Merge pull request #4372 from nia-e/multiple-native-libs

native-lib: allow multiple libraries and/or dirs

Author: oli-obk

src/tools/miri/README.md

@@ -398,9 +398,11 @@ to Miri failing to detect cases of undefined behavior in a program.
   **unsound** since the fallback body might not be checking for all UB.
 * `-Zmiri-native-lib=<path to a shared object file>` is an experimental flag for providing support
   for calling native functions from inside the interpreter via FFI. The flag is supported only on
-  Unix systems. Functions not provided by that file are still executed via the usual Miri shims.
+  Unix systems. Functions not provided by that file are still executed via the usual Miri shims. If
+  a path to a directory is specified, all files in that directory are included nonrecursively. This
+  flag can be passed multiple times to specify multiple files and/or directories.
   **WARNING**: If an invalid/incorrect `.so` file is specified, this can cause Undefined Behavior in
-  Miri itself! And of course, Miri cannot do any checks on the actions taken by the native code.
+  Miri itself! And of course, Miri often cannot do any checks on the actions taken by the native code.
   Note that Miri has its own handling of file descriptors, so if you want to replace *some*
   functions working on file descriptors, you will have to replace *all* of them, or the two kinds of
   file descriptors will be mixed up. This is **work in progress**; currently, only integer and

src/tools/miri/src/alloc_addresses/mod.rs

@@ -132,7 +132,7 @@ trait EvalContextExtPriv<'tcx>: crate::MiriInterpCxExt<'tcx> {
         assert!(!matches!(info.kind, AllocKind::Dead));
 
         // This allocation does not have a base address yet, pick or reuse one.
-        if this.machine.native_lib.is_some() {
+        if !this.machine.native_lib.is_empty() {
             // In native lib mode, we use the "real" address of the bytes for this allocation.
             // This ensures the interpreted program and native code have the same view of memory.
             let params = this.machine.get_default_alloc_params();
@@ -413,7 +413,7 @@ pub trait EvalContextExt<'tcx>: crate::MiriInterpCxExt<'tcx> {
     ) -> InterpResult<'tcx, MiriAllocBytes> {
         let this = self.eval_context_ref();
         assert!(this.tcx.try_get_global_alloc(id).is_some());
-        if this.machine.native_lib.is_some() {
+        if !this.machine.native_lib.is_empty() {
             // In native lib mode, MiriAllocBytes for global allocations are handled via `prepared_alloc_bytes`.
             // This additional call ensures that some `MiriAllocBytes` are always prepared, just in case
             // this function gets called before the first time `addr_from_alloc_id` gets called.

src/tools/miri/src/bin/miri.rs

@@ -692,11 +692,18 @@ fn main() {
             };
         } else if let Some(param) = arg.strip_prefix("-Zmiri-native-lib=") {
             let filename = param.to_string();
-            if std::path::Path::new(&filename).exists() {
-                if let Some(other_filename) = miri_config.native_lib {
-                    show_error!("-Zmiri-native-lib is already set to {}", other_filename.display());
+            let file_path = std::path::Path::new(&filename);
+            if file_path.exists() {
+                // For directories, nonrecursively add all normal files inside
+                if let Ok(dir) = file_path.read_dir() {
+                    for lib in dir.filter_map(|res| res.ok()) {
+                        if lib.file_type().unwrap().is_file() {
+                            miri_config.native_lib.push(lib.path().to_owned());
+                        }
+                    }
+                } else {
+                    miri_config.native_lib.push(filename.into());
                 }
-                miri_config.native_lib = Some(filename.into());
             } else {
                 show_error!("-Zmiri-native-lib `{}` does not exist", filename);
             }
@@ -731,12 +738,12 @@ fn main() {
                 "Tree Borrows does not support integer-to-pointer casts, and hence requires strict provenance"
             );
         }
-        if miri_config.native_lib.is_some() {
+        if !miri_config.native_lib.is_empty() {
             show_error!("Tree Borrows is not compatible with calling native functions");
         }
     }
     // Native calls and strict provenance are not compatible.
-    if miri_config.native_lib.is_some() && miri_config.provenance_mode == ProvenanceMode::Strict {
+    if !miri_config.native_lib.is_empty() && miri_config.provenance_mode == ProvenanceMode::Strict {
         show_error!("strict provenance is not compatible with calling native functions");
     }
     // You can set either one seed or many.

