOAD(stream, sizeof(FILE));
+#endif
   char *r = __lq_libc_fgets(s, size, stream);
   QASAN_DEBUG("\t\t = %p\n", r);
 
@@ -186,7 +216,7 @@ int memcmp(const void *s1, const void *s2, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: memcmp(%p, %p, %ld)\n", rtv, s1, s2, n);
+  QASAN_DEBUG("%14p: memcmp(%p, %p, %zu)\n", rtv, s1, s2, n);
   QASAN_LOAD(s1, n);
   QASAN_LOAD(s2, n);
   int r = __libqasan_memcmp(s1, s2, n);
@@ -200,7 +230,7 @@ void *memcpy(void *dest, const void *src, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: memcpy(%p, %p, %ld)\n", rtv, dest, src, n);
+  QASAN_DEBUG("%14p: memcpy(%p, %p, %zu)\n", rtv, dest, src, n);
   QASAN_LOAD(src, n);
   QASAN_STORE(dest, n);
   void *r = __libqasan_memcpy(dest, src, n);
@@ -214,7 +244,7 @@ void *mempcpy(void *dest, const void *src, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: mempcpy(%p, %p, %ld)\n", rtv, dest, src, n);
+  QASAN_DEBUG("%14p: mempcpy(%p, %p, %zu)\n", rtv, dest, src, n);
   QASAN_LOAD(src, n);
   QASAN_STORE(dest, n);
   void *r = (uint8_t *)__libqasan_memcpy(dest, src, n) + n;
@@ -228,7 +258,7 @@ void *memmove(void *dest, const void *src, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: memmove(%p, %p, %ld)\n", rtv, dest, src, n);
+  QASAN_DEBUG("%14p: memmove(%p, %p, %zu)\n", rtv, dest, src, n);
   QASAN_LOAD(src, n);
   QASAN_STORE(dest, n);
   void *r = __libqasan_memmove(dest, src, n);
@@ -242,7 +272,7 @@ void *memset(void *s, int c, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: memset(%p, %d, %ld)\n", rtv, s, c, n);
+  QASAN_DEBUG("%14p: memset(%p, %d, %zu)\n", rtv, s, c, n);
   QASAN_STORE(s, n);
   void *r = __libqasan_memset(s, c, n);
   QASAN_DEBUG("\t\t = %p\n", r);
@@ -255,7 +285,7 @@ void *memchr(const void *s, int c, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: memchr(%p, %d, %ld)\n", rtv, s, c, n);
+  QASAN_DEBUG("%14p: memchr(%p, %d, %zu)\n", rtv, s, c, n);
   void *r = __libqasan_memchr(s, c, n);
   if (r == NULL)
     QASAN_LOAD(s, n);
@@ -271,7 +301,7 @@ void *memrchr(const void *s, int c, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: memrchr(%p, %d, %ld)\n", rtv, s, c, n);
+  QASAN_DEBUG("%14p: memrchr(%p, %d, %zu)\n", rtv, s, c, n);
   QASAN_LOAD(s, n);
   void *r = __libqasan_memrchr(s, c, n);
   QASAN_DEBUG("\t\t = %p\n", r);
@@ -285,7 +315,7 @@ void *memmem(const void *haystack, size_t haystacklen, const void *needle,
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: memmem(%p, %ld, %p, %ld)\n", rtv, haystack, haystacklen,
+  QASAN_DEBUG("%14p: memmem(%p, %zu, %p, %zu)\n", rtv, haystack, haystacklen,
               needle, needlelen);
   QASAN_LOAD(haystack, haystacklen);
   QASAN_LOAD(needle, needlelen);
@@ -301,7 +331,7 @@ void bzero(void *s, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: bzero(%p, %ld)\n", rtv, s, n);
+  QASAN_DEBUG("%14p: bzero(%p, %zu)\n", rtv, s, n);
   QASAN_STORE(s, n);
   __libqasan_memset(s, 0, n);
 
@@ -313,7 +343,7 @@ void explicit_bzero(void *s, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: bzero(%p, %ld)\n", rtv, s, n);
+  QASAN_DEBUG("%14p: bzero(%p, %zu)\n", rtv, s, n);
   QASAN_STORE(s, n);
   __libqasan_memset(s, 0, n);
 
@@ -323,7 +353,7 @@ int bcmp(const void *s1, const void *s2, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: bcmp(%p, %p, %ld)\n", rtv, s1, s2, n);
+  QASAN_DEBUG("%14p: bcmp(%p, %p, %zu)\n", rtv, s1, s2, n);
   QASAN_LOAD(s1, n);
   QASAN_LOAD(s2, n);
   int r = __libqasan_bcmp(s1, s2, n);
@@ -381,7 +411,7 @@ int strncasecmp(const char *s1, const char *s2, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: strncasecmp(%p, %p, %ld)\n", rtv, s1, s2, n);
+  QASAN_DEBUG("%14p: strncasecmp(%p, %p, %zu)\n", rtv, s1, s2, n);
   size_t l1 = __libqasan_strnlen(s1, n);
   QASAN_LOAD(s1, l1);
   size_t l2 = __libqasan_strnlen(s2, n);
@@ -431,7 +461,7 @@ int strncmp(const char *s1, const char *s2, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: strncmp(%p, %p, %ld)\n", rtv, s1, s2, n);
+  QASAN_DEBUG("%14p: strncmp(%p, %p, %zu)\n", rtv, s1, s2, n);
   size_t l1 = __libqasan_strnlen(s1, n);
   QASAN_LOAD(s1, l1);
   size_t l2 = __libqasan_strnlen(s2, n);
@@ -462,7 +492,7 @@ char *strncpy(char *dest, const char *src, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: strncpy(%p, %p, %ld)\n", rtv, dest, src, n);
+  QASAN_DEBUG("%14p: strncpy(%p, %p, %zu)\n", rtv, dest, src, n);
   size_t l = __libqasan_strnlen(src, n);
   QASAN_STORE(dest, n);
   void *r;
@@ -521,7 +551,7 @@ size_t strlen(const char *s) {
   QASAN_DEBUG("%14p: strlen(%p)\n", rtv, s);
   size_t r = __libqasan_strlen(s);
   QASAN_LOAD(s, r + 1);
-  QASAN_DEBUG("\t\t = %ld\n", r);
+  QASAN_DEBUG("\t\t = %zu\n", r);
 
   return r;
 
@@ -531,10 +561,10 @@ size_t strnlen(const char *s, size_t n) {
 
   void *rtv = __builtin_return_address(0);
 
-  QASAN_DEBUG("%14p: strnlen(%p, %ld)\n", rtv, s, n);
+  QASAN_DEBUG("%14p: strnlen(%p, %zu)\n", rtv, s, n);
   size_t r = __libqasan_strnlen(s, n);
   QASAN_LOAD(s, r);
-  QASAN_DEBUG("\t\t = %ld\n", r);
+  QASAN_DEBUG("\t\t = %zu\n", r);
 
   return r;
 
@@ -621,7 +651,7 @@ size_t wcslen(const wchar_t *s) {
   QASAN_DEBUG("%14p: wcslen(%p)\n", rtv, s);
   size_t r = __libqasan_wcslen(s);
   QASAN_LOAD(s, sizeof(wchar_t) * (r + 1));
-  QASAN_DEBUG("\t\t = %ld\n", r);
+  QASAN_DEBUG("\t\t = %zu\n", r);
 
   return r;
 

qemu_mode/libqasan/libqasan.c

@@ -72,7 +72,7 @@ void __libqasan_print_maps(void) {
 
   QASAN_LOG("QEMU-AddressSanitizer (v%s)\n", QASAN_VERSTR);
   QASAN_LOG(
-      "Copyright (C) 2019-2020 Andrea Fioraldi <andreafioraldi@gmail.com>\n");
+      "Copyright (C) 2019-2021 Andrea Fioraldi <andreafioraldi@gmail.com>\n");
   QASAN_LOG("\n");
 
   if (__qasan_log) __libqasan_print_maps();

qemu_mode/libqasan/malloc.c

@@ -24,6 +24,7 @@ SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 *******************************************************************************/
 
 #include "libqasan.h"
+#include <features.h>
 #include <errno.h>
 #include <stddef.h>
 #include <assert.h>
@@ -65,9 +66,26 @@ struct chunk_struct {
 
 };
 
+#ifdef __GLIBC__
+
+void *(*__lq_libc_malloc)(size_t);
+void (*__lq_libc_free)(void *);
+  #define backend_malloc __lq_libc_malloc
+  #define backend_free __lq_libc_free
+
+  #define TMP_ZONE_SIZE 4096
+static int           __tmp_alloc_zone_idx;
+static unsigned char __tmp_alloc_zone[TMP_ZONE_SIZE];
+
+#else
+
 // From dlmalloc.c
-void *dlmalloc(size_t);
-void  dlfree(void *);
+void *                    dlmalloc(size_t);
+void                      dlfree(void *);
+  #define backend_malloc dlmalloc
+  #define backend_free dlfree
+
+#endif
 
 int __libqasan_malloc_initialized;
 
@@ -102,9 +120,9 @@ static int quanratine_push(struct chunk_begin *ck) {
     quarantine_bytes -= tmp->requested_size;
 
     if (tmp->aligned_orig)
-      dlfree(tmp->aligned_orig);
+      backend_free(tmp->aligned_orig);
     else
-      dlfree(tmp);
+      backend_free(tmp);
 
   }
 
@@ -122,6 +140,11 @@ void __libqasan_init_malloc(void) {
 
   if (__libqasan_malloc_initialized) return;
 
+#ifdef __GLIBC__
+  __lq_libc_malloc = dlsym(RTLD_NEXT, "malloc");
+  __lq_libc_free = dlsym(RTLD_NEXT, "free");
+#endif
+
   LOCK_INIT(&quarantine_lock, PTHREAD_PROCESS_PRIVATE);
 
   __libqasan_malloc_initialized = 1;
@@ -142,13 +165,27 @@ size_t __libqasan_malloc_usable_size(void *ptr) {
 
 void *__libqasan_malloc(size_t size) {
 
-  if (!__libqasan_malloc_initialized) { __libqasan_init_malloc(); }
+  if (!__libqasan_malloc_initialized) {
 
-  if (!__libqasan_malloc_initialized) __libqasan_init_malloc();
+    __libqasan_init_malloc();
+
+#ifdef __GLIBC__
+    void *r = &__tmp_alloc_zone[__tmp_alloc_zone_idx];
+
+    if (size & (ALLOC_ALIGN_SIZE - 1))
+      __tmp_alloc_zone_idx +=
+          (size & ~(ALLOC_ALIGN_SIZE - 1)) + ALLOC_ALIGN_SIZE;
+    else
+      __tmp_alloc_zone_idx += size;
+
+    return r;
+#endif
+
+  }
 
   int state = QASAN_SWAP(QASAN_DISABLED);  // disable qasan for this thread
 
-  struct chunk_begin *p = dlmalloc(sizeof(struct chunk_struct) + size);
+  struct chunk_begin *p = backend_malloc(sizeof(struct chunk_struct) + size);
 
   QASAN_SWAP(state);
 
@@ -179,6 +216,12 @@ void __libqasan_free(void *ptr) {
 
   if (!ptr) return;
 
+#ifdef __GLIBC__
+  if (ptr >= (void *)__tmp_alloc_zone &&
+      ptr < ((void *)__tmp_alloc_zone + TMP_ZONE_SIZE))
+    return;
+#endif
+
   struct chunk_begin *p = ptr;
   p -= 1;
 
@@ -190,9 +233,9 @@ void __libqasan_free(void *ptr) {
   if (!quanratine_push(p)) {
 
     if (p->aligned_orig)
-      dlfree(p->aligned_orig);
+      backend_free(p->aligned_orig);
     else
-      dlfree(p);
+      backend_free(p);
 
   }
 
@@ -210,6 +253,17 @@ void *__libqasan_calloc(size_t nmemb, size_t size) {
 
   size *= nmemb;
 
+#ifdef __GLIBC__
+  if (!__libqasan_malloc_initialized) {
+
+    void *r = &__tmp_alloc_zone[__tmp_alloc_zone_idx];
+    __tmp_alloc_zone_idx += size;
+    return r;
+
+  }
+
+#endif
+
   char *p = __libqasan_malloc(size);
   if (!p) return NULL;
 
@@ -252,7 +306,7 @@ int __libqasan_posix_memalign(void **ptr, size_t align, size_t len) {
 
   int state = QASAN_SWAP(QASAN_DISABLED);  // disable qasan for this thread
 
-  char *orig = dlmalloc(sizeof(struct chunk_struct) + size);
+  char *orig = backend_malloc(sizeof(struct chunk_struct) + size);
 
   QASAN_SWAP(state);
 

qemu_mode/libqasan/uninstrument.c

@@ -1,7 +1,7 @@
 /*
 
 This code is DEPRECATED!
-I'm keeping it here cause maybe the unistrumentation of a function is needed
+I'm keeping it here cause maybe the uninstrumentation of a function is needed
 for some strange reason.
 
 */

src/afl-analyze.c

@@ -785,6 +785,7 @@ static void set_up_environment(void) {
          "abort_on_error=1:"
          "detect_leaks=0:"
          "allocator_may_return_null=1:"
+         "detect_odr_violation=0:"
          "symbolize=0:"
          "handle_segv=0:"
          "handle_sigbus=0:"

src/afl-cc.c

@@ -22,7 +22,7 @@
 #include "types.h"
 #include "debug.h"
 #include "alloc-inl.h"
-#include "llvm-ngram-coverage.h"
+#include "llvm-alternative-coverage.h"
 
 #include <stdio.h>
 #include <unistd.h>
@@ -31,8 +31,6 @@
 #include <strings.h>
 #include <limits.h>
 #include <assert.h>
-#include <sys/types.h>
-#include <sys/wait.h>
 
 #if (LLVM_MAJOR - 0 == 0)
   #undef LLVM_MAJOR
@@ -52,14 +50,13 @@ static u8 **cc_params;                 /* Parameters passed to the real CC  */
 static u32  cc_par_cnt = 1;            /* Param count, including argv0      */
 static u8   clang_mode;                /* Invoked as afl-clang*?            */
 static u8   llvm_fullpath[PATH_MAX];
-static u8   instrument_mode, instrument_opt_mode, ngram_size, lto_mode;
+static u8   instrument_mode, instrument_opt_mode, ngram_size, ctx_k, lto_mode;
 static u8   compiler_mode, plusplus_mode, have_instr_env = 0;
 static u8   have_gcc, have_llvm, have_gcc_plugin, have_lto, have_instr_list = 0;
 static u8 * lto_flag = AFL_CLANG_FLTO, *argvnull;
 static u8   debug;
 static u8   cwd[4096];
 static u8   cmplog_mode;
-static u8 * ll_file;
 u8          use_stdin;                                             /* dummy */
 // static u8 *march_opt = CFLAGS_OPT;
 
@@ -76,7 +73,9 @@ enum {
   INSTRUMENT_GCC = 6,
   INSTRUMENT_CLANG = 7,
   INSTRUMENT_OPT_CTX = 8,
-  INSTRUMENT_OPT_NGRAM = 16
+  INSTRUMENT_OPT_NGRAM = 16,
+  INSTRUMENT_OPT_CALLER = 32,
+  INSTRUMENT_OPT_CTX_K = 64,
 
 };
 
@@ -91,7 +90,7 @@ char instrument_mode_string[18][18] = {
     "GCC",
     "CLANG",
     "CTX",
-    "",
+    "CALLER",
     "",
     "",
     "",
@@ -318,7 +317,6 @@ static void edit_params(u32 argc, char **argv, char **envp) {
   u8 fortify_set = 0, asan_set = 0, x_set = 0, bit_mode = 0, shared_linking = 0,
      preprocessor_only = 0, have_unroll = 0, have_o = 0, have_pic = 0,
      have_c = 0;
-  u32 out_idx = 0;
 
   cc_params = ck_alloc((argc + 128) * sizeof(u8 *));
 
@@ -558,6 +556,11 @@ static void edit_params(u32 argc, char **argv, char **envp) {
 
     }
 
+#if LLVM_MAJOR >= 13
+    // fuck you llvm 13
+    cc_params[cc_par_cnt++] = "-fno-experimental-new-pass-manager";
+#endif
+
     if (lto_mode && !have_c) {
 
       u8 *ld_path = strdup(AFL_REAL_LD);
@@ -745,8 +748,6 @@ static void edit_params(u32 argc, char **argv, char **envp) {
 
     cc_params[cc_par_cnt++] = cur;
 
-    if (!strcmp(cur, "-o")) { out_idx = cc_par_cnt; }
-
   }
 
   if (getenv("AFL_HARDEN")) {
@@ -807,7 +808,7 @@ static void edit_params(u32 argc, char **argv, char **envp) {
 
   }
 
-  if (!getenv("AFL_DONT_OPTIMIZE") && !getenv("AFL_LLVM_LTO_CALLGRAPH")) {
+  if (!getenv("AFL_DONT_OPTIMIZE")) {
 
     cc_params[cc_par_cnt++] = "-g";
     if (!have_o) cc_params[cc_par_cnt++] = "-O3";
@@ -939,7 +940,10 @@ static void edit_params(u32 argc, char **argv, char **envp) {
 
   }
 
-  if (preprocessor_only) {
+  // prevent unnecessary build errors
+  cc_params[cc_par_cnt++] = "-Wno-unused-command-line-argument";
+
+  if (preprocessor_only || have_c) {
 
