Add build script for symbolizer which can be linked into instrumented process.

Reviewers: eugenis

Subscribers: llvm-commits, kubabrecka

Differential Revision: https://reviews.llvm.org/D24974

git-svn-id: https://llvm.org/svn/llvm-project/compiler-rt/trunk@282617 91177308-0d34-0410-b5e6-96231b3b80d8

2 files changed, 313 insertions, 0 deletions

diff --git a/lib/sanitizer_common/symbolizer/scripts/build_symbolizer.sh b/lib/sanitizer_common/symbolizer/scripts/build_symbolizer.sh
new file mode 100755
index 000000000..a1206b182
--- /dev/null
+++ b/lib/sanitizer_common/symbolizer/scripts/build_symbolizer.sh
@@ -0,0 +1,183 @@
+#!/bin/bash -eu
+#
+# Run as: CLANG=bin/clang ZLIB_SRC=src/zlib \
+#             build_symbolizer.sh runtime_build/lib/clang/4.0.0/lib/linux/
+# zlib can be downloaded from from http://www.zlib.net.
+#
+# Script compiles self-contained object file with symbolization code and injects
+# it into the given set of runtime libraries. Script updates only libraries
+# which has unresolved __sanitizer_symbolize_* symbols and matches architecture.
+# Object file is be compiled from LLVM sources with dependencies like libc++ and
+# zlib. Then it internalizes symbols in the file, so that it can be linked
+# into arbitrary programs, avoiding conflicts with the program own symbols and
+# avoiding dependencies on any program symbols. The only acceptable dependencies
+# are libc and __sanitizer::internal_* from sanitizer runtime.
+#
+# Symbols exported by the object file will be used by Sanitizer runtime
+# libraries to symbolize code/data in-process.
+#
+# The script will modify the output directory which is given as the first
+# argument to the script.
+
+SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
+SRC_DIR=$(readlink -f $SCRIPT_DIR/..)
+BUILD_DIR=$(readlink -f ./symbolizer)
+TARGE_DIR=$(readlink -f $1)
+mkdir -p $BUILD_DIR
+
+LLVM_SRC=$(readlink -f $SCRIPT_DIR/../../../../../..)
+
+if [[ "$LLVM_SRC" == "" ||
+      ! -d "${LLVM_SRC}/projects/libcxxabi" ||
+      ! -d "${LLVM_SRC}/projects/libcxx" ]]; then
+  echo "Missing or incomplete LLVM_SRC"
+  exit 1
+fi
+LLVM_SRC=$(readlink -f $LLVM_SRC)
+
+if [[ "$ZLIB_SRC" == ""  ||
+      ! -x "${ZLIB_SRC}/configure" ||
+      ! -f "${ZLIB_SRC}/zlib.h" ]]; then
+  echo "Missing or incomplete ZLIB_SRC"
+  exit 1
+fi
+ZLIB_SRC=$(readlink -f $ZLIB_SRC)
+
+J="${J:-50}"
+
+CLANG="${CLANG:-`which clang`}"
+CLANG_DIR=$(readlink -f $(dirname "$CLANG"))
+
+CC=$CLANG_DIR/clang
+CXX=$CLANG_DIR/clang++
+TBLGEN=$CLANG_DIR/llvm-tblgen
+LINK=$CLANG_DIR/llvm-link
+OPT=$CLANG_DIR/opt
+AR=$CLANG_DIR/llvm-ar
+
+if [[ ! -x "$CC" ||
+      ! -x "$CXX" ||
+      ! -x "$TBLGEN" ||
+      ! -x "$LINK" ||
+      ! -x "$OPT" ||
+      ! -x "$AR" ]]; then
+ echo "Missing or incomplete CLANG_DIR"
+ exit 1
+fi
+
+ZLIB_BUILD=${BUILD_DIR}/zlib
+LIBCXX_BUILD=${BUILD_DIR}/libcxx
+LLVM_BUILD=${BUILD_DIR}/llvm
+SYMBOLIZER_BUILD=${BUILD_DIR}/symbolizer
+
+FLAGS=${FLAGS:-}
