MSan support for Rust

Previously you have to unpoison memory from the Rust, however Rust does
supports MSan, so let's simply use it.

But for this we need nightly Rust and recompile standard library.

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>

docker/packager/binary/Dockerfile

@@ -46,10 +46,12 @@ ENV CXX=clang++-${LLVM_VERSION}
 # Rust toolchain and libraries
 ENV RUSTUP_HOME=/rust/rustup
 ENV CARGO_HOME=/rust/cargo
-ENV PATH="/rust/cargo/env:${PATH}"
 ENV PATH="/rust/cargo/bin:${PATH}"
 RUN curl https://sh.rustup.rs -sSf | bash -s -- -y && \
     chmod 777 -R /rust && \
+    rustup toolchain install nightly && \
+    rustup default nightly && \
+    rustup component add rust-src && \
     rustup target add aarch64-unknown-linux-gnu && \
     rustup target add x86_64-apple-darwin && \
     rustup target add x86_64-unknown-freebsd && \

rust/.cargo/config.toml.in

@@ -1,3 +1,10 @@
 [env]
 CFLAGS = "@RUST_CFLAGS@"
 CXXFLAGS = "@RUST_CXXFLAGS@"
+
+[build]
+rustflags = @RUSTFLAGS@
+rustdocflags = @RUSTFLAGS@
+
+[unstable]
+@RUST_CARGO_BUILD_STD@

rust/BLAKE3/include/blake3.h

@@ -8,8 +8,6 @@ extern "C" {
 
 char *blake3_apply_shim(const char *begin, uint32_t _size, uint8_t *out_char_data);
 
-char *blake3_apply_shim_msan_compat(const char *begin, uint32_t size, uint8_t *out_char_data);
-
 void blake3_free_char_pointer(char *ptr_to_free);
 
 } // extern "C"

rust/BLAKE3/src/lib.rs

@@ -3,7 +3,6 @@ extern crate libc;
 
 use std::ffi::{CStr, CString};
 use std::os::raw::c_char;
-use std::mem;
 
 #[no_mangle]
 pub unsafe extern "C" fn blake3_apply_shim(
@@ -24,30 +23,6 @@ pub unsafe extern "C" fn blake3_apply_shim(
     std::ptr::null_mut()
 }
 
-#[no_mangle]
-pub unsafe extern "C" fn blake3_apply_shim_msan_compat(
-    mut begin: *const c_char,
-    size: u32,
-    out_char_data: *mut u8,
-) -> *mut c_char {
-    if begin.is_null() {
-        let err_str = CString::new("input was a null pointer").unwrap();
-        return err_str.into_raw();
-    }
-    libc::memset(out_char_data as *mut libc::c_void, 0, mem::size_of::<u8>());
-    let mut hasher = blake3::Hasher::new();
-    let mut vec = Vec::<u8>::new();
-    for _ in 0..size {
-        vec.push(*begin as u8);
-        begin = begin.add(1);
-    }
-    let input_res = vec.as_mut_slice();
-    hasher.update(input_res);
-    let mut reader = hasher.finalize_xof();
-    reader.fill(std::slice::from_raw_parts_mut(out_char_data, blake3::OUT_LEN));
-    std::ptr::null_mut()
-}
-
 // Freeing memory according to docs: https://doc.rust-lang.org/std/ffi/struct.CString.html#method.into_raw
 #[no_mangle]
 pub unsafe extern "C" fn blake3_free_char_pointer(ptr_to_free: *mut c_char) {

rust/CMakeLists.txt

@@ -14,8 +14,18 @@ macro(configure_rustc)
         set(RUST_CFLAGS "${RUST_CFLAGS} --sysroot ${CMAKE_SYSROOT}")
     endif()
 
+    set(RUSTFLAGS "[]")
+    set(RUST_CARGO_BUILD_STD "")
+    # For more info: https://doc.rust-lang.org/beta/unstable-book/compiler-flags/sanitizer.html#memorysanitizer
+    if (SANITIZE STREQUAL "memory")
+        set(RUST_CARGO_BUILD_STD "build-std = [\"std\", \"panic_abort\", \"core\", \"alloc\"]")
+        set(RUSTFLAGS "[\"-Zsanitizer=memory\", \"-Zsanitizer-memory-track-origins\"]")
+    endif()
+
     message(STATUS "RUST_CFLAGS: ${RUST_CFLAGS}")
     message(STATUS "RUST_CXXFLAGS: ${RUST_CXXFLAGS}")
+    message(STATUS "RUSTFLAGS: ${RUSTFLAGS}")
+    message(STATUS "RUST_CARGO_BUILD_STD: ${RUST_CARGO_BUILD_STD}")
 
     # NOTE: requires RW access for the source dir
     configure_file("${CMAKE_CURRENT_SOURCE_DIR}/.cargo/config.toml.in" "${CMAKE_CURRENT_SOURCE_DIR}/.cargo/config.toml" @ONLY)

src/Functions/FunctionsHashing.h

@@ -816,12 +816,7 @@ struct ImplBLAKE3
 #else
     static void apply(const char * begin, const size_t size, unsigned char* out_char_data)
     {
-#    if defined(MEMORY_SANITIZER)
-            auto err_msg = blake3_apply_shim_msan_compat(begin, safe_cast<uint32_t>(size), out_char_data);
-            __msan_unpoison(out_char_data, length);
-#    else
-            auto err_msg = blake3_apply_shim(begin, safe_cast<uint32_t>(size), out_char_data);
-#    endif
+        auto err_msg = blake3_apply_shim(begin, safe_cast<uint32_t>(size), out_char_data);
         if (err_msg != nullptr)
         {
             auto err_st = std::string(err_msg);