     /* In the preprocessor_only case (-E), we are not actually compiling at
        all but requesting the compiler to output preprocessed sources only.
@@ -953,7 +957,7 @@ static void edit_params(u32 argc, char **argv, char **envp) {
 
 #ifndef __ANDROID__
 
-  if (compiler_mode != GCC && compiler_mode != CLANG && !getenv("AFL_LLVM_LTO_CALLGRAPH")) {
+  if (compiler_mode != GCC && compiler_mode != CLANG) {
 
     switch (bit_mode) {
 
@@ -1000,39 +1004,19 @@ static void edit_params(u32 argc, char **argv, char **envp) {
     }
 
   #if !defined(__APPLE__) && !defined(__sun)
-    if (!shared_linking && !have_c)
+    if (!shared_linking)
       cc_params[cc_par_cnt++] =
           alloc_printf("-Wl,--dynamic-list=%s/dynamic_list.txt", obj_path);
   #endif
 
   #if defined(USEMMAP) && !defined(__HAIKU__)
-    if (!have_c) cc_params[cc_par_cnt++] = "-lrt";
+    cc_params[cc_par_cnt++] = "-lrt";
   #endif
 
-    // prevent unnecessary build errors
-    cc_params[cc_par_cnt++] = "-Wno-unused-command-line-argument";
-
   }
 
 #endif
 
-  if (getenv("AFL_LLVM_LTO_CALLGRAPH")) {
-
-    if (!lto_mode) { FATAL("AFL_LLVM_LTO_CALLGRAPH requires LTO mode"); }
-    cc_params[cc_par_cnt++] = "-O0";
-    cc_params[cc_par_cnt++] = "-w";
-  
-    if (!have_c) {
-
-      ll_file = alloc_printf("%s.ll", cc_params[out_idx]);
-      cc_params[out_idx] = ll_file;
-      cc_params[cc_par_cnt++] = "-S";
-      cc_params[cc_par_cnt++] = "-emit-llvm";
-
-    }
-
-  }
-
   cc_params[cc_par_cnt] = NULL;
 
 }
@@ -1041,7 +1025,7 @@ static void edit_params(u32 argc, char **argv, char **envp) {
 
 int main(int argc, char **argv, char **envp) {
 
-  int   i;
+  int   i, passthrough = 0;
   char *callname = argv[0], *ptr = NULL;
 
   if (getenv("AFL_DEBUG")) {
@@ -1061,6 +1045,13 @@ int main(int argc, char **argv, char **envp) {
 
   }
 
+  if (getenv("AFL_PASSTHROUGH") || getenv("AFL_NOOPT")) {
+
+    passthrough = 1;
+    if (!debug) { be_quiet = 1; }
+
+  }
+
   if ((ptr = strrchr(callname, '/')) != NULL) callname = ptr + 1;
   argvnull = (u8 *)argv[0];
   check_environment_vars(envp);
@@ -1291,6 +1282,7 @@ int main(int argc, char **argv, char **envp) {
   }
 
   if (getenv("AFL_LLVM_CTX")) instrument_opt_mode |= INSTRUMENT_OPT_CTX;
+  if (getenv("AFL_LLVM_CALLER")) instrument_opt_mode |= INSTRUMENT_OPT_CALLER;
 
   if (getenv("AFL_LLVM_NGRAM_SIZE")) {
 
@@ -1304,6 +1296,26 @@ int main(int argc, char **argv, char **envp) {
 
   }
 
+  if (getenv("AFL_LLVM_CTX_K")) {
+
+    ctx_k = atoi(getenv("AFL_LLVM_CTX_K"));
+    if (ctx_k < 1 || ctx_k > CTX_MAX_K)
+      FATAL("K-CTX instrumentation mode must be between 1 and CTX_MAX_K (%u)",
+            CTX_MAX_K);
+    if (ctx_k == 1) {
+
+      setenv("AFL_LLVM_CALLER", "1", 1);
+      unsetenv("AFL_LLVM_CTX_K");
+      instrument_opt_mode |= INSTRUMENT_OPT_CALLER;
+
+    } else {
+
+      instrument_opt_mode |= INSTRUMENT_OPT_CTX_K;
+
+    }
+
+  }
+
   if (getenv("AFL_LLVM_INSTRUMENT")) {
 
     u8 *ptr2 = strtok(getenv("AFL_LLVM_INSTRUMENT"), ":,;");
@@ -1399,6 +1411,44 @@ int main(int argc, char **argv, char **envp) {
 
       }
 
+      if (strncasecmp(ptr2, "ctx-", strlen("ctx-")) == 0) {
+
+        u8 *ptr3 = ptr2 + strlen("ctx-");
+        while (*ptr3 && (*ptr3 < '0' || *ptr3 > '9'))
+          ptr3++;
+
+        if (!*ptr3) {
+
+          if ((ptr3 = getenv("AFL_LLVM_CTX_K")) == NULL)
+            FATAL(
+                "you must set the K-CTX K with (e.g. for value 2) "
+                "AFL_LLVM_INSTRUMENT=ctx-2");
+
+        }
+
+        ctx_k = atoi(ptr3);
+        if (ctx_k < 1 || ctx_k > CTX_MAX_K)
+          FATAL(
+              "K-CTX instrumentation option must be between 1 and CTX_MAX_K "
+              "(%u)",
+              CTX_MAX_K);
+
+        if (ctx_k == 1) {
+
+          instrument_opt_mode |= INSTRUMENT_OPT_CALLER;
+          setenv("AFL_LLVM_CALLER", "1", 1);
+          unsetenv("AFL_LLVM_CTX_K");
+
+        } else {
+
+          instrument_opt_mode |= (INSTRUMENT_OPT_CTX_K);
+          u8 *ptr4 = alloc_printf("%u", ctx_k);
+          setenv("AFL_LLVM_CTX_K", ptr4, 1);
+
+        }
+
+      }
+
       if (strncasecmp(ptr2, "ctx", strlen("ctx")) == 0) {
 
         instrument_opt_mode |= INSTRUMENT_OPT_CTX;
@@ -1406,6 +1456,13 @@ int main(int argc, char **argv, char **envp) {
 
       }
 
+      if (strncasecmp(ptr2, "caller", strlen("caller")) == 0) {
+
+        instrument_opt_mode |= INSTRUMENT_OPT_CALLER;
+        setenv("AFL_LLVM_CALLER", "1", 1);
+
+      }
+
       if (strncasecmp(ptr2, "ngram", strlen("ngram")) == 0) {
 
         u8 *ptr3 = ptr2 + strlen("ngram");
@@ -1439,6 +1496,35 @@ int main(int argc, char **argv, char **envp) {
 
   }
 
+  if ((instrument_opt_mode & INSTRUMENT_OPT_CTX) &&
+      (instrument_opt_mode & INSTRUMENT_OPT_CALLER)) {
+
+    FATAL("you cannot set CTX and CALLER together");
+
+  }
+
+  if ((instrument_opt_mode & INSTRUMENT_OPT_CTX) &&
+      (instrument_opt_mode & INSTRUMENT_OPT_CTX_K)) {
+
+    FATAL("you cannot set CTX and K-CTX together");
+
+  }
+
+  if ((instrument_opt_mode & INSTRUMENT_OPT_CALLER) &&
+      (instrument_opt_mode & INSTRUMENT_OPT_CTX_K)) {
+
+    FATAL("you cannot set CALLER and K-CTX together");
+
+  }
+
+  if (instrument_opt_mode && instrument_mode == INSTRUMENT_DEFAULT &&
+      (compiler_mode == LLVM || compiler_mode == UNSET)) {
+
+    instrument_mode = INSTRUMENT_CLASSIC;
+    compiler_mode = LLVM;
+
+  }
+
   if (!compiler_mode) {
 
     // lto is not a default because outside of afl-cc RANLIB and AR have to
@@ -1508,12 +1594,13 @@ int main(int argc, char **argv, char **envp) {
         "      CLASSIC              %s      no  yes     module yes yes    "
         "yes\n"
         "        - NORMAL\n"
+        "        - CALLER\n"
         "        - CTX\n"
         "        - NGRAM-{2-16}\n"
         "      INSTRIM                           no  yes     module yes yes "
         "   yes\n"
         "        - NORMAL\n"
-        "        - CTX\n"
+        "        - CALLER\n"
         "        - NGRAM-{2-16}\n"
         "  [GCC_PLUGIN] gcc plugin: %s%s\n"
         "      CLASSIC              DEFAULT      no  yes     no     no  no     "
@@ -1560,7 +1647,10 @@ int main(int argc, char **argv, char **envp) {
         NATIVE_MSG
 
         "  CLASSIC: decision target instrumentation (README.llvm.md)\n"
-        "  CTX:     CLASSIC + callee context (instrumentation/README.ctx.md)\n"
+        "  CALLER:  CLASSIC + single callee context "
+        "(instrumentation/README.ctx.md)\n"
+        "  CTX:     CLASSIC + full callee context "
+        "(instrumentation/README.ctx.md)\n"
         "  NGRAM-x: CLASSIC + previous path "
         "((instrumentation/README.ngram.md)\n"
         "  INSTRIM: Dominator tree (for LLVM <= 6.0) "
@@ -1603,8 +1693,11 @@ int main(int argc, char **argv, char **envp) {
           "  AFL_DONT_OPTIMIZE: disable optimization instead of -O3\n"
           "  AFL_NO_BUILTIN: no builtins for string compare functions (for "
           "libtokencap.so)\n"
+          "  AFL_NOOP: behave like a normal compiler (to pass configure "
+          "tests)\n"
           "  AFL_PATH: path to instrumenting pass and runtime  "
           "(afl-compiler-rt.*o)\n"
+          "  AFL_IGNORE_UNKNOWN_ENVS: don't warn on unknown env vars\n"
           "  AFL_INST_RATIO: percentage of branches to instrument\n"
           "  AFL_QUIET: suppress verbose output\n"
           "  AFL_HARDEN: adds code hardening to catch memory bugs\n"
@@ -1653,15 +1746,17 @@ int main(int argc, char **argv, char **envp) {
             "  AFL_LLVM_CMPLOG: log operands of comparisons (RedQueen "
             "mutator)\n"
             "  AFL_LLVM_INSTRUMENT: set instrumentation mode:\n"
-            "    CLASSIC, INSTRIM, PCGUARD, LTO, GCC, CLANG, CTX, NGRAM-2 ... "
-            "NGRAM-16\n"
+            "    CLASSIC, INSTRIM, PCGUARD, LTO, GCC, CLANG, CALLER, CTX, "
+            "NGRAM-2 ..-16\n"
             " You can also use the old environment variables instead:\n"
             "  AFL_LLVM_USE_TRACE_PC: use LLVM trace-pc-guard instrumentation\n"
             "  AFL_LLVM_INSTRIM: use light weight instrumentation InsTrim\n"
             "  AFL_LLVM_INSTRIM_LOOPHEAD: optimize loop tracing for speed "
             "(option to INSTRIM)\n"
-            "  AFL_LLVM_CTX: use context sensitive coverage (for CLASSIC and "
-            "INSTRIM)\n"
+            "  AFL_LLVM_CALLER: use single context sensitive coverage (for "
+            "CLASSIC)\n"
+            "  AFL_LLVM_CTX: use full context sensitive coverage (for "
+            "CLASSIC)\n"
             "  AFL_LLVM_NGRAM_SIZE: use ngram prev_loc count coverage (for "
             "CLASSIC & INSTRIM)\n");
 
@@ -1678,9 +1773,8 @@ int main(int argc, char **argv, char **envp) {
             "  AFL_LLVM_LTO_DONTWRITEID: don't write the highest ID used to a "
             "global var\n"
             "  AFL_LLVM_LTO_STARTID: from which ID to start counting from for "
-            "a bb\n"
-            "  AFL_LLVM_LTO_CALLGRAPH: will not generate a binary but a "
-            "callgraph instead\n"
+            "a "
+            "bb\n"
             "  AFL_REAL_LD: use this lld linker instead of the compiled in "
             "path\n"
             "If anything fails - be sure to read README.lto.md!\n");
@@ -1717,7 +1811,10 @@ int main(int argc, char **argv, char **envp) {
         "Do not be overwhelmed :) afl-cc uses good defaults if no options are "
         "selected.\n"
         "Read the documentation for FEATURES though, all are good but few are "
-        "defaults.\n\n");
+        "defaults.\n"
+        "Recommended is afl-clang-lto with AFL_LLVM_CMPLOG or afl-clang-fast "
+        "with\n"
+        "AFL_LLVM_CMPLOG and AFL_LLVM_DICT2FILE.\n\n");
 
     exit(1);
 
@@ -1777,7 +1874,7 @@ int main(int argc, char **argv, char **envp) {
   }
 
   if (instrument_opt_mode && compiler_mode != LLVM)
-    FATAL("CTX and NGRAM can only be used in LLVM mode");
+    FATAL("CTX, CALLER and NGRAM can only be used in LLVM mode");
 
   if (!instrument_opt_mode) {
 
@@ -1787,15 +1884,18 @@ int main(int argc, char **argv, char **envp) {
 
   } else {
 
-    if (instrument_opt_mode == INSTRUMENT_OPT_CTX)
+    char *ptr2 = alloc_printf(" + NGRAM-%u", ngram_size);
+    char *ptr3 = alloc_printf(" + K-CTX-%u", ctx_k);
 
-      ptr = alloc_printf("%s + CTX", instrument_mode_string[instrument_mode]);
-    else if (instrument_opt_mode == INSTRUMENT_OPT_NGRAM)
-      ptr = alloc_printf("%s + NGRAM-%u",
-                         instrument_mode_string[instrument_mode], ngram_size);
-    else
-      ptr = alloc_printf("%s + CTX + NGRAM-%u",
-                         instrument_mode_string[instrument_mode], ngram_size);
+    ptr = alloc_printf(
+        "%s%s%s%s%s", instrument_mode_string[instrument_mode],
+        (instrument_opt_mode & INSTRUMENT_OPT_CTX) ? " + CTX" : "",
+        (instrument_opt_mode & INSTRUMENT_OPT_CALLER) ? " + CALLER" : "",
+        (instrument_opt_mode & INSTRUMENT_OPT_NGRAM) ? ptr2 : "",
+        (instrument_opt_mode & INSTRUMENT_OPT_CTX_K) ? ptr3 : "");
+
+    ck_free(ptr2);
+    ck_free(ptr3);
 
   }
 
@@ -1806,11 +1906,14 @@ int main(int argc, char **argv, char **envp) {
         "(requires LLVM 11 or higher)");
 #endif
 
-  if (instrument_opt_mode && instrument_mode != INSTRUMENT_CLASSIC &&
-      instrument_mode != INSTRUMENT_CFG)
+  if (instrument_opt_mode && instrument_mode == INSTRUMENT_CFG &&
+      instrument_opt_mode & INSTRUMENT_OPT_CTX)
+    FATAL("CFG instrumentation mode supports NGRAM and CALLER, but not CTX.");
+  else if (instrument_opt_mode && instrument_mode != INSTRUMENT_CLASSIC)
+    // we will drop CFG/INSTRIM in the future so do not advertise
     FATAL(
-        "CTX and NGRAM instrumentation options can only be used with CFG "
-        "(recommended) and CLASSIC instrumentation modes!");
+        "CALLER, CTX and NGRAM instrumentation options can only be used with "
+        "the LLVM CLASSIC instrumentation mode.");
 
   if (getenv("AFL_LLVM_SKIP_NEVERZERO") && getenv("AFL_LLVM_NOT_ZERO"))
     FATAL(
@@ -1904,79 +2007,18 @@ int main(int argc, char **argv, char **envp) {
 
   }
 
-  s32 pid = 0;
-  if (ll_file) { pid = fork(); }  // CG run
+  if (passthrough) {
 
-  if (!pid) {
-
-    execvp(cc_params[0], (char **)cc_params);
-    FATAL("Oops, failed to execute '%s' - check your PATH", cc_params[0]);
-
-  }
-
-  // only for AFL_LLVM_LTO_CALLGRAPH
-  s32 status;
-  if (waitpid(pid, &status, 0) <= 0) { PFATAL("waitpid() failed"); }
-
-  u8 *cg_file = (u8 *)ck_strdup(ll_file);
-  cg_file[strlen(cg_file) - 2] = 0;
-  strcat(cg_file, "cg");
-  if (!be_quiet) SAYF("Generating callgraph into %s ...\n", cg_file);
-  u8 *opt_cmd = (u8 *)ck_alloc(strlen(cc_params[0]) + 16);
-  strcpy(opt_cmd, cc_params[0]);
-  if ((ptr = (u8 *)strrchr(opt_cmd, '/'))) {
-
-    ++ptr;
+    argv[0] = cc_params[0];
+    execvp(cc_params[0], (char **)argv);
 
   } else {
 
-    ptr = opt_cmd;
+    execvp(cc_params[0], (char **)cc_params);
 