src/tools/miri/src/eval.rs

@@ -148,9 +148,8 @@ pub struct MiriConfig {
     pub report_progress: Option<u32>,
     /// Whether Stacked Borrows and Tree Borrows retagging should recurse into fields of datatypes.
     pub retag_fields: RetagFields,
-    /// The location of a shared object file to load when calling external functions
-    /// FIXME! consider allowing users to specify paths to multiple files, or to a directory
-    pub native_lib: Option<PathBuf>,
+    /// The location of the shared object files to load when calling external functions
+    pub native_lib: Vec<PathBuf>,
     /// Run a garbage collector for BorTags every N basic blocks.
     pub gc_interval: u32,
     /// The number of CPUs to be reported by miri.
@@ -197,7 +196,7 @@ impl Default for MiriConfig {
             preemption_rate: 0.01, // 1%
             report_progress: None,
             retag_fields: RetagFields::Yes,
-            native_lib: None,
+            native_lib: vec![],
             gc_interval: 10_000,
             num_cpus: 1,
             page_size: None,

src/tools/miri/src/machine.rs

@@ -558,9 +558,9 @@ pub struct MiriMachine<'tcx> {
 
     /// Handle of the optional shared object file for native functions.
     #[cfg(unix)]
-    pub native_lib: Option<(libloading::Library, std::path::PathBuf)>,
+    pub native_lib: Vec<(libloading::Library, std::path::PathBuf)>,
     #[cfg(not(unix))]
-    pub native_lib: Option<!>,
+    pub native_lib: Vec<!>,
 
     /// Run a garbage collector for BorTags every N basic blocks.
     pub(crate) gc_interval: u32,
@@ -720,7 +720,7 @@ impl<'tcx> MiriMachine<'tcx> {
             extern_statics: FxHashMap::default(),
             rng: RefCell::new(rng),
             #[cfg(target_os = "linux")]
-            allocator: if config.native_lib.is_some() {
+            allocator: if !config.native_lib.is_empty() {
                 Some(Rc::new(RefCell::new(crate::alloc::isolated_alloc::IsolatedAlloc::new())))
             } else { None },
             tracked_alloc_ids: config.tracked_alloc_ids.clone(),
@@ -732,7 +732,7 @@ impl<'tcx> MiriMachine<'tcx> {
             basic_block_count: 0,
             monotonic_clock: MonotonicClock::new(config.isolated_op == IsolatedOp::Allow),
             #[cfg(unix)]
-            native_lib: config.native_lib.as_ref().map(|lib_file_path| {
+            native_lib: config.native_lib.iter().map(|lib_file_path| {
                 let host_triple = rustc_session::config::host_tuple();
                 let target_triple = tcx.sess.opts.target_triple.tuple();
                 // Check if host target == the session target.
@@ -752,11 +752,11 @@ impl<'tcx> MiriMachine<'tcx> {
                     },
                     lib_file_path.clone(),
                 )
-            }),
+            }).collect(),
             #[cfg(not(unix))]
-            native_lib: config.native_lib.as_ref().map(|_| {
+            native_lib: config.native_lib.iter().map(|_| {
                 panic!("calling functions from native libraries via FFI is only supported on Unix")
-            }),
+            }).collect(),
             gc_interval: config.gc_interval,
             since_gc: 0,
             num_cpus: config.num_cpus,

src/tools/miri/src/shims/foreign_items.rs

@@ -238,7 +238,7 @@ trait EvalContextExtPriv<'tcx>: crate::MiriInterpCxExt<'tcx> {
 
         // First deal with any external C functions in linked .so file.
         #[cfg(unix)]
-        if this.machine.native_lib.as_ref().is_some() {
+        if !this.machine.native_lib.is_empty() {
             use crate::shims::native_lib::EvalContextExt as _;
             // An Ok(false) here means that the function being called was not exported
             // by the specified `.so` file; we should continue and check if it corresponds to

src/tools/miri/src/shims/native_lib.rs

@@ -87,47 +87,52 @@ trait EvalContextExtPriv<'tcx>: crate::MiriInterpCxExt<'tcx> {
     }
 