+FLAGS="$FLAGS -fPIC -flto -Os -g0 -DNDEBUG"
+
+# Build zlib.
+mkdir -p ${ZLIB_BUILD}
+cd ${ZLIB_BUILD}
+cp -r ${ZLIB_SRC}/* .
+CC=$CC CFLAGS="$FLAGS" RANLIB=/bin/true ./configure --static
+make -j${J} libz.a
+
+# Build and install libcxxabi and libcxx.
+if [[ ! -d ${LIBCXX_BUILD} ]]; then
+  mkdir -p ${LIBCXX_BUILD}
+  cd ${LIBCXX_BUILD}
+  LIBCXX_FLAGS="${FLAGS} -I${LLVM_SRC}/projects/libcxxabi/include"
+  cmake -GNinja \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DCMAKE_C_COMPILER=$CC \
+    -DCMAKE_CXX_COMPILER=$CXX \
+    -DCMAKE_C_FLAGS_RELEASE="${LIBCXX_FLAGS}" \
+    -DCMAKE_CXX_FLAGS_RELEASE="${LIBCXX_FLAGS}" \
+    -DLIBCXXABI_ENABLE_ASSERTIONS=OFF \
+    -DLIBCXXABI_ENABLE_EXCEPTIONS=OFF \
+    -DLIBCXXABI_ENABLE_SHARED=OFF \
+    -DLIBCXXABI_ENABLE_THREADS=OFF \
+    -DLIBCXX_ENABLE_ASSERTIONS=OFF \
+    -DLIBCXX_ENABLE_EXCEPTIONS=OFF \
+    -DLIBCXX_ENABLE_RTTI=OFF \
+    -DLIBCXX_ENABLE_SHARED=OFF \
+    -DLIBCXX_ENABLE_THREADS=OFF \
+  $LLVM_SRC
+fi
+cd ${LIBCXX_BUILD}
+ninja cxx cxxabi
+
+FLAGS="${FLAGS} -fno-rtti -fno-exceptions"
+
+# Build LLVM.
+if [[ ! -d ${LLVM_BUILD} ]]; then
+  mkdir -p ${LLVM_BUILD}
+  cd ${LLVM_BUILD}
+  LLVM_FLAGS="${FLAGS} -I${ZLIB_BUILD} -I${LIBCXX_BUILD}/include/c++/v1"
+  cmake -GNinja \
+    -DCMAKE_BUILD_TYPE=Release \
+    -DCMAKE_C_COMPILER=$CC \
+    -DCMAKE_CXX_COMPILER=$CXX \
+    -DCMAKE_C_FLAGS_RELEASE="${LLVM_FLAGS}" \
+    -DCMAKE_CXX_FLAGS_RELEASE="${LLVM_FLAGS}" \
+    -DLLVM_TABLEGEN=$TBLGEN \
+    -DLLVM_ENABLE_ZLIB=ON \
+    -DLLVM_ENABLE_TERMINFO=OFF \
+    -DLLVM_ENABLE_THREADS=OFF \
+  $LLVM_SRC
+fi
+cd ${LLVM_BUILD}
+ninja LLVMSymbolize LLVMObject LLVMDebugInfoDWARF LLVMSupport LLVMDebugInfoPDB LLVMMC
+
+cd ${BUILD_DIR}
+rm -rf ${SYMBOLIZER_BUILD}
+mkdir ${SYMBOLIZER_BUILD}
+cd ${SYMBOLIZER_BUILD}
+
+for A in $LIBCXX_BUILD/lib/libc++.a \
+         $LIBCXX_BUILD/lib/libc++abi.a \
+         $LLVM_BUILD/lib/libLLVMSymbolize.a \
+         $LLVM_BUILD/lib/libLLVMObject.a \
+         $LLVM_BUILD/lib/libLLVMDebugInfoDWARF.a \
+         $LLVM_BUILD/lib/libLLVMSupport.a \
+         $LLVM_BUILD/lib/libLLVMDebugInfoPDB.a \
+         $LLVM_BUILD/lib/libLLVMMC.a \
+         $ZLIB_BUILD/libz.a ; do
+  for O in $($AR t $A); do
+    $AR x $A $O
+    mv -f $O "$(basename $A).$O" # Rename to avoid collisions between libs.
+  done
+done
+
+echo "Compiling..."
+SYMBOLIZER_FLAGS="$FLAGS -std=c++11 -I${LLVM_SRC}/include -I${LLVM_BUILD}/include -I${LIBCXX_BUILD}/include/c++/v1"
+$CXX $SYMBOLIZER_FLAGS ${SRC_DIR}/sanitizer_symbolize.cc ${SRC_DIR}/sanitizer_wrappers.cc -c
+
+SYMBOLIZER_API_LIST=__sanitizer_symbolize_code,__sanitizer_symbolize_data,__sanitizer_symbolize_flush,__sanitizer_symbolize_demangle
+
+# Merge all the object files together and copy the resulting library back.
+$LINK *.o -o all.bc
+echo "Optimizing..."
+$OPT -internalize -internalize-public-api-list=${SYMBOLIZER_API_LIST} all.bc -o opt.bc