   }
 
-  u8 *dash = (u8 *)strchr(opt_cmd, '-');
-  strcpy((char *)opt_cmd, "opt");
-  if (dash) {
-
-    dash = (u8 *)strrchr(cc_params[0], '/');
-    dash = (u8 *)strchr(dash, '-');
-    strcat((char *)opt_cmd, (char *)dash);
-
-  }
-
-  u8 *opt_params[8];
-  u32 opt_par_cnt = 0;
-  opt_params[opt_par_cnt++] = opt_cmd;
-  opt_params[opt_par_cnt++] = "-O0";
-  opt_params[opt_par_cnt++] = "--analyze";
-  opt_params[opt_par_cnt++] = "--dot-callgraph";
-  opt_params[opt_par_cnt++] = "-o";
-  opt_params[opt_par_cnt++] = cg_file;
-  opt_params[opt_par_cnt++] = ll_file;
-  opt_params[opt_par_cnt] = NULL;
-
-  if (debug) {
-
-    DEBUGF("cd '%s';", getthecwd());
-    for (i = 0; i < (s32)opt_par_cnt; i++)
-      SAYF(" '%s'", opt_params[i]);
-    SAYF("\n");
-    fflush(stdout);
-    fflush(stderr);
-
-  }
-
-  pid = fork();
-
-  if (!pid) {
-
-    execvp(opt_params[0], (char **)opt_params);
-    FATAL("Oops, failed to execute '%s' - check your PATH", opt_params[0]);
-
-  }
-
-  if (waitpid(pid, &status, 0) <= 0) { PFATAL("waitpid() failed"); }
-  if (!be_quiet) SAYF("Done, delete '%s' now.\n", ll_file);
+  FATAL("Oops, failed to execute '%s' - check your PATH", cc_params[0]);
 
   return 0;
 

src/afl-common.c

@@ -47,6 +47,10 @@ u8  be_quiet = 0;
 u8 *doc_path = "";
 u8  last_intr = 0;
 
+#ifndef AFL_PATH
+  #define AFL_PATH "/usr/local/lib/afl/"
+#endif
+
 void detect_file_args(char **argv, u8 *prog_in, bool *use_stdin) {
 
   u32 i = 0;
@@ -372,11 +376,11 @@ u8 *get_libqasan_path(u8 *own_loc) {
 
   }
 
-  if (!access(BIN_PATH "/libqasan.so", X_OK)) {
+  if (!access(AFL_PATH "/libqasan.so", X_OK)) {
 
     if (cp) { ck_free(cp); }
 
-    return ck_strdup(BIN_PATH "/libqasan.so");
+    return ck_strdup(AFL_PATH "/libqasan.so");
 
   }
 
@@ -518,12 +522,147 @@ int parse_afl_kill_signal_env(u8 *afl_kill_signal_env, int default_signal) {
 
 }
 
+static inline unsigned int helper_min3(unsigned int a, unsigned int b,
+                                       unsigned int c) {
+
+  return a < b ? (a < c ? a : c) : (b < c ? b : c);
+
+}
+
+// from
+// https://en.wikibooks.org/wiki/Algorithm_Implementation/Strings/Levenshtein_distance#C
+static int string_distance_levenshtein(char *s1, char *s2) {
+
+  unsigned int s1len, s2len, x, y, lastdiag, olddiag;
+  s1len = strlen(s1);
+  s2len = strlen(s2);
+  unsigned int column[s1len + 1];
+  column[s1len] = 1;
+
+  for (y = 1; y <= s1len; y++)
+    column[y] = y;
+  for (x = 1; x <= s2len; x++) {
+
+    column[0] = x;
+    for (y = 1, lastdiag = x - 1; y <= s1len; y++) {
+
+      olddiag = column[y];
+      column[y] = helper_min3(column[y] + 1, column[y - 1] + 1,
+                              lastdiag + (s1[y - 1] == s2[x - 1] ? 0 : 1));
+      lastdiag = olddiag;
+
+    }
+
+  }
+
+  return column[s1len];
+
+}
+
+#define ENV_SIMILARITY_TRESHOLD 3
+
+void print_suggested_envs(char *mispelled_env) {
+
+  size_t env_name_len =
+      strcspn(mispelled_env, "=") - 4;  // remove the AFL_prefix
+  char *env_name = ck_alloc(env_name_len + 1);
+  memcpy(env_name, mispelled_env + 4, env_name_len);
+
+  char *seen = ck_alloc(sizeof(afl_environment_variables) / sizeof(char *));
+  int   found = 0;
+
+  int j;
+  for (j = 0; afl_environment_variables[j] != NULL; ++j) {
+
+    char *afl_env = afl_environment_variables[j] + 4;
+    int   distance = string_distance_levenshtein(afl_env, env_name);
+    if (distance < ENV_SIMILARITY_TRESHOLD && seen[j] == 0) {
+
+      SAYF("Did you mean %s?\n", afl_environment_variables[j]);
+      seen[j] = 1;
+      found = 1;
+
+    }
+
+  }
+
+  if (found) goto cleanup;
+
+  for (j = 0; afl_environment_variables[j] != NULL; ++j) {
+
+    char * afl_env = afl_environment_variables[j] + 4;
+    size_t afl_env_len = strlen(afl_env);
+    char * reduced = ck_alloc(afl_env_len + 1);
+
+    size_t start = 0;
+    while (start < afl_env_len) {
+
+      size_t end = start + strcspn(afl_env + start, "_") + 1;
+      memcpy(reduced, afl_env, start);
+      if (end < afl_env_len)
+        memcpy(reduced + start, afl_env + end, afl_env_len - end);
+      reduced[afl_env_len - end + start] = 0;
+
+      int distance = string_distance_levenshtein(reduced, env_name);
+      if (distance < ENV_SIMILARITY_TRESHOLD && seen[j] == 0) {
+
+        SAYF("Did you mean %s?\n", afl_environment_variables[j]);
+        seen[j] = 1;
+        found = 1;
+
+      }
+
+      start = end;
+
+    };
+
+    ck_free(reduced);
+
+  }
+
+  if (found) goto cleanup;
+
+  char * reduced = ck_alloc(env_name_len + 1);
+  size_t start = 0;
+  while (start < env_name_len) {
+
+    size_t end = start + strcspn(env_name + start, "_") + 1;
+    memcpy(reduced, env_name, start);
+    if (end < env_name_len)
+      memcpy(reduced + start, env_name + end, env_name_len - end);
+    reduced[env_name_len - end + start] = 0;
+
+    for (j = 0; afl_environment_variables[j] != NULL; ++j) {
+
+      int distance = string_distance_levenshtein(
+          afl_environment_variables[j] + 4, reduced);
+      if (distance < ENV_SIMILARITY_TRESHOLD && seen[j] == 0) {
+
+        SAYF("Did you mean %s?\n", afl_environment_variables[j]);
+        seen[j] = 1;
+
+      }
+
+    }
+
+    start = end;
+
+  };
+
+  ck_free(reduced);
+
+cleanup:
+  ck_free(env_name);
+  ck_free(seen);
+
+}
+
 void check_environment_vars(char **envp) {
 
   if (be_quiet) { return; }
 
   int   index = 0, issue_detected = 0;
-  char *env, *val;
+  char *env, *val, *ignore = getenv("AFL_IGNORE_UNKNOWN_ENVS");
   while ((env = envp[index++]) != NULL) {
 
     if (strncmp(env, "ALF_", 4) == 0 || strncmp(env, "_ALF", 4) == 0 ||
@@ -543,6 +682,7 @@ void check_environment_vars(char **envp) {
             env[strlen(afl_environment_variables[i])] == '=') {
 
           match = 1;
+
           if ((val = getenv(afl_environment_variables[i])) && !*val) {
 
             WARNF(
@@ -582,11 +722,13 @@ void check_environment_vars(char **envp) {
 
       }
 
-      if (match == 0) {
+      if (match == 0 && !ignore) {
 
         WARNF("Mistyped AFL environment variable: %s", env);
         issue_detected = 1;
 
+        print_suggested_envs(env);
+
       }
 
     }
@@ -981,7 +1123,7 @@ u8 *u_stringify_time_diff(u8 *buf, u64 cur_ms, u64 event_ms) {
 /* Reads the map size from ENV */
 u32 get_map_size(void) {
 
-  uint32_t map_size = (MAP_SIZE << 2);  // needed for target ctors :(
+  uint32_t map_size = 8000000;  // a very large default map
   char *   ptr;
 
   if ((ptr = getenv("AFL_MAP_SIZE")) || (ptr = getenv("AFL_MAPSIZE"))) {
@@ -989,12 +1131,12 @@ u32 get_map_size(void) {
     map_size = atoi(ptr);
     if (!map_size || map_size > (1 << 29)) {
 
-      FATAL("illegal AFL_MAP_SIZE %u, must be between %u and %u", map_size, 32U,
+      FATAL("illegal AFL_MAP_SIZE %u, must be between %u and %u", map_size, 64U,
             1U << 29);
 
     }
 
-    if (map_size % 32) { map_size = (((map_size >> 5) + 1) << 5); }
+    if (map_size % 64) { map_size = (((map_size >> 6) + 1) << 6); }
 
   }
 

src/afl-forkserver.c

@@ -481,27 +481,28 @@ void afl_fsrv_start(afl_forkserver_t *fsrv, char **argv,
     /* This should improve performance a bit, since it stops the linker from
        doing extra work post-fork(). */
 
-    if (!getenv("LD_BIND_LAZY")) { setenv("LD_BIND_NOW", "1", 0); }
+    if (!getenv("LD_BIND_LAZY")) { setenv("LD_BIND_NOW", "1", 1); }
 
     /* Set sane defaults for ASAN if nothing else specified. */
 
-    if (fsrv->debug == true && !getenv("ASAN_OPTIONS"))
+    if (!getenv("ASAN_OPTIONS"))
       setenv("ASAN_OPTIONS",
              "abort_on_error=1:"
              "detect_leaks=0:"
              "malloc_context_size=0:"
              "symbolize=0:"
              "allocator_may_return_null=1:"
+             "detect_odr_violation=0:"
              "handle_segv=0:"
              "handle_sigbus=0:"
              "handle_abort=0:"
              "handle_sigfpe=0:"
              "handle_sigill=0",
-             0);
+             1);
 
     /* Set sane defaults for UBSAN if nothing else specified. */
 
-    if (fsrv->debug == true && !getenv("UBSAN_OPTIONS"))
+    if (!getenv("UBSAN_OPTIONS"))
       setenv("UBSAN_OPTIONS",
              "halt_on_error=1:"
              "abort_on_error=1:"
@@ -513,7 +514,7 @@ void afl_fsrv_start(afl_forkserver_t *fsrv, char **argv,
              "handle_abort=0:"
              "handle_sigfpe=0:"
              "handle_sigill=0",
-             0);
+             1);
 
     /* Envs for QASan */
     setenv("QASAN_MAX_CALL_STACK", "0", 0);
@@ -522,7 +523,7 @@ void afl_fsrv_start(afl_forkserver_t *fsrv, char **argv,
     /* MSAN is tricky, because it doesn't support abort_on_error=1 at this
        point. So, we do this in a very hacky way. */
 
-    if (fsrv->debug == true && !getenv("MSAN_OPTIONS"))
+    if (!getenv("MSAN_OPTIONS"))
       setenv("MSAN_OPTIONS",
            "exit_code=" STRINGIFY(MSAN_ERROR) ":"
            "symbolize=0:"
@@ -535,7 +536,7 @@ void afl_fsrv_start(afl_forkserver_t *fsrv, char **argv,
            "handle_abort=0:"
            "handle_sigfpe=0:"
            "handle_sigill=0",
-           0);
+           1);
 
     fsrv->init_child_func(fsrv, argv);
 
@@ -656,11 +657,11 @@ void afl_fsrv_start(afl_forkserver_t *fsrv, char **argv,
 
         if (!fsrv->map_size) { fsrv->map_size = MAP_SIZE; }
 
-        if (unlikely(tmp_map_size % 32)) {
+        if (unlikely(tmp_map_size % 64)) {
 
           // should not happen
           WARNF("Target reported non-aligned map size of %u", tmp_map_size);
-          tmp_map_size = (((tmp_map_size + 31) >> 5) << 5);
+          tmp_map_size = (((tmp_map_size + 63) >> 6) << 6);
 
         }
 
@@ -820,7 +821,7 @@ void afl_fsrv_start(afl_forkserver_t *fsrv, char **argv,
 
            "    - The target binary requires a large map and crashes before "
            "reporting.\n"
-           "      Set a high value (e.g. AFL_MAP_SIZE=1024000) or use "
+           "      Set a high value (e.g. AFL_MAP_SIZE=8000000) or use "
            "AFL_DEBUG=1 to see the\n"
            "      message from the target binary\n\n"
 
@@ -847,7 +848,7 @@ void afl_fsrv_start(afl_forkserver_t *fsrv, char **argv,
 
            "    - The target binary requires a large map and crashes before "
            "reporting.\n"
-           "      Set a high value (e.g. AFL_MAP_SIZE=1024000) or use "
+           "      Set a high value (e.g. AFL_MAP_SIZE=8000000) or use "
            "AFL_DEBUG=1 to see the\n"
            "      message from the target binary\n\n"
 
@@ -908,10 +909,12 @@ void afl_fsrv_start(afl_forkserver_t *fsrv, char **argv,
   } else if (!fsrv->mem_limit) {
 
     SAYF("\n" cLRD "[-] " cRST
-         "Hmm, looks like the target binary terminated before we could"
-         " complete a handshake with the injected code.\n"
-         "If the target was compiled with afl-clang-lto and AFL_LLVM_MAP_ADDR"
-         " then recompiling without this parameter.\n"
+         "Hmm, looks like the target binary terminated before we could complete"
+         " a\n"
+         "handshake with the injected code.\n"
+         "Most likely the target has a huge coverage map, retry with setting"
+         " the\n"
+         "environment variable AFL_MAP_SIZE=8000000\n"
          "Otherwise there is a horrible bug in the fuzzer.\n"
          "Poke <afl-users@googlegroups.com> for troubleshooting tips.\n");
 
@@ -927,6 +930,11 @@ void afl_fsrv_start(afl_forkserver_t *fsrv, char **argv,
         "explanations:\n\n"
 
         "%s"
+
+        "    - Most likely the target has a huge coverage map, retry with "
+        "setting the\n"
+        "      environment variable AFL_MAP_SIZE=8000000\n\n"
+
         "    - The current memory limit (%s) is too restrictive, causing an "
         "OOM\n"
         "      fault in the dynamic linker. This can be fixed with the -m "

src/afl-fuzz-extras.c

@@ -387,6 +387,130 @@ static inline u8 memcmp_nocase(u8 *m1, u8 *m2, u32 len) {
 
 }
 
+/* add an extra/dict/token - no checks performed, no sorting */
+
+static void add_extra_nocheck(afl_state_t *afl, u8 *mem, u32 len) {
+
+  afl->extras = afl_realloc((void **)&afl->extras,
+                            (afl->extras_cnt + 1) * sizeof(struct extra_data));
+
+  if (unlikely(!afl->extras)) { PFATAL("alloc"); }
+
+  afl->extras[afl->extras_cnt].data = ck_alloc(len);
+  afl->extras[afl->extras_cnt].len = len;
+  memcpy(afl->extras[afl->extras_cnt].data, mem, len);
+  afl->extras_cnt++;
+
+  /* We only want to print this once */
+
+  if (afl->extras_cnt == afl->max_det_extras + 1) {
+
+    WARNF("More than %u tokens - will use them probabilistically.",
+          afl->max_det_extras);
+
+  }
+
+}
+
+/* Sometimes strings in input is transformed to unicode internally, so for
+   fuzzing we should attempt to de-unicode if it looks like simple unicode */
+
+void deunicode_extras(afl_state_t *afl) {
+
+  if (!afl->extras_cnt) return;
+
+  u32 i, j, orig_cnt = afl->extras_cnt;
+  u8  buf[64];
+
+  for (i = 0; i < orig_cnt; ++i) {
+
+    if (afl->extras[i].len < 6 || afl->extras[i].len > 64 ||
+        afl->extras[i].len % 2) {
+
+      continue;
+
+    }
+
+    u32 k = 0, z1 = 0, z2 = 0, z3 = 0, z4 = 0, half = afl->extras[i].len >> 1;
+    u32 quarter = half >> 1;
+
+    for (j = 0; j < afl->extras[i].len; ++j) {
+
+      switch (j % 4) {
+
+        case 2:
+          if (!afl->extras[i].data[j]) { ++z3; }
+          // fall through
+        case 0:
+          if (!afl->extras[i].data[j]) { ++z1; }
+          break;
+        case 3:
+          if (!afl->extras[i].data[j]) { ++z4; }
+          // fall through
+        case 1:
+          if (!afl->extras[i].data[j]) { ++z2; }
+          break;
+
+      }
+
+    }
+
+    if ((z1 < half && z2 < half) || z1 + z2 == afl->extras[i].len) { continue; }
+
+    // also maybe 32 bit unicode?
+    if (afl->extras[i].len % 4 == 0 && afl->extras[i].len >= 12 &&
+        (z3 == quarter || z4 == quarter) && z1 + z2 == quarter * 3) {
+
+      for (j = 0; j < afl->extras[i].len; ++j) {
+
+        if (z4 < quarter) {
+
+          if (j % 4 == 3) { buf[k++] = afl->extras[i].data[j]; }
+
+        } else if (z3 < quarter) {
+
+          if (j % 4 == 2) { buf[k++] = afl->extras[i].data[j]; }
+
+        } else if (z2 < half) {
+
+          if (j % 4 == 1) { buf[k++] = afl->extras[i].data[j]; }
+
+        } else {
+
+          if (j % 4 == 0) { buf[k++] = afl->extras[i].data[j]; }
+
+        }
+
+      }
+
+      add_extra_nocheck(afl, buf, k);
+      k = 0;
+
+    }
+
+    for (j = 0; j < afl->extras[i].len; ++j) {
+
+      if (z1 < half) {
+
+        if (j % 2 == 0) { buf[k++] = afl->extras[i].data[j]; }
+
+      } else {
+
+        if (j % 2 == 1) { buf[k++] = afl->extras[i].data[j]; }
+
+      }
+
+    }
+
+    add_extra_nocheck(afl, buf, k);
+
+  }
+
+  qsort(afl->extras, afl->extras_cnt, sizeof(struct extra_data),
+        compare_extras_len);
+
+}
+
 /* Removes duplicates from the loaded extras. This can happen if multiple files
    are loaded */
 
