Merge #18288: build: Add MemorySanitizer (MSan) in Travis to detect use of uninitialized memory

870f0cd2a0534d54bba18190e9f024f88e832933 build: Add MemorySanitizer (MSan) in Travis to detect use of uninitialized memory (practicalswift)

Pull request description:

  Add MemorySanitizer (MSan) in Travis to detect use of uninitialized memory.

  First UBSan, then ASan followed by TSan... and now: yes, the wait is over -- **MSan is finally here!** :)

  Some historical context:
  * 2017: Continuous compilation with Clang Thread Safety analysis enabled (#10866, #10923)
  * 2018: Continuous testing with trapping on signed integer overflows (`-ftrapv`) (#12686)
  * 2018: Continuous testing of use of locale dependent functions (#13041)
  * 2018: Continuous testing of format strings (#13705)
  * 2018: Continuous compilation with MSVC `TreatWarningAsError` (#14151)
  * 2018: Continuous testing under UndefinedBehaviorSanitizer – UBSan (#14252, #14673, #17006)
  * 2018: Continuous testing under AddressSanitizer – ASan (#14794, #17205, #17674)
  * 2018: Continuous testing under ThreadSanitizer – TSan (#14829)
  * 2019: Continuous testing in an unsigned char environment (`-funsigned-char`) (#15134)
  * 2019: Continuous compile-time testing of assumptions we're making (#15391)
  * 2019: Continuous testing of fuzz test cases under Valgrind (#17633, #18159, #18166)
  * 2020: Finally... MemorySanitizer – MSAN! :)

  What is the next step? What tools should we add to CI to keep bugs from entering `master`? :)

ACKs for top commit:
  MarcoFalke:
    ACK 870f0cd2a0534d54bba18190e9f024f88e832933

Tree-SHA512: 38327c8b75679d97d469fe42e704cacd1217447a5a603701dd8a58ee50b3be2c10248f8d68a479ed081c0c4b254589d3081c9183f991640b06ef689061f75578

5 files changed, 52 insertions, 0 deletions

diff --git a/.travis.yml b/.travis.yml
index 108a4c968d..9184257d47 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -105,6 +105,11 @@ jobs:
         FILE_ENV="./ci/test/00_setup_env_native_tsan.sh"
 
     - stage: test
+      name: 'x86_64 Linux  [GOAL: install]  [focal]  [depends, sanitizers: memory (MSan)]'
+      env: >-
+        FILE_ENV="./ci/test/00_setup_env_native_msan.sh"
+
+    - stage: test
       name: 'x86_64 Linux  [GOAL: install]  [focal]  [no depends, only system libs, sanitizers: fuzzer,address,undefined]'
       env: >-
         FILE_ENV="./ci/test/00_setup_env_native_fuzz.sh"
diff --git a/ci/test/00_setup_env_native_msan.sh b/ci/test/00_setup_env_native_msan.sh
new file mode 100644
index 0000000000..cc583edf17
--- /dev/null
+++ b/ci/test/00_setup_env_native_msan.sh
@@ -0,0 +1,22 @@
+#!/usr/bin/env bash
+#
+# Copyright (c) 2020 The Bitcoin Core developers
+# Distributed under the MIT software license, see the accompanying
+# file COPYING or http://www.opensource.org/licenses/mit-license.php.
+
+export LC_ALL=C.UTF-8
+
+export DOCKER_NAME_TAG="ubuntu:20.04"
+LIBCXX_DIR="${BASE_ROOT_DIR}/ci/scratch/msan/build/"
+export MSAN_FLAGS="-fsanitize=memory -fsanitize-memory-track-origins=2 -fno-omit-frame-pointer -g -O1 -fno-optimize-sibling-calls"
+LIBCXX_FLAGS="-nostdinc++ -stdlib=libc++ -L${LIBCXX_DIR}lib -lc++abi -I${LIBCXX_DIR}include -I${LIBCXX_DIR}include/c++/v1 -lpthread -Wl,-rpath,${LIBCXX_DIR}lib -Wno-unused-command-line-argument"
+export MSAN_AND_LIBCXX_FLAGS="${MSAN_FLAGS} ${LIBCXX_FLAGS}"
+export BDB_PREFIX="${BASE_ROOT_DIR}/db4"
+
+export CONTAINER_NAME="ci_native_msan"
+export PACKAGES="clang-9 llvm-9 cmake"
+export DEP_OPTS="NO_WALLET=1 NO_QT=1 CC='clang' CXX='clang++' CFLAGS='${MSAN_FLAGS}' CXXFLAGS='${MSAN_AND_LIBCXX_FLAGS}' boost_cxxflags='-std=c++11 -fvisibility=hidden -fPIC ${MSAN_AND_LIBCXX_FLAGS}' zeromq_cxxflags='-std=c++11 ${MSAN_AND_LIBCXX_FLAGS}'"
+export GOAL="install"
+export BITCOIN_CONFIG="--enable-wallet --with-sanitizers=memory --with-asm=no --prefix=${BASE_ROOT_DIR}/depends/x86_64-pc-linux-gnu/ CC=clang CXX=clang++ CFLAGS='${MSAN_FLAGS}' CXXFLAGS='${MSAN_AND_LIBCXX_FLAGS}' BDB_LIBS='-L${BDB_PREFIX}/lib -ldb_cxx-4.8' BDB_CFLAGS='-I${BDB_PREFIX}/include'"
+export USE_MEMORY_SANITIZER="true"
+export RUN_FUNCTIONAL_TESTS="false"
diff --git a/ci/test/04_install.sh b/ci/test/04_install.sh
index 165983d906..7cbf6f9d10 100755
--- a/ci/test/04_install.sh
+++ b/ci/test/04_install.sh
@@ -90,6 +90,15 @@ export DIR_FUZZ_IN=${DIR_QA_ASSETS}/fuzz_seed_corpus/
 