+$CC $FLAGS -fno-lto -c opt.bc -o symbolizer.o
+
+echo "Checking undefined symbols..."
+nm -f posix -g symbolizer.o | cut -f 1,2 -d \  | LC_COLLATE=C sort -u > undefined.new
+(diff -u $SCRIPT_DIR/global_symbols.txt undefined.new | grep -E "^\+[^+]") && \
+  (echo "Failed: unexpected symbols"; exit 1)
+
+arch() {
+  objdump -f $1 | grep -m1 -Po "(?<=file format ).*$"
+}
+
+SYMBOLIZER_FORMAT=$(arch symbolizer.o)
+echo "Injecting $SYMBOLIZER_FORMAT symbolizer..."
+for A in $TARGE_DIR/*.a; do
+  A_FORMAT=$(arch $A)
+  if [[ "$A_FORMAT" != "$SYMBOLIZER_FORMAT" ]] ; then
+    continue
+  fi
+  (nm -u $A | grep -E "__sanitizer_symbolize_code" >/dev/null) || continue
+  echo "$A"
+  $AR rcs $A symbolizer.o
+done
+
+echo "Success!"
diff --git a/lib/sanitizer_common/symbolizer/scripts/global_symbols.txt b/lib/sanitizer_common/symbolizer/scripts/global_symbols.txt
new file mode 100644
index 000000000..dcaa8cdf3
--- /dev/null
+++ b/lib/sanitizer_common/symbolizer/scripts/global_symbols.txt
@@ -0,0 +1,130 @@
+_ZN11__sanitizer13internal_mmapEPvmiiiy U
+_ZN11__sanitizer13internal_openEPKcij U
+_ZN11__sanitizer13internal_statEPKcPv U
+_ZN11__sanitizer14internal_closeEi U
+_ZN11__sanitizer14internal_fstatEiPv U
+_ZN11__sanitizer14internal_lstatEPKcPv U
+_ZN11__sanitizer15internal_strlenEPKc U
+_ZN11__sanitizer16internal_iserrorEmPi U
+_ZN11__sanitizer17internal_snprintfEPcmPKcz U
+__ctype_b_loc U
+__ctype_get_mb_cur_max U
+__cxa_atexit U
+__dso_handle U
+__errno_location U
+__interceptor_pread w
+__interceptor_read w
+__interceptor_realpath w
+__sanitizer_symbolize_code T
+__sanitizer_symbolize_data T
+__sanitizer_symbolize_demangle T
+__sanitizer_symbolize_flush T
+__strdup U
+_exit U
+abort U
+access U
+calloc U
+catclose U
+catgets U
+catopen U
+ceil U
+cfgetospeed U
+dl_iterate_phdr U
+dlsym U
+dup2 U
+environ U
+execv U
+exit U
+fclose U
+fflush U
+fileno U
+fopen U
+fork U
+fprintf U
+fputc U
+free U
+freelocale U
+fwrite U
+getc U
+getcwd U
+getenv U
+getpagesize U
+getpid U
+gettimeofday U
+ioctl U
+isatty U
+isprint U
+isupper U
+isxdigit U
+log10 U
+lseek U
+malloc U
+mbrlen U
+mbrtowc U
+mbsnrtowcs U
+mbsrtowcs U
+mbtowc U
+memchr U
+memcmp U
+memcpy U
+memmove U
+memset U
+mkdir U
+munmap U
+newlocale U
+perror U
+posix_spawn U
+posix_spawn_file_actions_adddup2 U
+posix_spawn_file_actions_addopen U
+posix_spawn_file_actions_destroy U
+posix_spawn_file_actions_init U
+qsort U
+rand U
+readlink U
+realloc U
+remove U
+setvbuf U
+sigfillset U
+sigprocmask U
+snprintf U
+sprintf U
+srand U
+sscanf U
+stderr U
+stdin U
+stdout U
+strcat U
+strchr U
+strcmp U
+strcpy U
+strdup U
+strerror U
+strerror_r U
+strftime_l U
+strncmp U
+strncpy U
+strrchr U
+strsep U
+strtod_l U
+strtof_l U
+strtol U
+strtold_l U
+strtoll_l U
+strtoull_l U
+tcgetattr U
+uname U
+ungetc U
+unlink U
+uselocale U
+vasprintf U
+vfprintf U
+vsnprintf U
+vsscanf U
+waitpid U
+wcrtomb U
+wcslen U
+wcsnrtombs U
+wmemcpy U
+wmemmove U
+wmemset U
+write U