@@ -396,9 +520,9 @@ void dedup_extras(afl_state_t *afl) {
 
   u32 i, j, orig_cnt = afl->extras_cnt;
 
-  for (i = 0; i < afl->extras_cnt - 1; i++) {
+  for (i = 0; i < afl->extras_cnt - 1; ++i) {
 
-    for (j = i + 1; j < afl->extras_cnt; j++) {
+    for (j = i + 1; j < afl->extras_cnt; ++j) {
 
     restart_dedup:
 
@@ -462,30 +586,11 @@ void add_extra(afl_state_t *afl, u8 *mem, u32 len) {
 
   }
 
-  afl->extras = afl_realloc((void **)&afl->extras,
-                            (afl->extras_cnt + 1) * sizeof(struct extra_data));
-
-  if (unlikely(!afl->extras)) { PFATAL("alloc"); }
-
-  afl->extras[afl->extras_cnt].data = ck_alloc(len);
-  afl->extras[afl->extras_cnt].len = len;
-
-  memcpy(afl->extras[afl->extras_cnt].data, mem, len);
-
-  afl->extras_cnt++;
+  add_extra_nocheck(afl, mem, len);
 
   qsort(afl->extras, afl->extras_cnt, sizeof(struct extra_data),
         compare_extras_len);
 
-  /* We only want to print this once */
-
-  if (afl->extras_cnt == afl->max_det_extras + 1) {
-
-    WARNF("More than %u tokens - will use them probabilistically.",
-          afl->max_det_extras);
-
-  }
-
 }
 
 /* Maybe add automatic extra. */

src/afl-fuzz-init.c

@@ -828,7 +828,7 @@ void perform_dry_run(afl_state_t *afl) {
   for (idx = 0; idx < afl->queued_paths; idx++) {
 
     q = afl->queue_buf[idx];
-    if (unlikely(q->disabled)) { continue; }
+    if (unlikely(!q || q->disabled)) { continue; }
 
     u8  res;
     s32 fd;
@@ -882,32 +882,23 @@ void perform_dry_run(afl_state_t *afl) {
 
         if (afl->timeout_given) {
 
-          /* The -t nn+ syntax in the command line sets afl->timeout_given to
-             '2' and instructs afl-fuzz to tolerate but skip queue entries that
-             time out. */
+          /* if we have a timeout but a timeout value was given then always
+             skip. The '+' meaning has been changed! */
+          WARNF("Test case results in a timeout (skipping)");
+          ++cal_failures;
+          q->cal_failed = CAL_CHANCES;
+          q->disabled = 1;
+          q->perf_score = 0;
 
-          if (afl->timeout_given > 1) {
+          if (!q->was_fuzzed) {
 
-            WARNF("Test case results in a timeout (skipping)");
-            q->cal_failed = CAL_CHANCES;
-            ++cal_failures;
-            break;
+            q->was_fuzzed = 1;
+            --afl->pending_not_fuzzed;
+            --afl->active_paths;
 
           }
 
-          SAYF("\n" cLRD "[-] " cRST
-               "The program took more than %u ms to process one of the initial "
-               "test cases.\n"
-               "    Usually, the right thing to do is to relax the -t option - "
-               "or to delete it\n"
-               "    altogether and allow the fuzzer to auto-calibrate. That "
-               "said, if you know\n"
-               "    what you are doing and want to simply skip the unruly test "
-               "cases, append\n"
-               "    '+' at the end of the value passed to -t ('-t %u+').\n",
-               afl->fsrv.exec_tmout, afl->fsrv.exec_tmout);
-
-          FATAL("Test case '%s' results in a timeout", fn);
+          break;
 
         } else {
 
@@ -1060,16 +1051,25 @@ void perform_dry_run(afl_state_t *afl) {
         p->perf_score = 0;
 
         u32 i = 0;
-        while (unlikely(afl->queue_buf[i]->disabled)) {
+        while (unlikely(i < afl->queued_paths && afl->queue_buf[i] &&
+                        afl->queue_buf[i]->disabled)) {
 
           ++i;
 
         }
 
-        afl->queue = afl->queue_buf[i];
+        if (i < afl->queued_paths && afl->queue_buf[i]) {
+
+          afl->queue = afl->queue_buf[i];
+
+        } else {
+
+          afl->queue = afl->queue_buf[0];
+
+        }
 
         afl->max_depth = 0;
-        for (i = 0; i < afl->queued_paths; i++) {
+        for (i = 0; i < afl->queued_paths && likely(afl->queue_buf[i]); i++) {
 
           if (!afl->queue_buf[i]->disabled &&
               afl->queue_buf[i]->depth > afl->max_depth)
@@ -1136,10 +1136,11 @@ void perform_dry_run(afl_state_t *afl) {
   for (idx = 0; idx < afl->queued_paths; idx++) {
 
     q = afl->queue_buf[idx];
-    if (q->disabled || q->cal_failed || !q->exec_cksum) { continue; }
+    if (!q || q->disabled || q->cal_failed || !q->exec_cksum) { continue; }
 
     u32 done = 0;
-    for (i = idx + 1; i < afl->queued_paths && !done; i++) {
+    for (i = idx + 1;
+         i < afl->queued_paths && !done && likely(afl->queue_buf[i]); i++) {
 
       struct queue_entry *p = afl->queue_buf[i];
       if (p->disabled || p->cal_failed || !p->exec_cksum) { continue; }
@@ -1191,7 +1192,7 @@ void perform_dry_run(afl_state_t *afl) {
 
     for (idx = 0; idx < afl->queued_paths; idx++) {
 
-      if (!afl->queue_buf[idx]->disabled &&
+      if (afl->queue_buf[idx] && !afl->queue_buf[idx]->disabled &&
           afl->queue_buf[idx]->depth > afl->max_depth)
         afl->max_depth = afl->queue_buf[idx]->depth;
 
@@ -1247,7 +1248,7 @@ void pivot_inputs(afl_state_t *afl) {
 
   ACTF("Creating hard links for all input files...");
 
-  for (i = 0; i < afl->queued_paths; i++) {
+  for (i = 0; i < afl->queued_paths && likely(afl->queue_buf[i]); i++) {
 
     q = afl->queue_buf[i];
 
@@ -2017,7 +2018,7 @@ void setup_dirs_fds(afl_state_t *afl) {
   fprintf(afl->fsrv.plot_file,
           "# unix_time, cycles_done, cur_path, paths_total, "
           "pending_total, pending_favs, map_size, unique_crashes, "
-          "unique_hangs, max_depth, execs_per_sec\n");
+          "unique_hangs, max_depth, execs_per_sec, total_execs, edges_found\n");
   fflush(afl->fsrv.plot_file);
 
   /* ignore errors */
@@ -2457,7 +2458,7 @@ void check_asan_opts(afl_state_t *afl) {
 
     }
 
-    if (!strstr(x, "symbolize=0")) {
+    if (!afl->debug && !strstr(x, "symbolize=0")) {
 
       FATAL("Custom MSAN_OPTIONS set without symbolize=0 - please fix!");
 

src/afl-fuzz-queue.c

@@ -198,34 +198,35 @@ void create_alias_table(afl_state_t *afl) {
   while (nS)
     afl->alias_probability[S[--nS]] = 1;
 
-#ifdef INTROSPECTION
-  u8 fn[PATH_MAX];
-  snprintf(fn, PATH_MAX, "%s/introspection_corpus.txt", afl->out_dir);
-  FILE *f = fopen(fn, "a");
-  if (f) {
+  /*
+  #ifdef INTROSPECTION
+    u8 fn[PATH_MAX];
+    snprintf(fn, PATH_MAX, "%s/introspection_corpus.txt", afl->out_dir);
+    FILE *f = fopen(fn, "a");
+    if (f) {
 
-    for (i = 0; i < n; i++) {
+      for (i = 0; i < n; i++) {
 
-      struct queue_entry *q = afl->queue_buf[i];
-      fprintf(
-          f,
-          "entry=%u name=%s favored=%s variable=%s disabled=%s len=%u "
-          "exec_us=%u "
-          "bitmap_size=%u bitsmap_size=%u tops=%u weight=%f perf_score=%f\n",
-          i, q->fname, q->favored ? "true" : "false",
-          q->var_behavior ? "true" : "false", q->disabled ? "true" : "false",
-          q->len, (u32)q->exec_us, q->bitmap_size, q->bitsmap_size, q->tc_ref,
-          q->weight, q->perf_score);
+        struct queue_entry *q = afl->queue_buf[i];
+        fprintf(
+            f,
+            "entry=%u name=%s favored=%s variable=%s disabled=%s len=%u "
+            "exec_us=%u "
+            "bitmap_size=%u bitsmap_size=%u tops=%u weight=%f perf_score=%f\n",
+            i, q->fname, q->favored ? "true" : "false",
+            q->var_behavior ? "true" : "false", q->disabled ? "true" : "false",
+            q->len, (u32)q->exec_us, q->bitmap_size, q->bitsmap_size, q->tc_ref,
+            q->weight, q->perf_score);
+
+      }
+
+      fprintf(f, "\n");
+      fclose(f);
 
     }
 
-    fprintf(f, "\n");
-    fclose(f);
-
-  }
-
-#endif
-
+  #endif
+  */
   /*
   fprintf(stderr, "  entry  alias  probability  perf_score   weight
   filename\n"); for (u32 i = 0; i < n; ++i) fprintf(stderr, "  %5u  %5u  %11u
@@ -324,7 +325,7 @@ static u8 check_if_text(afl_state_t *afl, struct queue_entry *q) {
 
   if (len >= MAX_FILE) len = MAX_FILE - 1;
   if ((fd = open(q->fname, O_RDONLY)) < 0) return 0;
-  buf = afl_realloc(AFL_BUF_PARAM(in_scratch), len);
+  buf = afl_realloc(AFL_BUF_PARAM(in_scratch), len + 1);
   comp = read(fd, buf, len);
   close(fd);
   if (comp != (ssize_t)len) return 0;

src/afl-fuzz-redqueen.c

@@ -30,8 +30,6 @@
 
 //#define _DEBUG
 //#define CMPLOG_INTROSPECTION
-#define COMBINE
-#define ARITHMETIC_LESSER_GREATER
 
 // CMP attribute enum
 enum {
@@ -206,14 +204,31 @@ static void type_replace(afl_state_t *afl, u8 *buf, u32 len) {
         case '\t':
           c = ' ';
           break;
-          /*
-                case '\r':
-                case '\n':
-                  // nothing ...
-                  break;
-          */
+        case '\r':
+          c = '\n';
+          break;
+        case '\n':
+          c = '\r';
+          break;
+        case 0:
+          c = 1;
+          break;
+        case 1:
+          c = 0;
+          break;
+        case 0xff:
+          c = 0;
+          break;
         default:
-          c = (buf[i] ^ 0xff);
+          if (buf[i] < 32) {
+
+            c = (buf[i] ^ 0x1f);
+
+          } else {
+
+            c = (buf[i] ^ 0x7f);  // we keep the highest bit
+
+          }
 
       }
 
@@ -383,6 +398,7 @@ static u8 colorization(afl_state_t *afl, u8 *buf, u32 len,
     rng = ranges;
     ranges = rng->next;
     ck_free(rng);
+    rng = NULL;
 
   }
 
@@ -421,8 +437,9 @@ static u8 colorization(afl_state_t *afl, u8 *buf, u32 len,
 
   if (taint) {
 
-    if (len / positions == 1 && positions > CMPLOG_POSITIONS_MAX &&
-        afl->active_paths / afl->colorize_success > CMPLOG_CORPUS_PERCENT) {
+    if (afl->colorize_success &&
+        (len / positions == 1 && positions > CMPLOG_POSITIONS_MAX &&
+         afl->active_paths / afl->colorize_success > CMPLOG_CORPUS_PERCENT)) {
 
 #ifdef _DEBUG
       fprintf(stderr, "Colorization unsatisfactory\n");
@@ -456,6 +473,15 @@ static u8 colorization(afl_state_t *afl, u8 *buf, u32 len,
   return 0;
 
 checksum_fail:
+  while (ranges) {
+
+    rng = ranges;
+    ranges = rng->next;
+    ck_free(rng);
+    rng = NULL;
+
+  }
+
   ck_free(backup);
   ck_free(changed);
 
@@ -496,7 +522,7 @@ static u8 its_fuzz(afl_state_t *afl, u8 *buf, u32 len, u8 *status) {
 
 }
 
-#ifdef CMPLOG_TRANSFORM
+//#ifdef CMPLOG_SOLVE_TRANSFORM
 static int strntoll(const char *str, size_t sz, char **end, int base,
                     long long *out) {
 
@@ -504,6 +530,8 @@ static int strntoll(const char *str, size_t sz, char **end, int base,
   long long   ret;
   const char *beg = str;
 
+  if (!str || !sz) { return 1; }
+
   for (; beg && sz && *beg == ' '; beg++, sz--) {};
 
   if (!sz) return 1;
@@ -527,6 +555,8 @@ static int strntoull(const char *str, size_t sz, char **end, int base,
   unsigned long long ret;
   const char *       beg = str;
 
+  if (!str || !sz) { return 1; }
+
   for (; beg && sz && *beg == ' '; beg++, sz--)
     ;
 
@@ -577,7 +607,7 @@ static int is_hex(const char *str) {
 
 }
 
-  #ifdef CMPLOG_TRANSFORM_BASE64
+#ifdef CMPLOG_SOLVE_TRANSFORM_BASE64
 // tests 4 bytes at location
 static int is_base64(const char *str) {
 
@@ -690,10 +720,10 @@ static void to_base64(u8 *src, u8 *dst, u32 dst_len) {
 
 }
 
-  #endif
-
 #endif
 
+//#endif
+
 static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
                               u64 pattern, u64 repl, u64 o_pattern,
                               u64 changed_val, u8 attr, u32 idx, u32 taint_len,
@@ -717,9 +747,9 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
   //         o_pattern, pattern, repl, changed_val, idx, taint_len,
   //         h->shape + 1, attr);
 
-#ifdef CMPLOG_TRANSFORM
+  //#ifdef CMPLOG_SOLVE_TRANSFORM
   // reverse atoi()/strnu?toll() is expensive, so we only to it in lvl 3
-  if (lvl & LVL3) {
+  if (afl->cmplog_enable_transform && (lvl & LVL3)) {
 
     u8 *               endptr;
     u8                 use_num = 0, use_unum = 0;
@@ -740,11 +770,11 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
 
     }
 
-  #ifdef _DEBUG
+#ifdef _DEBUG
     if (idx == 0)
       fprintf(stderr, "ASCII is=%u use_num=%u use_unum=%u idx=%u %llx==%llx\n",
               afl->queue_cur->is_ascii, use_num, use_unum, idx, num, pattern);
-  #endif
+#endif
 
     // num is likely not pattern as atoi("AAA") will be zero...
     if (use_num && ((u64)num == pattern || !num)) {
@@ -794,37 +824,82 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
     // Try to identify transform magic
     if (pattern != o_pattern && repl == changed_val && attr <= IS_EQUAL) {
 
-      u64 *ptr = (u64 *)&buf[idx];
-      u64 *o_ptr = (u64 *)&orig_buf[idx];
-      u64  b_val, o_b_val, mask;
+      u64 b_val, o_b_val, mask;
+      u8  bytes;
 
       switch (SHAPE_BYTES(h->shape)) {
 
         case 0:
         case 1:
-          b_val = (u64)(*ptr % 0x100);
+          bytes = 1;
+          break;
+        case 2:
+          bytes = 2;
+          break;
+        case 3:
+        case 4:
+          bytes = 4;
+          break;
+        default:
+          bytes = 8;
+
+      }
+
+      // necessary for preventing heap access overflow
+      bytes = MIN(bytes, len - idx);
+
+      switch (bytes) {
+
+        case 0:                        // cannot happen
+          b_val = o_b_val = mask = 0;  // keep the linters happy
+          break;
+        case 1: {
+
+          u8 *ptr = (u8 *)&buf[idx];
+          u8 *o_ptr = (u8 *)&orig_buf[idx];
+          b_val = (u64)(*ptr);
           o_b_val = (u64)(*o_ptr % 0x100);
           mask = 0xff;
           break;
+
+        }
+
         case 2:
-        case 3:
-          b_val = (u64)(*ptr % 0x10000);
-          o_b_val = (u64)(*o_ptr % 0x10000);
+        case 3: {
+
+          u16 *ptr = (u16 *)&buf[idx];
+          u16 *o_ptr = (u16 *)&orig_buf[idx];
+          b_val = (u64)(*ptr);
+          o_b_val = (u64)(*o_ptr);
           mask = 0xffff;
           break;
+
+        }
+
         case 4:
         case 5:
         case 6:
-        case 7:
-          b_val = (u64)(*ptr % 0x100000000);
-          o_b_val = (u64)(*o_ptr % 0x100000000);
+        case 7: {
+
+          u32 *ptr = (u32 *)&buf[idx];
+          u32 *o_ptr = (u32 *)&orig_buf[idx];
+          b_val = (u64)(*ptr);
+          o_b_val = (u64)(*o_ptr);
           mask = 0xffffffff;
           break;
-        default:
-          b_val = *ptr;
-          o_b_val = *o_ptr;
+
+        }
+
+        default: {
+
+          u64 *ptr = (u64 *)&buf[idx];
+          u64 *o_ptr = (u64 *)&orig_buf[idx];
+          b_val = (u64)(*ptr);
+          o_b_val = (u64)(*o_ptr);
           mask = 0xffffffffffffffff;
 