 DOCKER_EXEC mkdir -p "${BASE_SCRATCH_DIR}/sanitizer-output/"
 
+if [[ ${USE_MEMORY_SANITIZER} == "true" ]]; then
+  DOCKER_EXEC "update-alternatives --install /usr/bin/clang++ clang++ \$(which clang++-9) 100"
+  DOCKER_EXEC "update-alternatives --install /usr/bin/clang clang \$(which clang-9) 100"
+  DOCKER_EXEC "mkdir -p ${BASE_SCRATCH_DIR}/msan/build/"
+  DOCKER_EXEC "git clone --depth=1 https://github.com/llvm/llvm-project -b llvmorg-10.0.0 ${BASE_SCRATCH_DIR}/msan/llvm-project"
+  DOCKER_EXEC "cd ${BASE_SCRATCH_DIR}/msan/build/ && cmake -DLLVM_ENABLE_PROJECTS='libcxx;libcxxabi' -DCMAKE_BUILD_TYPE=Release -DLLVM_USE_SANITIZER=Memory -DCMAKE_C_COMPILER=clang -DCMAKE_CXX_COMPILER=clang++ -DLLVM_TARGETS_TO_BUILD=X86 ../llvm-project/llvm/"
+  DOCKER_EXEC "cd ${BASE_SCRATCH_DIR}/msan/build/ && make $MAKEJOBS cxx"
+fi
+
 if [ -z "$DANGER_RUN_CI_ON_HOST" ]; then
   echo "Create $BASE_ROOT_DIR"
   DOCKER_EXEC rsync -a /ro_base/ $BASE_ROOT_DIR
diff --git a/ci/test/05_before_script.sh b/ci/test/05_before_script.sh
index de33881419..057395eba6 100755
--- a/ci/test/05_before_script.sh
+++ b/ci/test/05_before_script.sh
@@ -21,6 +21,14 @@ OSX_SDK_PATH="${DEPENDS_DIR}/sdk-sources/${OSX_SDK_BASENAME}"
 if [ -n "$XCODE_VERSION" ] && [ ! -f "$OSX_SDK_PATH" ]; then
   curl --location --fail "${SDK_URL}/${OSX_SDK_BASENAME}" -o "$OSX_SDK_PATH"
 fi
+
+if [[ ${USE_MEMORY_SANITIZER} == "true" ]]; then
+  # Use BDB compiled using install_db4.sh script to work around linking issue when using BDB
+  # from depends. See https://github.com/bitcoin/bitcoin/pull/18288#discussion_r433189350 for
+  # details.
+  DOCKER_EXEC "contrib/install_db4.sh \$(pwd) --enable-umrw CC=clang CXX=clang++ CFLAGS='${MSAN_FLAGS}' CXXFLAGS='${MSAN_AND_LIBCXX_FLAGS}'"
+fi
+
 if [ -n "$XCODE_VERSION" ] && [ -f "$OSX_SDK_PATH" ]; then
   DOCKER_EXEC tar -C "${DEPENDS_DIR}/SDKs" -xf "$OSX_SDK_PATH"
 fi
diff --git a/ci/test/06_script_a.sh b/ci/test/06_script_a.sh
index b68cd9d3f8..17d765b862 100755
--- a/ci/test/06_script_a.sh
+++ b/ci/test/06_script_a.sh
@@ -37,6 +37,14 @@ END_FOLD
 set -o errtrace
 trap 'DOCKER_EXEC "cat ${BASE_SCRATCH_DIR}/sanitizer-output/* 2> /dev/null"' ERR
 
+if [[ ${USE_MEMORY_SANITIZER} == "true" ]]; then
+  # MemorySanitizer (MSAN) does not support tracking memory initialization done by
+  # using the Linux getrandom syscall. Avoid using getrandom by undefining
+  # HAVE_SYS_GETRANDOM. See https://github.com/google/sanitizers/issues/852 for
+  # details.
+  DOCKER_EXEC 'grep -v HAVE_SYS_GETRANDOM src/config/bitcoin-config.h > src/config/bitcoin-config.h.tmp && mv src/config/bitcoin-config.h.tmp src/config/bitcoin-config.h'
+fi
+
 BEGIN_FOLD build
 DOCKER_EXEC make $MAKEJOBS $GOAL || ( echo "Build failure. Verbose build follows." && DOCKER_EXEC make $GOAL V=1 ; false )
 END_FOLD