+        }
+
       }
 
       // test for arithmetic, eg. "if ((user_val - 0x1111) == 0x1234) ..."
@@ -984,7 +1059,7 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
 
   }
 
-#endif
+  //#endif
 
   // we only allow this for ascii2integer (above) so leave if this is the case
   if (unlikely(pattern == o_pattern)) { return 0; }
@@ -1009,7 +1084,7 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
         u64 tmp_64 = *buf_64;
         *buf_64 = repl;
         if (unlikely(its_fuzz(afl, buf, len, status))) { return 1; }
-#ifdef COMBINE
+#ifdef CMPLOG_COMBINE
         if (*status == 1) { memcpy(cbuf + idx, buf_64, 8); }
 #endif
         *buf_64 = tmp_64;
@@ -1050,7 +1125,7 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
         u32 tmp_32 = *buf_32;
         *buf_32 = (u32)repl;
         if (unlikely(its_fuzz(afl, buf, len, status))) { return 1; }
-#ifdef COMBINE
+#ifdef CMPLOG_COMBINE
         if (*status == 1) { memcpy(cbuf + idx, buf_32, 4); }
 #endif
         *buf_32 = tmp_32;
@@ -1084,7 +1159,7 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
         u16 tmp_16 = *buf_16;
         *buf_16 = (u16)repl;
         if (unlikely(its_fuzz(afl, buf, len, status))) { return 1; }
-#ifdef COMBINE
+#ifdef CMPLOG_COMBINE
         if (*status == 1) { memcpy(cbuf + idx, buf_16, 2); }
 #endif
         *buf_16 = tmp_16;
@@ -1122,7 +1197,7 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
         u8 tmp_8 = *buf_8;
         *buf_8 = (u8)repl;
         if (unlikely(its_fuzz(afl, buf, len, status))) { return 1; }
-#ifdef COMBINE
+#ifdef CMPLOG_COMBINE
         if (*status == 1) { cbuf[idx] = *buf_8; }
 #endif
         *buf_8 = tmp_8;
@@ -1139,8 +1214,12 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
   //       16 = modified float, 32 = modified integer (modified = wont match
   //                                                   in original buffer)
 
-#ifdef ARITHMETIC_LESSER_GREATER
-  if (lvl < LVL3 || attr == IS_TRANSFORM) { return 0; }
+  //#ifdef CMPLOG_SOLVE_ARITHMETIC
+  if (!afl->cmplog_enable_arith || lvl < LVL3 || attr == IS_TRANSFORM) {
+
+    return 0;
+
+  }
 
   if (!(attr & (IS_GREATER | IS_LESSER)) || SHAPE_BYTES(h->shape) < 4) {
 
@@ -1245,11 +1324,11 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
       double *f = (double *)&repl;
       float   g = (float)*f;
       repl_new = 0;
-  #if (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+#if (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
       memcpy((char *)&repl_new, (char *)&g, 4);
-  #else
+#else
       memcpy(((char *)&repl_new) + 4, (char *)&g, 4);
-  #endif
+#endif
       changed_val = repl_new;
       h->shape = 3;  // modify shape
 
@@ -1304,7 +1383,7 @@ static u8 cmp_extend_encoding(afl_state_t *afl, struct cmp_header *h,
 
   }
 
-#endif                                         /* ARITHMETIC_LESSER_GREATER */
+  //#endif                                           /* CMPLOG_SOLVE_ARITHMETIC
 
   return 0;
 
@@ -1366,7 +1445,7 @@ static u8 cmp_extend_encodingN(afl_state_t *afl, struct cmp_header *h,
 
       if (unlikely(its_fuzz(afl, buf, len, status))) { return 1; }
 
-  #ifdef COMBINE
+  #ifdef CMPLOG_COMBINE
       if (*status == 1) { memcpy(cbuf + idx, r, shape); }
   #endif
 
@@ -1463,7 +1542,7 @@ static void try_to_add_to_dictN(afl_state_t *afl, u128 v, u8 size) {
   for (k = 0; k < size; ++k) {
 
   #else
-  u32 off = 16 - size;
+  u32    off = 16 - size;
   for (k = 16 - size; k < 16; ++k) {
 
   #endif
@@ -1499,11 +1578,12 @@ static u8 cmp_fuzz(afl_state_t *afl, u32 key, u8 *orig_buf, u8 *buf, u8 *cbuf,
   struct cmp_header *h = &afl->shm.cmp_map->headers[key];
   struct tainted *   t;
   u32                i, j, idx, taint_len, loggeds;
-  u32                have_taint = 1, is_n = 0;
+  u32                have_taint = 1;
   u8                 status = 0, found_one = 0;
 
   /* loop cmps are useless, detect and ignore them */
 #ifdef WORD_SIZE_64
+  u32  is_n = 0;
   u128 s128_v0 = 0, s128_v1 = 0, orig_s128_v0 = 0, orig_s128_v1 = 0;
 #endif
   u64 s_v0, s_v1;
@@ -1521,6 +1601,7 @@ static u8 cmp_fuzz(afl_state_t *afl, u32 key, u8 *orig_buf, u8 *buf, u8 *cbuf,
 
   }
 
+#ifdef WORD_SIZE_64
   switch (SHAPE_BYTES(h->shape)) {
 
     case 1:
@@ -1533,6 +1614,8 @@ static u8 cmp_fuzz(afl_state_t *afl, u32 key, u8 *orig_buf, u8 *buf, u8 *cbuf,
 
   }
 
+#endif
+
   for (i = 0; i < loggeds; ++i) {
 
     struct cmp_operands *o = &afl->shm.cmp_map->log[key][i];
@@ -1770,20 +1853,20 @@ static u8 cmp_fuzz(afl_state_t *afl, u32 key, u8 *orig_buf, u8 *buf, u8 *cbuf,
 }
 
 static u8 rtn_extend_encoding(afl_state_t *afl, u8 *pattern, u8 *repl,
-                              u8 *o_pattern, u8 *changed_val, u32 idx,
+                              u8 *o_pattern, u8 *changed_val, u8 plen, u32 idx,
                               u32 taint_len, u8 *orig_buf, u8 *buf, u8 *cbuf,
                               u32 len, u8 lvl, u8 *status) {
 
-#ifndef COMBINE
+#ifndef CMPLOG_COMBINE
   (void)(cbuf);
 #endif
-#ifndef CMPLOG_TRANSFORM
-  (void)(changed_val);
-#endif
+  //#ifndef CMPLOG_SOLVE_TRANSFORM
+  //  (void)(changed_val);
+  //#endif
 
   u8  save[40];
   u32 saved_idx = idx, pre, from = 0, to = 0, i, j;
-  u32 its_len = MIN((u32)32, len - idx);
+  u32 its_len = MIN((u32)plen, len - idx);
   its_len = MIN(its_len, taint_len);
   u32 saved_its_len = its_len;
 
@@ -1847,7 +1930,7 @@ static u8 rtn_extend_encoding(afl_state_t *afl, u8 *pattern, u8 *repl,
 
         if (unlikely(its_fuzz(afl, buf, len, status))) { return 1; }
 
-#ifdef COMBINE
+#ifdef CMPLOG_COMBINE
         if (*status == 1) { memcpy(cbuf + idx, &buf[idx], i); }
 #endif
 
@@ -1859,16 +1942,16 @@ static u8 rtn_extend_encoding(afl_state_t *afl, u8 *pattern, u8 *repl,
 
   }
 
-#ifdef CMPLOG_TRANSFORM
+  //#ifdef CMPLOG_SOLVE_TRANSFORM
 
   if (*status == 1) return 0;
 
-  if (lvl & LVL3) {
+  if (afl->cmplog_enable_transform && (lvl & LVL3)) {
 
     u32 toupper = 0, tolower = 0, xor = 0, arith = 0, tohex = 0, fromhex = 0;
-  #ifdef CMPLOG_TRANSFORM_BASE64
+#ifdef CMPLOG_SOLVE_TRANSFORM_BASE64
     u32 tob64 = 0, fromb64 = 0;
-  #endif
+#endif
     u32 from_0 = 0, from_x = 0, from_X = 0, from_slash = 0, from_up = 0;
     u32 to_0 = 0, to_x = 0, to_slash = 0, to_up = 0;
     u8  xor_val[32], arith_val[32], tmp[48];
@@ -1964,7 +2047,7 @@ static u8 rtn_extend_encoding(afl_state_t *afl, u8 *pattern, u8 *repl,
 
       }
 
-  #ifdef CMPLOG_TRANSFORM_BASE64
+#ifdef CMPLOG_SOLVE_TRANSFORM_BASE64
       if (i % 3 == 2 && i < 24) {
 
         if (is_base64(repl + ((i / 3) << 2))) tob64 += 3;
@@ -1977,7 +2060,7 @@ static u8 rtn_extend_encoding(afl_state_t *afl, u8 *pattern, u8 *repl,
 
       }
 
-  #endif
+#endif
 
       if ((o_pattern[i] ^ orig_buf[idx + i]) == xor_val[i] && xor_val[i]) {
 
@@ -2005,20 +2088,20 @@ static u8 rtn_extend_encoding(afl_state_t *afl, u8 *pattern, u8 *repl,
 
       }
 
-  #ifdef _DEBUG
+#ifdef _DEBUG
       fprintf(stderr,
               "RTN idx=%u loop=%u xor=%u arith=%u tolower=%u toupper=%u "
               "tohex=%u fromhex=%u to_0=%u to_slash=%u to_x=%u "
               "from_0=%u from_slash=%u from_x=%u\n",
               idx, i, xor, arith, tolower, toupper, tohex, fromhex, to_0,
               to_slash, to_x, from_0, from_slash, from_x);
-    #ifdef CMPLOG_TRANSFORM_BASE64
+  #ifdef CMPLOG_SOLVE_TRANSFORM_BASE64
       fprintf(stderr, "RTN idx=%u loop=%u tob64=%u from64=%u\n", tob64,
               fromb64);
-    #endif
   #endif
+#endif
 
-  #ifdef CMPLOG_TRANSFORM_BASE64
+#ifdef CMPLOG_SOLVE_TRANSFORM_BASE64
       // input is base64 and converted to binary? convert repl to base64!
       if ((i % 4) == 3 && i < 24 && fromb64 > i) {
 
@@ -2041,7 +2124,7 @@ static u8 rtn_extend_encoding(afl_state_t *afl, u8 *pattern, u8 *repl,
 
       }
 
-  #endif
+#endif
 
       // input is converted to hex? convert repl to binary!
       if (i < 16 && tohex > i) {
@@ -2170,16 +2253,16 @@ static u8 rtn_extend_encoding(afl_state_t *afl, u8 *pattern, u8 *repl,
 
       }
 
-  #ifdef COMBINE
+#ifdef CMPLOG_COMBINE
       if (*status == 1) { memcpy(cbuf + idx, &buf[idx], i + 1); }
-  #endif
+#endif
 
       if ((i >= 7 &&
            (i >= xor&&i >= arith &&i >= tolower &&i >= toupper &&i > tohex &&i >
                 (fromhex + from_0 + from_x + from_slash + 1)
-  #ifdef CMPLOG_TRANSFORM_BASE64
+#ifdef CMPLOG_SOLVE_TRANSFORM_BASE64
             && i > tob64 + 3 && i > fromb64 + 4
-  #endif
+#endif
             )) ||
           repl[i] != changed_val[i] || *status == 1) {
 
@@ -2193,7 +2276,7 @@ static u8 rtn_extend_encoding(afl_state_t *afl, u8 *pattern, u8 *repl,
 
   }
 
-#endif
+  //#endif
 
   return 0;
 
@@ -2282,9 +2365,9 @@ static u8 rtn_fuzz(afl_state_t *afl, u32 key, u8 *orig_buf, u8 *buf, u8 *cbuf,
 
       status = 0;
 
-      if (unlikely(rtn_extend_encoding(afl, o->v0, o->v1, orig_o->v0,
-                                       orig_o->v1, idx, taint_len, orig_buf,
-                                       buf, cbuf, len, lvl, &status))) {
+      if (unlikely(rtn_extend_encoding(
+              afl, o->v0, o->v1, orig_o->v0, orig_o->v1, SHAPE_BYTES(h->shape),
+              idx, taint_len, orig_buf, buf, cbuf, len, lvl, &status))) {
 
         return 1;
 
@@ -2299,9 +2382,9 @@ static u8 rtn_fuzz(afl_state_t *afl, u32 key, u8 *orig_buf, u8 *buf, u8 *cbuf,
 
       status = 0;
 
-      if (unlikely(rtn_extend_encoding(afl, o->v1, o->v0, orig_o->v1,
-                                       orig_o->v0, idx, taint_len, orig_buf,
-                                       buf, cbuf, len, lvl, &status))) {
+      if (unlikely(rtn_extend_encoding(
+              afl, o->v1, o->v0, orig_o->v1, orig_o->v0, SHAPE_BYTES(h->shape),
+              idx, taint_len, orig_buf, buf, cbuf, len, lvl, &status))) {
 
         return 1;
 
@@ -2469,7 +2552,7 @@ u8 input_to_state_stage(afl_state_t *afl, u8 *orig_buf, u8 *buf, u32 len) {
   u32 lvl = (afl->queue_cur->colorized ? 0 : LVL1) +
             (afl->cmplog_lvl == CMPLOG_LVL_MAX ? LVL3 : 0);
 
-#ifdef COMBINE
+#ifdef CMPLOG_COMBINE
   u8 *cbuf = afl_realloc((void **)&afl->in_scratch_buf, len + 128);
   memcpy(cbuf, orig_buf, len);
   u8 *virgin_backup = afl_realloc((void **)&afl->ex_buf, afl->shm.map_size);
@@ -2526,9 +2609,9 @@ u8 input_to_state_stage(afl_state_t *afl, u8 *orig_buf, u8 *buf, u32 len) {
 
     } else if ((lvl & LVL1)
 
-#ifdef CMPLOG_TRANSFORM
-               || (lvl & LVL3)
-#endif
+               //#ifdef CMPLOG_SOLVE_TRANSFORM
+               || ((lvl & LVL3) && afl->cmplog_enable_transform)
+               //#endif
     ) {
 
       if (unlikely(rtn_fuzz(afl, k, orig_buf, buf, cbuf, len, lvl, taint))) {
@@ -2583,7 +2666,7 @@ exit_its:
 
   }
 
-#ifdef COMBINE
+#ifdef CMPLOG_COMBINE
   if (afl->queued_paths + afl->unique_crashes > orig_hit_cnt + 1) {
 
     // copy the current virgin bits so we can recover the information
@@ -2607,9 +2690,9 @@ exit_its:
     }
 
   #else
-    u32 *v = (u64 *)afl->virgin_bits;
-    u32 *s = (u64 *)virgin_save;
-    u32 i;
+    u32 *v = (u32 *)afl->virgin_bits;
+    u32 *s = (u32 *)virgin_save;
+    u32  i;
     for (i = 0; i < (afl->shm.map_size >> 2); i++) {
 
       v[i] &= s[i];
@@ -2622,7 +2705,7 @@ exit_its:
     dump("COMB", cbuf, len);
     if (status == 1) {
 
-      fprintf(stderr, "NEW COMBINED\n");
+      fprintf(stderr, "NEW CMPLOG_COMBINED\n");
 
     } else {
 
@@ -2672,7 +2755,3 @@ exit_its:
 
 }
 
-#ifdef COMBINE
-  #undef COMBINE
-#endif
-

src/afl-fuzz-run.c

@@ -707,6 +707,8 @@ void sync_fuzzers(afl_state_t *afl) {
 
   if (afl->foreign_sync_cnt) read_foreign_testcases(afl, 0);
 
+  afl->last_sync_time = get_cur_time();
+
 }
 
 /* Trim all new test cases to save cycles when doing deterministic checks. The

src/afl-fuzz-state.c

@@ -486,6 +486,8 @@ void read_afl_environment(afl_state_t *afl, char **envp) {
         WARNF("Mistyped AFL environment variable: %s", env);
         issue_detected = 1;
 
+        print_suggested_envs(env);
+
       }
 
     }

src/afl-fuzz-stats.c

@@ -185,15 +185,14 @@ void load_stats_file(afl_state_t *afl) {
 
 /* Update stats file for unattended monitoring. */
 
-void write_stats_file(afl_state_t *afl, double bitmap_cvg, double stability,
-                      double eps) {
+void write_stats_file(afl_state_t *afl, u32 t_bytes, double bitmap_cvg,
+                      double stability, double eps) {
 
 #ifndef __HAIKU__
   struct rusage rus;
 #endif
 
   u64   cur_time = get_cur_time();
-  u32   t_bytes = count_non_255_bytes(afl, afl->virgin_bits);
   u8    fn[PATH_MAX];
   FILE *f;
 
@@ -353,9 +352,11 @@ void write_stats_file(afl_state_t *afl, double bitmap_cvg, double stability,
 
 /* Update the plot file if there is a reason to. */
 
-void maybe_update_plot_file(afl_state_t *afl, double bitmap_cvg, double eps) {
+void maybe_update_plot_file(afl_state_t *afl, u32 t_bytes, double bitmap_cvg,
+                            double eps) {
 
-  if (unlikely(afl->plot_prev_qp == afl->queued_paths &&
+  if (unlikely(afl->stop_soon) ||
+      unlikely(afl->plot_prev_qp == afl->queued_paths &&
                afl->plot_prev_pf == afl->pending_favored &&
                afl->plot_prev_pnf == afl->pending_not_fuzzed &&
                afl->plot_prev_ce == afl->current_entry &&
@@ -384,16 +385,16 @@ void maybe_update_plot_file(afl_state_t *afl, double bitmap_cvg, double eps) {
   /* Fields in the file:
 
      unix_time, afl->cycles_done, cur_path, paths_total, paths_not_fuzzed,
-     favored_not_fuzzed, afl->unique_crashes, afl->unique_hangs, afl->max_depth,
-     execs_per_sec */
+     favored_not_fuzzed, unique_crashes, unique_hangs, max_depth,
+     execs_per_sec, edges_found */
 
-  fprintf(
-      afl->fsrv.plot_file,
-      "%llu, %llu, %u, %u, %u, %u, %0.02f%%, %llu, %llu, %u, %0.02f, %llu\n",
-      get_cur_time() / 1000, afl->queue_cycle - 1, afl->current_entry,
-      afl->queued_paths, afl->pending_not_fuzzed, afl->pending_favored,
-      bitmap_cvg, afl->unique_crashes, afl->unique_hangs, afl->max_depth, eps,
-      afl->plot_prev_ed);                                  /* ignore errors */
+  fprintf(afl->fsrv.plot_file,
+          "%llu, %llu, %u, %u, %u, %u, %0.02f%%, %llu, %llu, %u, %0.02f, %llu, "
+          "%u\n",
+          get_cur_time() / 1000, afl->queue_cycle - 1, afl->current_entry,
+          afl->queued_paths, afl->pending_not_fuzzed, afl->pending_favored,
+          bitmap_cvg, afl->unique_crashes, afl->unique_hangs, afl->max_depth,
+          eps, afl->plot_prev_ed, t_bytes);                /* ignore errors */
 
   fflush(afl->fsrv.plot_file);
 
@@ -532,7 +533,8 @@ void show_stats(afl_state_t *afl) {
   if (cur_ms - afl->stats_last_stats_ms > STATS_UPDATE_SEC * 1000) {
 
     afl->stats_last_stats_ms = cur_ms;
-    write_stats_file(afl, t_byte_ratio, stab_ratio, afl->stats_avg_exec);
+    write_stats_file(afl, t_bytes, t_byte_ratio, stab_ratio,
+                     afl->stats_avg_exec);
     save_auto(afl);
     write_bitmap(afl);
 
@@ -555,7 +557,7 @@ void show_stats(afl_state_t *afl) {
   if (cur_ms - afl->stats_last_plot_ms > PLOT_UPDATE_SEC * 1000) {
 
     afl->stats_last_plot_ms = cur_ms;
-    maybe_update_plot_file(afl, t_byte_ratio, afl->stats_avg_exec);
+    maybe_update_plot_file(afl, t_bytes, t_byte_ratio, afl->stats_avg_exec);
 
   }
 
@@ -643,6 +645,13 @@ void show_stats(afl_state_t *afl) {
 #define SP10 SP5 SP5
 #define SP20 SP10 SP10
 
+  /* Since `total_crashes` does not get reloaded from disk on restart,
+    it indicates if we found crashes this round already -> paint red.
+    If it's 0, but `unique_crashes` is set from a past run, paint in yellow. */
+  char *crash_color = afl->total_crashes    ? cLRD
+                      : afl->unique_crashes ? cYEL
+                                            : cRST;
+
   /* Lord, forgive me this. */
 
   SAYF(SET_G1 bSTG bLT bH bSTOP                         cCYA
@@ -730,7 +739,7 @@ void show_stats(afl_state_t *afl) {
   u_stringify_time_diff(time_tmp, cur_ms, afl->last_crash_time);
   SAYF(bV bSTOP " last uniq crash : " cRST "%-33s " bSTG bV bSTOP
                 " uniq crashes : %s%-6s" bSTG               bV "\n",
-       time_tmp, afl->unique_crashes ? cLRD : cRST, tmp);
+       time_tmp, crash_color, tmp);
 
   sprintf(tmp, "%s%s", u_stringify_int(IB(0), afl->unique_hangs),
           (afl->unique_hangs >= KEEP_UNIQUE_HANG) ? "+" : "");
@@ -813,15 +822,13 @@ void show_stats(afl_state_t *afl) {
 
     SAYF(bV bSTOP " total execs : " cRST "%-20s " bSTG bV bSTOP
                   "   new crashes : %s%-22s" bSTG         bV "\n",
-         u_stringify_int(IB(0), afl->fsrv.total_execs),
-         afl->unique_crashes ? cLRD : cRST, tmp);
+         u_stringify_int(IB(0), afl->fsrv.total_execs), crash_color, tmp);
 
   } else {
 
     SAYF(bV bSTOP " total execs : " cRST "%-20s " bSTG bV bSTOP
                   " total crashes : %s%-22s" bSTG         bV "\n",
-         u_stringify_int(IB(0), afl->fsrv.total_execs),
-         afl->unique_crashes ? cLRD : cRST, tmp);
+         u_stringify_int(IB(0), afl->fsrv.total_execs), crash_color, tmp);
 
   }
 
@@ -1217,7 +1224,7 @@ void show_init_stats(afl_state_t *afl) {
       stringify_int(IB(0), min_us), stringify_int(IB(1), max_us),
       stringify_int(IB(2), avg_us));
 
-  if (!afl->timeout_given) {
+  if (afl->timeout_given != 1) {
 
     /* Figure out the appropriate timeout. The basic idea is: 5x average or
        1x max, rounded up to EXEC_TM_ROUND ms and capped at 1 second.

src/afl-fuzz.c

@@ -103,7 +103,10 @@ static void usage(u8 *argv0, int more_help) {
       "                  quad -- see docs/power_schedules.md\n"
       "  -f file       - location read by the fuzzed program (default: stdin "
       "or @@)\n"
-      "  -t msec       - timeout for each run (auto-scaled, 50-%u ms)\n"
+      "  -t msec       - timeout for each run (auto-scaled, default %u ms). "
+      "Add a '+'\n"
+      "                  to auto-calculate the timeout, the value being the "
+      "maximum.\n"
       "  -m megs       - memory limit for child process (%u MB, 0 = no limit "
       "[default])\n"
       "  -Q            - use binary-only instrumentation (QEMU mode)\n"
@@ -122,10 +125,10 @@ static void usage(u8 *argv0, int more_help) {
       "  -c program    - enable CmpLog by specifying a binary compiled for "
       "it.\n"
       "                  if using QEMU, just use -c 0.\n"
-      "  -l cmplog_level - set the complexity/intensivity of CmpLog.\n"
-      "                  Values: 1 (basic), 2 (larger files) and 3 "
-      "(transform)\n\n"
-
+      "  -l cmplog_opts - CmpLog configuration values (e.g. \"2AT\"):\n"
+      "                  1=small files (default), 2=larger files, 3=all "
+      "files,\n"
+      "                  A=arithmetic solving, T=transformational solving.\n\n"
       "Fuzzing behavior settings:\n"
       "  -Z            - sequential queue selection instead of weighted "
       "random\n"
@@ -137,8 +140,8 @@ static void usage(u8 *argv0, int more_help) {
 
       "Testing settings:\n"
       "  -s seed       - use a fixed seed for the RNG\n"
-      "  -V seconds    - fuzz for a specific time then terminate\n"
-      "  -E execs      - fuzz for a approx. no of total executions then "
+      "  -V seconds    - fuzz for a specified time then terminate\n"
+      "  -E execs      - fuzz for an approx. no. of total executions then "
       "terminate\n"
       "                  Note: not precise and can have several more "
       "executions.\n\n"
@@ -198,6 +201,7 @@ static void usage(u8 *argv0, int more_help) {
       "AFL_FORKSRV_INIT_TMOUT: time spent waiting for forkserver during startup (in milliseconds)\n"
       "AFL_HANG_TMOUT: override timeout value (in milliseconds)\n"
       "AFL_I_DONT_CARE_ABOUT_MISSING_CRASHES: don't warn about core dump handlers\n"
+      "AFL_IGNORE_UNKNOWN_ENVS: don't warn on unknown env vars\n"
       "AFL_IMPORT_FIRST: sync and import test cases from other fuzzer instances first\n"
       "AFL_KILL_SIGNAL: Signal ID delivered to child processes on timeout, etc. (default: SIGKILL)\n"
       "AFL_MAP_SIZE: the shared memory size for that target. must be >= the size\n"
@@ -347,7 +351,7 @@ int main(int argc, char **argv_orig, char **envp) {
   exit_1 = !!afl->afl_env.afl_bench_just_one;
 
   SAYF(cCYA "afl-fuzz" VERSION cRST
-            " based on afl by Michal Zalewski and a big online community\n");
+            " based on afl by Michal Zalewski and a large online community\n");
 
   doc_path = access(DOC_PATH, F_OK) != 0 ? (u8 *)"docs" : (u8 *)DOC_PATH;
 
@@ -552,13 +556,22 @@ int main(int argc, char **argv_orig, char **envp) {
 
       case 'F':                                         /* foreign sync dir */
 
-        if (!afl->is_main_node)
+        if (!optarg) { FATAL("Missing path for -F"); }
+        if (!afl->is_main_node) {
+
           FATAL(
               "Option -F can only be specified after the -M option for the "
               "main fuzzer of a fuzzing campaign");
-        if (afl->foreign_sync_cnt >= FOREIGN_SYNCS_MAX)
+
+        }
+
+        if (afl->foreign_sync_cnt >= FOREIGN_SYNCS_MAX) {
+
           FATAL("Maximum %u entried of -F option can be specified",
                 FOREIGN_SYNCS_MAX);
+
+        }
+
         afl->foreign_syncs[afl->foreign_sync_cnt].dir = optarg;
         while (afl->foreign_syncs[afl->foreign_sync_cnt]
                    .dir[strlen(afl->foreign_syncs[afl->foreign_sync_cnt].dir) -
@@ -802,13 +815,36 @@ int main(int argc, char **argv_orig, char **envp) {
 
       case 'l': {
 
-        afl->cmplog_lvl = atoi(optarg);
-        if (afl->cmplog_lvl < 1 || afl->cmplog_lvl > CMPLOG_LVL_MAX) {
+        if (!optarg) { FATAL("missing parameter for 'l'"); }
+        char *c = optarg;
+        while (*c) {
 
-          FATAL(
-              "Bad complog level value, accepted values are 1 (default), 2 and "
-              "%u.",
-              CMPLOG_LVL_MAX);
+          switch (*c) {
+
+            case '0':
+            case '1':
+              afl->cmplog_lvl = 1;
+              break;
+            case '2':
+              afl->cmplog_lvl = 2;
+              break;
+            case '3':
+              afl->cmplog_lvl = 3;
+              break;
+            case 'a':
+            case 'A':
+              afl->cmplog_enable_arith = 1;
+              break;
+            case 't':
+            case 'T':
+              afl->cmplog_enable_transform = 1;
+              break;
+            default:
+              FATAL("Unknown option value '%c' in -l %s", *c, optarg);
+
+          }
+
+          ++c;
 
         }
 
@@ -1367,6 +1403,15 @@ int main(int argc, char **argv_orig, char **envp) {
   set_scheduler_mode(SCHEDULER_MODE_LOW_LATENCY);
   #endif
 
+  #ifdef __APPLE__
+  if (pthread_set_qos_class_self_np(QOS_CLASS_USER_INTERACTIVE, 0) != 0) {
+
+    WARNF("general thread priority settings failed");
+
+  }
+
+  #endif
+
   init_count_class16();
 
   if (afl->is_main_node && check_main_node_exists(afl) == 1) {
@@ -1401,24 +1446,9 @@ int main(int argc, char **argv_orig, char **envp) {
   // read_foreign_testcases(afl, 1); for the moment dont do this
   OKF("Loaded a total of %u seeds.", afl->queued_paths);
 
-  load_auto(afl);
-
   pivot_inputs(afl);
 
-  if (extras_dir_cnt) {
-
-    for (i = 0; i < extras_dir_cnt; i++) {
-
-      load_extras(afl, extras_dir[i]);
-
-    }
-
-    dedup_extras(afl);
-    OKF("Loaded a total of %u extras.", afl->extras_cnt);
-
-  }
-
-  if (!afl->timeout_given) { find_timeout(afl); }
+  if (!afl->timeout_given) { find_timeout(afl); }  // only for resumes!
 
   if ((afl->tmp_dir = afl->afl_env.afl_tmpdir) != NULL &&
       !afl->in_place_resume) {
@@ -1532,6 +1562,21 @@ int main(int argc, char **argv_orig, char **envp) {
 
   }
 
+  if (afl->non_instrumented_mode || afl->fsrv.qemu_mode || afl->unicorn_mode) {
+
+    map_size = afl->fsrv.map_size = MAP_SIZE;
+    afl->virgin_bits = ck_realloc(afl->virgin_bits, map_size);
+    afl->virgin_tmout = ck_realloc(afl->virgin_tmout, map_size);
+    afl->virgin_crash = ck_realloc(afl->virgin_crash, map_size);
+    afl->var_bytes = ck_realloc(afl->var_bytes, map_size);
+    afl->top_rated = ck_realloc(afl->top_rated, map_size * sizeof(void *));
+    afl->clean_trace = ck_realloc(afl->clean_trace, map_size);
+    afl->clean_trace_custom = ck_realloc(afl->clean_trace_custom, map_size);
+    afl->first_trace = ck_realloc(afl->first_trace, map_size);
+    afl->map_tmp_buf = ck_realloc(afl->map_tmp_buf, map_size);
+
+  }
+
   afl->argv = use_argv;
   afl->fsrv.trace_bits =
       afl_shm_init(&afl->shm, afl->fsrv.map_size, afl->non_instrumented_mode);
@@ -1539,42 +1584,44 @@ int main(int argc, char **argv_orig, char **envp) {
   if (!afl->non_instrumented_mode && !afl->fsrv.qemu_mode &&
       !afl->unicorn_mode) {
 
-    afl->fsrv.map_size = 4194304;  // dummy temporary value
-    setenv("AFL_MAP_SIZE", "4194304", 1);
+    if (map_size <= 8000000 && !afl->non_instrumented_mode &&
+        !afl->fsrv.qemu_mode && !afl->unicorn_mode) {
+
+      afl->fsrv.map_size = 8000000;  // dummy temporary value
+      setenv("AFL_MAP_SIZE", "8000000", 1);
+
+    }
 
     u32 new_map_size = afl_fsrv_get_mapsize(
         &afl->fsrv, afl->argv, &afl->stop_soon, afl->afl_env.afl_debug_child);
 
-    if (new_map_size && new_map_size != 4194304) {
+    // only reinitialize when it makes sense
+    if ((map_size < new_map_size ||
+         (new_map_size != MAP_SIZE && new_map_size < map_size &&
+          map_size - new_map_size > MAP_SIZE))) {
 
-      // only reinitialize when it makes sense
-      if (map_size < new_map_size ||
-          (new_map_size > map_size && new_map_size - map_size > MAP_SIZE)) {
+      OKF("Re-initializing maps to %u bytes", new_map_size);
 
-        OKF("Re-initializing maps to %u bytes", new_map_size);
+      afl->virgin_bits = ck_realloc(afl->virgin_bits, new_map_size);
+      afl->virgin_tmout = ck_realloc(afl->virgin_tmout, new_map_size);
+      afl->virgin_crash = ck_realloc(afl->virgin_crash, new_map_size);
+      afl->var_bytes = ck_realloc(afl->var_bytes, new_map_size);
+      afl->top_rated =
+          ck_realloc(afl->top_rated, new_map_size * sizeof(void *));
+      afl->clean_trace = ck_realloc(afl->clean_trace, new_map_size);
+      afl->clean_trace_custom =
+          ck_realloc(afl->clean_trace_custom, new_map_size);
+      afl->first_trace = ck_realloc(afl->first_trace, new_map_size);
+      afl->map_tmp_buf = ck_realloc(afl->map_tmp_buf, new_map_size);
 
-        afl->virgin_bits = ck_realloc(afl->virgin_bits, new_map_size);
-        afl->virgin_tmout = ck_realloc(afl->virgin_tmout, new_map_size);
-        afl->virgin_crash = ck_realloc(afl->virgin_crash, new_map_size);
-        afl->var_bytes = ck_realloc(afl->var_bytes, new_map_size);
-        afl->top_rated =
-            ck_realloc(afl->top_rated, new_map_size * sizeof(void *));
-        afl->clean_trace = ck_realloc(afl->clean_trace, new_map_size);
-        afl->clean_trace_custom =
-            ck_realloc(afl->clean_trace_custom, new_map_size);
-        afl->first_trace = ck_realloc(afl->first_trace, new_map_size);
-        afl->map_tmp_buf = ck_realloc(afl->map_tmp_buf, new_map_size);
-
-        afl_fsrv_kill(&afl->fsrv);
-        afl_shm_deinit(&afl->shm);
-        afl->fsrv.map_size = new_map_size;
-        afl->fsrv.trace_bits =
-            afl_shm_init(&afl->shm, new_map_size, afl->non_instrumented_mode);
-        setenv("AFL_NO_AUTODICT", "1", 1);  // loaded already
-        afl_fsrv_start(&afl->fsrv, afl->argv, &afl->stop_soon,
-                       afl->afl_env.afl_debug_child);
-
-      }
+      afl_fsrv_kill(&afl->fsrv);
+      afl_shm_deinit(&afl->shm);
+      afl->fsrv.map_size = new_map_size;
+      afl->fsrv.trace_bits =
+          afl_shm_init(&afl->shm, new_map_size, afl->non_instrumented_mode);
+      setenv("AFL_NO_AUTODICT", "1", 1);  // loaded already
+      afl_fsrv_start(&afl->fsrv, afl->argv, &afl->stop_soon,
+                     afl->afl_env.afl_debug_child);
 
       map_size = new_map_size;
 
@@ -1594,57 +1641,77 @@ int main(int argc, char **argv_orig, char **envp) {
     afl->cmplog_fsrv.cmplog_binary = afl->cmplog_binary;
     afl->cmplog_fsrv.init_child_func = cmplog_exec_child;
 
-    afl->cmplog_fsrv.map_size = 4194304;
+    if (map_size <= 8000000 && !afl->non_instrumented_mode &&
+        !afl->fsrv.qemu_mode && !afl->unicorn_mode) {
+
+      afl->cmplog_fsrv.map_size = 8000000;  // dummy temporary value
+      setenv("AFL_MAP_SIZE", "8000000", 1);
+
+    }
 
     u32 new_map_size =
         afl_fsrv_get_mapsize(&afl->cmplog_fsrv, afl->argv, &afl->stop_soon,
                              afl->afl_env.afl_debug_child);
 
-    if (new_map_size && new_map_size != 4194304) {
+    // only reinitialize when it needs to be larger
+    if (map_size < new_map_size) {
 
-      // only reinitialize when it needs to be larger
-      if (map_size < new_map_size) {
+      OKF("Re-initializing maps to %u bytes due cmplog", new_map_size);
 
-        OKF("Re-initializing maps to %u bytes due cmplog", new_map_size);
+      afl->virgin_bits = ck_realloc(afl->virgin_bits, new_map_size);
+      afl->virgin_tmout = ck_realloc(afl->virgin_tmout, new_map_size);
+      afl->virgin_crash = ck_realloc(afl->virgin_crash, new_map_size);
+      afl->var_bytes = ck_realloc(afl->var_bytes, new_map_size);
+      afl->top_rated =
+          ck_realloc(afl->top_rated, new_map_size * sizeof(void *));
+      afl->clean_trace = ck_realloc(afl->clean_trace, new_map_size);
+      afl->clean_trace_custom =
+          ck_realloc(afl->clean_trace_custom, new_map_size);
+      afl->first_trace = ck_realloc(afl->first_trace, new_map_size);
+      afl->map_tmp_buf = ck_realloc(afl->map_tmp_buf, new_map_size);
 
-        afl->virgin_bits = ck_realloc(afl->virgin_bits, new_map_size);
-        afl->virgin_tmout = ck_realloc(afl->virgin_tmout, new_map_size);
-        afl->virgin_crash = ck_realloc(afl->virgin_crash, new_map_size);
-        afl->var_bytes = ck_realloc(afl->var_bytes, new_map_size);
-        afl->top_rated =
-            ck_realloc(afl->top_rated, new_map_size * sizeof(void *));
-        afl->clean_trace = ck_realloc(afl->clean_trace, new_map_size);
-        afl->clean_trace_custom =
-            ck_realloc(afl->clean_trace_custom, new_map_size);
-        afl->first_trace = ck_realloc(afl->first_trace, new_map_size);
-        afl->map_tmp_buf = ck_realloc(afl->map_tmp_buf, new_map_size);
+      afl_fsrv_kill(&afl->fsrv);
+      afl_fsrv_kill(&afl->cmplog_fsrv);
+      afl_shm_deinit(&afl->shm);
 
-        afl_fsrv_kill(&afl->fsrv);
-        afl_fsrv_kill(&afl->cmplog_fsrv);
-        afl_shm_deinit(&afl->shm);
-        afl->cmplog_fsrv.map_size = new_map_size;  // non-cmplog stays the same
+      afl->cmplog_fsrv.map_size = new_map_size;  // non-cmplog stays the same
+      map_size = new_map_size;
 
-        afl->fsrv.trace_bits =
-            afl_shm_init(&afl->shm, new_map_size, afl->non_instrumented_mode);
-        setenv("AFL_NO_AUTODICT", "1", 1);  // loaded already
-        afl_fsrv_start(&afl->fsrv, afl->argv, &afl->stop_soon,
-                       afl->afl_env.afl_debug_child);
+      setenv("AFL_NO_AUTODICT", "1", 1);  // loaded already
+      afl->fsrv.trace_bits =
+          afl_shm_init(&afl->shm, new_map_size, afl->non_instrumented_mode);
+      afl->cmplog_fsrv.trace_bits = afl->fsrv.trace_bits;
+      afl_fsrv_start(&afl->fsrv, afl->argv, &afl->stop_soon,
+                     afl->afl_env.afl_debug_child);
+      afl_fsrv_start(&afl->cmplog_fsrv, afl->argv, &afl->stop_soon,
+                     afl->afl_env.afl_debug_child);
 
-        afl_fsrv_start(&afl->cmplog_fsrv, afl->argv, &afl->stop_soon,
-                       afl->afl_env.afl_debug_child);
+    } else {
 
-        map_size = new_map_size;
-
-      }
+      afl->cmplog_fsrv.map_size = new_map_size;
 
     }
 
-    afl->cmplog_fsrv.map_size = map_size;
-
     OKF("Cmplog forkserver successfully started");
 
   }
 
+  load_auto(afl);
+
+  if (extras_dir_cnt) {
+
+    for (i = 0; i < extras_dir_cnt; i++) {
+
+      load_extras(afl, extras_dir[i]);
+
+    }
+
+  }
+
+  deunicode_extras(afl);
+  dedup_extras(afl);
+  if (afl->extras_cnt) { OKF("Loaded a total of %u extras.", afl->extras_cnt); }
+
   // after we have the correct bitmap size we can read the bitmap -B option
   // and set the virgin maps
   if (afl->in_bitmap) {
@@ -1672,20 +1739,49 @@ int main(int argc, char **argv_orig, char **envp) {
 
   cull_queue(afl);
 
-  if (!afl->pending_not_fuzzed) {
+  // ensure we have at least one seed that is not disabled.
+  u32 entry, valid_seeds = 0;
+  for (entry = 0; entry < afl->queued_paths; ++entry)
+    if (!afl->queue_buf[entry]->disabled) { ++valid_seeds; }
+
+  if (!afl->pending_not_fuzzed || !valid_seeds) {
 
     FATAL("We need at least one valid input seed that does not crash!");
 
   }
 
+  if (afl->timeout_given == 2) {  // -t ...+ option
+
+    if (valid_seeds == 1) {
+
+      WARNF(
+          "Only one valid seed is present, auto-calculating the timeout is "
+          "disabled!");
+      afl->timeout_given = 1;
+
+    } else {
+
+      u64 max_ms = 0;
+
+      for (entry = 0; entry < afl->queued_paths; ++entry)
+        if (!afl->queue_buf[entry]->disabled)
+          if (afl->queue_buf[entry]->exec_us > max_ms)
+            max_ms = afl->queue_buf[entry]->exec_us;
+
+      afl->fsrv.exec_tmout = max_ms;
+
+    }
+
+  }
+
   show_init_stats(afl);
 
   if (unlikely(afl->old_seed_selection)) seek_to = find_start_position(afl);
 
   afl->start_time = get_cur_time();
   if (afl->in_place_resume || afl->afl_env.afl_autoresume) load_stats_file(afl);
-  write_stats_file(afl, 0, 0, 0);
-  maybe_update_plot_file(afl, 0, 0);
+  write_stats_file(afl, 0, 0, 0, 0);
+  maybe_update_plot_file(afl, 0, 0, 0);
   save_auto(afl);
 
   if (afl->stop_soon) { goto stop_fuzzing; }
@@ -1735,12 +1831,15 @@ int main(int argc, char **argv_orig, char **envp) {
       if (unlikely(afl->old_seed_selection)) {
 
         afl->current_entry = 0;
-        while (unlikely(afl->queue_buf[afl->current_entry]->disabled)) {
+        while (unlikely(afl->current_entry < afl->queued_paths &&
+                        afl->queue_buf[afl->current_entry]->disabled)) {
 
           ++afl->current_entry;
 
         }
 
+        if (afl->current_entry >= afl->queued_paths) { afl->current_entry = 0; }
+
         afl->queue_cur = afl->queue_buf[afl->current_entry];
 
         if (unlikely(seek_to)) {
@@ -1943,15 +2042,24 @@ int main(int argc, char **argv_orig, char **envp) {
 
         if (unlikely(afl->is_main_node)) {
 
-          if (!(sync_interval_cnt++ % (SYNC_INTERVAL / 3))) {
+          if (unlikely(get_cur_time() >
+                       (SYNC_TIME >> 1) + afl->last_sync_time)) {
 
-            sync_fuzzers(afl);
+            if (!(sync_interval_cnt++ % (SYNC_INTERVAL / 3))) {
+
+              sync_fuzzers(afl);
+
+            }
 
           }
 
         } else {
 
-          if (!(sync_interval_cnt++ % SYNC_INTERVAL)) { sync_fuzzers(afl); }
+          if (unlikely(get_cur_time() > SYNC_TIME + afl->last_sync_time)) {
+
+            if (!(sync_interval_cnt++ % SYNC_INTERVAL)) { sync_fuzzers(afl); }
+
+          }
 
         }
 
@@ -1966,12 +2074,12 @@ int main(int argc, char **argv_orig, char **envp) {
   }
 
   write_bitmap(afl);
-  maybe_update_plot_file(afl, 0, 0);
+  maybe_update_plot_file(afl, 0, 0, 0);
   save_auto(afl);
 
 stop_fuzzing:
 
-  write_stats_file(afl, 0, 0, 0);
+  write_stats_file(afl, 0, 0, 0, 0);
   afl->force_ui_update = 1;  // ensure the screen is reprinted
   show_stats(afl);           // print the screen one last time
 

src/afl-showmap.c

@@ -563,6 +563,7 @@ static void set_up_environment(afl_forkserver_t *fsrv) {
          "detect_leaks=0:"
          "allocator_may_return_null=1:"
          "symbolize=0:"
+         "detect_odr_violation=0:"
          "handle_segv=0:"
          "handle_sigbus=0:"
          "handle_abort=0:"
@@ -1013,7 +1014,6 @@ int main(int argc, char **argv_orig, char **envp) {
 
   if (in_dir) {
 
-    if (at_file) { PFATAL("Options -A and -i are mutually exclusive"); }
     detect_file_args(argv + optind, "", &fsrv->use_stdin);
 
   } else {
@@ -1169,8 +1169,9 @@ int main(int argc, char **argv_orig, char **envp) {
 
     }
 
-    stdin_file =
-        alloc_printf("%s/.afl-showmap-temp-%u", use_dir, (u32)getpid());
+    stdin_file = at_file ? strdup(at_file)
+                         : (char *)alloc_printf("%s/.afl-showmap-temp-%u",
+                                                use_dir, (u32)getpid());
     unlink(stdin_file);
     atexit(at_exit_handler);
     fsrv->out_file = stdin_file;

src/afl-tmin.c

@@ -717,6 +717,7 @@ static void set_up_environment(afl_forkserver_t *fsrv) {
          "detect_leaks=0:"
          "allocator_may_return_null=1:"
          "symbolize=0:"
+         "detect_odr_violation=0:"
          "handle_segv=0:"
          "handle_sigbus=0:"
          "handle_abort=0:"

test/test-basic.sh

@@ -7,7 +7,7 @@ AFL_GCC=afl-gcc
 $ECHO "$BLUE[*] Testing: ${AFL_GCC}, afl-showmap, afl-fuzz, afl-cmin and afl-tmin"
 test "$SYS" = "i686" -o "$SYS" = "x86_64" -o "$SYS" = "amd64" -o "$SYS" = "i86pc" -o "$SYS" = "i386" && {
  test -e ../${AFL_GCC} -a -e ../afl-showmap -a -e ../afl-fuzz && {
-  ../${AFL_GCC} -o test-instr.plain ../test-instr.c > /dev/null 2>&1
+  ../${AFL_GCC} -o test-instr.plain -O0 ../test-instr.c > /dev/null 2>&1
   AFL_HARDEN=1 ../${AFL_GCC} -o test-compcov.harden test-compcov.c > /dev/null 2>&1
   test -e test-instr.plain && {
     $ECHO "$GREEN[+] ${AFL_GCC} compilation succeeded"
@@ -39,7 +39,7 @@ test "$SYS" = "i686" -o "$SYS" = "x86_64" -o "$SYS" = "amd64" -o "$SYS" = "i86pc
     $ECHO "$RED[!] ${AFL_GCC} failed"
     echo CUT------------------------------------------------------------------CUT
     uname -a
-    ../${AFL_GCC} -o test-instr.plain ../test-instr.c
+    ../${AFL_GCC} -o test-instr.plain -O0 ../test-instr.c
     echo CUT------------------------------------------------------------------CUT
     CODE=1
   }
@@ -128,7 +128,7 @@ test "$SYS" = "i686" -o "$SYS" = "x86_64" -o "$SYS" = "amd64" -o "$SYS" = "i86pc
  $ECHO "$BLUE[*] Testing: ${AFL_GCC}, afl-showmap, afl-fuzz, afl-cmin and afl-tmin"
  SKIP=
  test -e ../${AFL_GCC} -a -e ../afl-showmap -a -e ../afl-fuzz && {
-  ../${AFL_GCC} -o test-instr.plain ../test-instr.c > /dev/null 2>&1
+  ../${AFL_GCC} -o test-instr.plain -O0 ../test-instr.c > /dev/null 2>&1
   AFL_HARDEN=1 ../${AFL_GCC} -o test-compcov.harden test-compcov.c > /dev/null 2>&1
   test -e test-instr.plain && {
     $ECHO "$GREEN[+] ${AFL_GCC} compilation succeeded"

test/test-custom-mutators.sh

@@ -37,9 +37,9 @@ test "1" = "`../afl-fuzz | grep -i 'without python' >/dev/null; echo $?`" && {
       echo "00000" > in/in
 
       # Run afl-fuzz w/ the C mutator
-      $ECHO "$GREY[*] running afl-fuzz for the C mutator, this will take approx 5 seconds"
+      $ECHO "$GREY[*] running afl-fuzz for the C mutator, this will take approx 10 seconds"
       {
-        AFL_CUSTOM_MUTATOR_LIBRARY=./libexamplemutator.so AFL_CUSTOM_MUTATOR_ONLY=1 ../afl-fuzz -V1 -m ${MEM_LIMIT} -i in -o out -- ./test-custom-mutator >>errors 2>&1
+        AFL_CUSTOM_MUTATOR_LIBRARY=./libexamplemutator.so AFL_CUSTOM_MUTATOR_ONLY=1 ../afl-fuzz -V10 -m ${MEM_LIMIT} -i in -o out -- ./test-custom-mutator >>errors 2>&1
       } >>errors 2>&1
 
       # Check results
@@ -57,9 +57,9 @@ test "1" = "`../afl-fuzz | grep -i 'without python' >/dev/null; echo $?`" && {
       rm -rf out errors core.*
 
       # Run afl-fuzz w/ multiple C mutators
-      $ECHO "$GREY[*] running afl-fuzz with multiple custom C mutators, this will take approx 5 seconds"
+      $ECHO "$GREY[*] running afl-fuzz with multiple custom C mutators, this will take approx 10 seconds"
       {
-        AFL_CUSTOM_MUTATOR_LIBRARY="./libexamplemutator.so;./libexamplemutator2.so" AFL_CUSTOM_MUTATOR_ONLY=1 ../afl-fuzz -V1 -m ${MEM_LIMIT} -i in -o out -- ./test-multiple-mutators >>errors 2>&1
+        AFL_CUSTOM_MUTATOR_LIBRARY="./libexamplemutator.so;./libexamplemutator2.so" AFL_CUSTOM_MUTATOR_ONLY=1 ../afl-fuzz -V10 -m ${MEM_LIMIT} -i in -o out -- ./test-multiple-mutators >>errors 2>&1
       } >>errors 2>&1
 
       test -n "$( ls out/default/crashes/id:000000* 2>/dev/null )" && {  # TODO: update here
@@ -76,11 +76,11 @@ test "1" = "`../afl-fuzz | grep -i 'without python' >/dev/null; echo $?`" && {
       rm -rf out errors core.*
 
       # Run afl-fuzz w/ the Python mutator
-      $ECHO "$GREY[*] running afl-fuzz for the Python mutator, this will take approx 5 seconds"
+      $ECHO "$GREY[*] running afl-fuzz for the Python mutator, this will take approx 10 seconds"
       {
         export PYTHONPATH=${CUSTOM_MUTATOR_PATH}
         export AFL_PYTHON_MODULE=example
-        AFL_CUSTOM_MUTATOR_ONLY=1 ../afl-fuzz -V5 -m ${MEM_LIMIT} -i in -o out -- ./test-custom-mutator >>errors 2>&1
+        AFL_CUSTOM_MUTATOR_ONLY=1 ../afl-fuzz -V10 -m ${MEM_LIMIT} -i in -o out -- ./test-custom-mutator >>errors 2>&1
         unset PYTHONPATH
         unset AFL_PYTHON_MODULE
       } >>errors 2>&1

test/test-llvm.sh

@@ -162,9 +162,9 @@ test -e ../afl-clang-fast -a -e ../split-switches-pass.so && {
   test -e test-floatingpoint && {
     mkdir -p in
     echo ZZZZ > in/in
-    $ECHO "$GREY[*] running afl-fuzz with floating point splitting, this will take max. 30 seconds"
+    $ECHO "$GREY[*] running afl-fuzz with floating point splitting, this will take max. 45 seconds"
     {
-      AFL_BENCH_UNTIL_CRASH=1 AFL_NO_UI=1 ../afl-fuzz -s 1 -V30 -m ${MEM_LIMIT} -i in -o out -D -- ./test-floatingpoint >>errors 2>&1
+      AFL_BENCH_UNTIL_CRASH=1 AFL_NO_UI=1 ../afl-fuzz -Z -s 123 -V50 -m ${MEM_LIMIT} -i in -o out -D -- ./test-floatingpoint >>errors 2>&1
     } >>errors 2>&1
     test -n "$( ls out/default/crashes/id:* 2>/dev/null )" && {
       $ECHO "$GREEN[+] llvm_mode laf-intel floatingpoint splitting feature works correctly"

test/test-qemu-mode.sh

@@ -39,14 +39,7 @@ test -e ../afl-qemu-trace && {
       $ECHO "$GREY[*] running afl-fuzz for qemu_mode AFL_ENTRYPOINT, this will take approx 6 seconds"
       {
         {
-          if file test-instr | grep -q "32-bit"; then
-            # for 32-bit reduce 8 nibbles to the lower 7 nibbles
-	    ADDR_LOWER_PART=`nm test-instr | grep "T main" | awk '{print $1}' | sed 's/^.//'`
-          else
-            # for 64-bit reduce 16 nibbles to the lower 9 nibbles
-	    ADDR_LOWER_PART=`nm test-instr | grep "T main" | awk '{print $1}' | sed 's/^.......//'`
-          fi
-          export AFL_ENTRYPOINT=`expr 0x4${ADDR_LOWER_PART}`
+          export AFL_ENTRYPOINT=`printf 1 | AFL_DEBUG=1 ../afl-qemu-trace ./test-instr 2>&1 >/dev/null | awk '/forkserver/{print $4; exit}'`
           $ECHO AFL_ENTRYPOINT=$AFL_ENTRYPOINT - $(nm test-instr | grep "T main") - $(file ./test-instr)
           ../afl-fuzz -m ${MEM_LIMIT} -V2 -Q -i in -o out -- ./test-instr
           unset AFL_ENTRYPOINT

test/test-unicorn-mode.sh

@@ -14,7 +14,7 @@ test -d ../unicorn_mode/unicornafl -a -e ../unicorn_mode/unicornafl/samples/shel
       EASY_INSTALL_FOUND=0
       for PYTHON in $PYTHONS ; do
 
-        if $PYTHON -c "help('easy_install');" </dev/null | grep -q module ; then
+        if $PYTHON -c "import setuptools" ; then
 
             EASY_INSTALL_FOUND=1
             PY=$PYTHON

unicorn_mode/README.md

@@ -8,19 +8,19 @@ The CompareCoverage and NeverZero counters features are by Andrea Fioraldi <andr
 
 ## 1) Introduction
 
-The code in ./unicorn_mode allows you to build a standalone feature that
-leverages the Unicorn Engine and allows callers to obtain instrumentation 
+The code in ./unicorn_mode allows you to build the (Unicorn Engine)[https://github.com/unicorn-engine/unicorn] with afl support.
+This means, you can run anything that can be emulated in unicorn and obtain instrumentation
 output for black-box, closed-source binary code snippets. This mechanism 
 can be then used by afl-fuzz to stress-test targets that couldn't be built 
-with afl-gcc or used in QEMU mode, or with other extensions such as 
-TriforceAFL.
+with afl-cc or used in QEMU mode.
 
 There is a significant performance penalty compared to native AFL,
 but at least we're able to use AFL++ on these binaries, right?
 
 ## 2) How to use
 
-Requirements: you need an installed python environment.
+First, you will need a working harness for your target in unicorn, using Python, C, or Rust.
+For some pointers for more advanced emulation, take a look at [BaseSAFE](https://github.com/fgsect/BaseSAFE) and [Qiling](https://github.com/qilingframework/qiling).
 
 ### Building AFL++'s Unicorn Mode
 
@@ -34,23 +34,23 @@ cd unicorn_mode
 ```
 
 NOTE: This script checks out a Unicorn Engine fork as submodule that has been tested 
-and is stable-ish, based on the unicorn engine master. 
+and is stable-ish, based on the unicorn engine `next` branch. 
 
 Building Unicorn will take a little bit (~5-10 minutes). Once it completes 
 it automatically compiles a sample application and verifies that it works.
 
 ### Fuzzing with Unicorn Mode
 
-To really use unicorn-mode effectively you need to prepare the following:
+To use unicorn-mode effectively you need to prepare the following:
 
 	* Relevant binary code to be fuzzed
 	* Knowledge of the memory map and good starting state
 	* Folder containing sample inputs to start fuzzing with
 		+ Same ideas as any other AFL inputs
-		+ Quality/speed of results will depend greatly on quality of starting 
+		+ Quality/speed of results will depend greatly on the quality of starting 
 		  samples
 		+ See AFL's guidance on how to create a sample corpus
-	* Unicornafl-based test harness which:
+	* Unicornafl-based test harness in Rust, C, or Python, which:
 		+ Adds memory map regions
 		+ Loads binary code into memory		
 		+ Calls uc.afl_fuzz() / uc.afl_start_forkserver
@@ -59,13 +59,13 @@ To really use unicorn-mode effectively you need to prepare the following:
 			  the test harness
 			+ Presumably the data to be fuzzed is at a fixed buffer address
 			+ If input constraints (size, invalid bytes, etc.) are known they 
-			  should be checked after the file is loaded. If a constraint 
-			  fails, just exit the test harness. AFL will treat the input as 
+			  should be checked in the place_input handler. If a constraint 
+			  fails, just return false from the handler. AFL will treat the input as 
 			  'uninteresting' and move on.
 		+ Sets up registers and memory state for beginning of test
-		+ Emulates the interested code from beginning to end
+		+ Emulates the interesting code from beginning to end
 		+ If a crash is detected, the test harness must 'crash' by 
-		  throwing a signal (SIGSEGV, SIGKILL, SIGABORT, etc.)
+		  throwing a signal (SIGSEGV, SIGKILL, SIGABORT, etc.), or indicate a crash in the crash validation callback.
 
 Once you have all those things ready to go you just need to run afl-fuzz in
 'unicorn-mode' by passing in the '-U' flag:
@@ -79,11 +79,12 @@ AFL's main documentation for more info about how to use afl-fuzz effectively.
 
 For a much clearer vision of what all of this looks like, please refer to the
 sample provided in the 'unicorn_mode/samples' directory. There is also a blog
-post that goes over the basics at:
+post that uses slightly older concepts, but describes the general ideas, at:
 
 [https://medium.com/@njvoss299/afl-unicorn-fuzzing-arbitrary-binary-code-563ca28936bf](https://medium.com/@njvoss299/afl-unicorn-fuzzing-arbitrary-binary-code-563ca28936bf)
 
-The 'helper_scripts' directory also contains several helper scripts that allow you 
+
+The ['helper_scripts'](./helper_scripts) directory also contains several helper scripts that allow you 
 to dump context from a running process, load it, and hook heap allocations. For details
 on how to use this check out the follow-up blog post to the one linked above.
 
@@ -92,10 +93,10 @@ A example use of AFL-Unicorn mode is discussed in the paper Unicorefuzz:
 
 ## 3) Options
 
-As for the QEMU-based instrumentation, the afl-unicorn twist of afl++
-comes with a sub-instruction based instrumentation similar in purpose to laf-intel.
+As for the QEMU-based instrumentation, unicornafl comes with a sub-instruction based instrumentation similar in purpose to laf-intel.
 
 The options that enable Unicorn CompareCoverage are the same used for QEMU.
+This will split up each multi-byte compare to give feedback for each correct byte.
 AFL_COMPCOV_LEVEL=1 is to instrument comparisons with only immediate values.
 
 AFL_COMPCOV_LEVEL=2 instruments all comparison instructions.
@@ -119,6 +120,20 @@ unicornafl.monkeypatch()
 
 This will replace all unicorn imports with unicornafl inputs.
 
-Refer to the [samples/arm_example/arm_tester.c](samples/arm_example/arm_tester.c) for an example
-of how to do this properly! If you don't get this right, AFL will not 
-load any mutated inputs and your fuzzing will be useless!
+5) Examples
+
+Apart from reading the documentation in `afl.c` and the python bindings of unicornafl, the best documentation are the [samples/](./samples).
+The following examples exist at the time of writing:
+
+- c: A simple example how to use the c bindings
+- compcov_x64: A python example that uses compcov to traverse hard-to-reach blocks
+- persistent: A c example using persistent mode for maximum speed, and resetting the target state between each iteration
+- simple: A simple python example
+- speedtest/c: The c harness for an example target, used to compare c, python, and rust bindings and fix speed issues
+- speedtest/python: Fuzzing the same target in python
+- speedtest/rust: Fuzzing the same target using a rust harness
+
+Usually, the place to look at is the `harness` in each folder. The source code in each harness is pretty well documented.
+Most harnesses also have the `afl-fuzz` commandline, or even offer a `make fuzz` Makefile target.
+Targets in these folders, if x86, can usually be made using `make target` in each folder or get shipped pre-built (plus their source).
+Especially take a look at the [speedtest documentation](./samples/speedtest/README.md) to see how the languages compare.

unicorn_mode/build_unicorn_support.sh

@@ -117,19 +117,19 @@ done
 
 # some python version should be available now
 PYTHONS="`command -v python3` `command -v python` `command -v python2`"
-EASY_INSTALL_FOUND=0
+SETUPTOOLS_FOUND=0
 for PYTHON in $PYTHONS ; do
 
   if $PYTHON -c "import setuptools" ; then
 
-    EASY_INSTALL_FOUND=1
+    SETUPTOOLS_FOUND=1
     PYTHONBIN=$PYTHON
     break
 
   fi
 
 done
-if [ "0" = $EASY_INSTALL_FOUND ]; then
+if [ "0" = $SETUPTOOLS_FOUND ]; then
 
   echo "[-] Error: Python setup-tools not found. Run 'sudo apt-get install python-setuptools', or install python3-setuptools, or run '$PYTHONBIN -m ensurepip', or create a virtualenv, or ..."
   PREREQ_NOTFOUND=1

unicorn_mode/helper_scripts/unicorn_dumper_gdb.py

@@ -45,30 +45,31 @@ MAX_SEG_SIZE = 128 * 1024 * 1024
 INDEX_FILE_NAME = "_index.json"
 
 
-#----------------------
-#---- Helper Functions
+# ----------------------
+# ---- Helper Functions
+
 
 def map_arch():
-    arch = get_arch() # from GEF
-    if 'x86_64' in arch or 'x86-64' in arch:
+    arch = get_arch()  # from GEF
+    if "x86_64" in arch or "x86-64" in arch:
         return "x64"
-    elif 'x86' in arch or 'i386' in arch:
+    elif "x86" in arch or "i386" in arch:
         return "x86"
-    elif 'aarch64' in arch or 'arm64' in arch:
+    elif "aarch64" in arch or "arm64" in arch:
         return "arm64le"
-    elif 'aarch64_be' in arch:
+    elif "aarch64_be" in arch:
         return "arm64be"
-    elif 'armeb' in arch:
+    elif "armeb" in arch:
         # check for THUMB mode
-        cpsr = get_register('$cpsr')
-        if (cpsr & (1 << 5)):
+        cpsr = get_register("$cpsr")
+        if cpsr & (1 << 5):
             return "armbethumb"
         else:
             return "armbe"
-    elif 'arm' in arch:
+    elif "arm" in arch:
         # check for THUMB mode
-        cpsr = get_register('$cpsr')
-        if (cpsr & (1 << 5)):
+        cpsr = get_register("$cpsr")
+        if cpsr & (1 << 5):
             return "armlethumb"
         else:
             return "armle"
@@ -76,8 +77,9 @@ def map_arch():
         return ""
 
 
-#-----------------------
-#---- Dumping functions
+# -----------------------
+# ---- Dumping functions
+
 
 def dump_arch_info():
     arch_info = {}
@@ -89,7 +91,7 @@ def dump_regs():
     reg_state = {}
     for reg in current_arch.all_registers:
         reg_val = get_register(reg)
-        reg_state[reg.strip().strip('$')] = reg_val
+        reg_state[reg.strip().strip("$")] = reg_val
 
     return reg_state
 
@@ -108,47 +110,76 @@ def dump_process_memory(output_dir):
         if entry.page_start == entry.page_end:
             continue
 
-        seg_info = {'start': entry.page_start, 'end': entry.page_end, 'name': entry.path, 'permissions': {
-            "r": entry.is_readable() > 0,
-            "w": entry.is_writable() > 0,
-            "x": entry.is_executable() > 0
-        }, 'content_file': ''}
+        seg_info = {
+            "start": entry.page_start,
+            "end": entry.page_end,
+            "name": entry.path,
+            "permissions": {
+                "r": entry.is_readable() > 0,
+                "w": entry.is_writable() > 0,
+                "x": entry.is_executable() > 0,
+            },
+            "content_file": "",
+        }
 
         # "(deleted)" may or may not be valid, but don't push it.
-        if entry.is_readable() and not '(deleted)' in entry.path:
+        if entry.is_readable() and not "(deleted)" in entry.path:
             try:
                 # Compress and dump the content to a file
                 seg_content = read_memory(entry.page_start, entry.size)
-                if(seg_content == None):
-                    print("Segment empty: @0x{0:016x} (size:UNKNOWN) {1}".format(entry.page_start, entry.path))
+                if seg_content == None:
+                    print(
+                        "Segment empty: @0x{0:016x} (size:UNKNOWN) {1}".format(
+                            entry.page_start, entry.path
+                        )
+                    )
                 else:
-                    print("Dumping segment @0x{0:016x} (size:0x{1:x}): {2} [{3}]".format(entry.page_start, len(seg_content), entry.path, repr(seg_info['permissions'])))
+                    print(
+                        "Dumping segment @0x{0:016x} (size:0x{1:x}): {2} [{3}]".format(
+                            entry.page_start,
+                            len(seg_content),
+                            entry.path,
+                            repr(seg_info["permissions"]),
+                        )
+                    )
                     compressed_seg_content = zlib.compress(seg_content)
                     md5_sum = hashlib.md5(compressed_seg_content).hexdigest() + ".bin"
                     seg_info["content_file"] = md5_sum
 
                     # Write the compressed contents to disk
-                    out_file = open(os.path.join(output_dir, md5_sum), 'wb')
+                    out_file = open(os.path.join(output_dir, md5_sum), "wb")
                     out_file.write(compressed_seg_content)
                     out_file.close()
 
             except:
-                print("Exception reading segment ({}): {}".format(entry.path, sys.exc_info()[0]))
+                print(
+                    "Exception reading segment ({}): {}".format(
+                        entry.path, sys.exc_info()[0]
+                    )
+                )
         else:
-            print("Skipping segment {0}@0x{1:016x}".format(entry.path, entry.page_start))
+            print(
+                "Skipping segment {0}@0x{1:016x}".format(entry.path, entry.page_start)
+            )
 
         # Add the segment to the list
         final_segment_list.append(seg_info)
 
-
     return final_segment_list
 
-#---------------------------------------------
-#---- ARM Extention (dump floating point regs)
+
+# ---------------------------------------------
+# ---- ARM Extention (dump floating point regs)
+
 
 def dump_float(rge=32):
     reg_convert = ""
-    if map_arch() == "armbe" or map_arch() == "armle" or map_arch() == "armbethumb" or map_arch() == "armbethumb":
+    if (
+        map_arch() == "armbe"
+        or map_arch() == "armle"
+        or map_arch() == "armbethumb"
+        or map_arch() == "armbethumb"
+    ):
         reg_state = {}
         for reg_num in range(32):
             value = gdb.selected_frame().read_register("d" + str(reg_num))
@@ -158,8 +189,10 @@ def dump_float(rge=32):
 
         return reg_state
 
-#----------
-#---- Main
+
+# ----------
+# ---- Main
+
 
 def main():
     print("----- Unicorn Context Dumper -----")
@@ -175,7 +208,9 @@ def main():
     try:
 
         # Create the output directory
-        timestamp = datetime.datetime.fromtimestamp(time.time()).strftime('%Y%m%d_%H%M%S')
+        timestamp = datetime.datetime.fromtimestamp(time.time()).strftime(
+            "%Y%m%d_%H%M%S"
+        )
         output_path = "UnicornContext_" + timestamp
         if not os.path.exists(output_path):
             os.makedirs(output_path)
@@ -190,7 +225,7 @@ def main():
         }
 
         # Write the index file
-        index_file = open(os.path.join(output_path, INDEX_FILE_NAME), 'w')
+        index_file = open(os.path.join(output_path, INDEX_FILE_NAME), "w")
         index_file.write(json.dumps(context, indent=4))
         index_file.close()
         print("Done.")
@@ -198,5 +233,6 @@ def main():
     except Exception as e:
         print("!!! ERROR:\n\t{}".format(repr(e)))
 
+
 if __name__ == "__main__":
     main()