     /// Get the pointer to the function of the specified name in the shared object file,
-    /// if it exists. The function must be in the shared object file specified: we do *not*
-    /// return pointers to functions in dependencies of the library.  
+    /// if it exists. The function must be in one of the shared object files specified:
+    /// we do *not* return pointers to functions in dependencies of libraries.
     fn get_func_ptr_explicitly_from_lib(&mut self, link_name: Symbol) -> Option<CodePtr> {
         let this = self.eval_context_mut();
-        // Try getting the function from the shared library.
-        let (lib, lib_path) = this.machine.native_lib.as_ref().unwrap();
-        let func: libloading::Symbol<'_, unsafe extern "C" fn()> =
-            unsafe { lib.get(link_name.as_str().as_bytes()).ok()? };
-        #[expect(clippy::as_conversions)] // fn-ptr to raw-ptr cast needs `as`.
-        let fn_ptr = *func.deref() as *mut std::ffi::c_void;
+        // Try getting the function from one of the shared libraries.
+        for (lib, lib_path) in &this.machine.native_lib {
+            let Ok(func): Result<libloading::Symbol<'_, unsafe extern "C" fn()>, _> =
+                (unsafe { lib.get(link_name.as_str().as_bytes()) })
+            else {
+                continue;
+            };
+            #[expect(clippy::as_conversions)] // fn-ptr to raw-ptr cast needs `as`.
+            let fn_ptr = *func.deref() as *mut std::ffi::c_void;
 
-        // FIXME: this is a hack!
-        // The `libloading` crate will automatically load system libraries like `libc`.
-        // On linux `libloading` is based on `dlsym`: https://docs.rs/libloading/0.7.3/src/libloading/os/unix/mod.rs.html#202
-        // and `dlsym`(https://linux.die.net/man/3/dlsym) looks through the dependency tree of the
-        // library if it can't find the symbol in the library itself.
-        // So, in order to check if the function was actually found in the specified
-        // `machine.external_so_lib` we need to check its `dli_fname` and compare it to
-        // the specified SO file path.
-        // This code is a reimplementation of the mechanism for getting `dli_fname` in `libloading`,
-        // from: https://docs.rs/libloading/0.7.3/src/libloading/os/unix/mod.rs.html#411
-        // using the `libc` crate where this interface is public.
-        let mut info = std::mem::MaybeUninit::<libc::Dl_info>::zeroed();
-        unsafe {
-            if libc::dladdr(fn_ptr, info.as_mut_ptr()) != 0 {
-                let info = info.assume_init();
-                #[cfg(target_os = "cygwin")]
-                let fname_ptr = info.dli_fname.as_ptr();
-                #[cfg(not(target_os = "cygwin"))]
-                let fname_ptr = info.dli_fname;
-                assert!(!fname_ptr.is_null());
-                if std::ffi::CStr::from_ptr(fname_ptr).to_str().unwrap()
-                    != lib_path.to_str().unwrap()
-                {
-                    return None;
+            // FIXME: this is a hack!
+            // The `libloading` crate will automatically load system libraries like `libc`.
+            // On linux `libloading` is based on `dlsym`: https://docs.rs/libloading/0.7.3/src/libloading/os/unix/mod.rs.html#202
+            // and `dlsym`(https://linux.die.net/man/3/dlsym) looks through the dependency tree of the
+            // library if it can't find the symbol in the library itself.
+            // So, in order to check if the function was actually found in the specified
+            // `machine.external_so_lib` we need to check its `dli_fname` and compare it to
+            // the specified SO file path.
+            // This code is a reimplementation of the mechanism for getting `dli_fname` in `libloading`,
+            // from: https://docs.rs/libloading/0.7.3/src/libloading/os/unix/mod.rs.html#411
+            // using the `libc` crate where this interface is public.
+            let mut info = std::mem::MaybeUninit::<libc::Dl_info>::zeroed();
+            unsafe {
+                if libc::dladdr(fn_ptr, info.as_mut_ptr()) != 0 {
+                    let info = info.assume_init();
+                    #[cfg(target_os = "cygwin")]
+                    let fname_ptr = info.dli_fname.as_ptr();
+                    #[cfg(not(target_os = "cygwin"))]
+                    let fname_ptr = info.dli_fname;
+                    assert!(!fname_ptr.is_null());
+                    if std::ffi::CStr::from_ptr(fname_ptr).to_str().unwrap()
+                        != lib_path.to_str().unwrap()
+                    {
+                        return None;
+                    }
                 }
             }
-        }
 
-        // Return a pointer to the function.
-        Some(CodePtr(fn_ptr))
+            // Return a pointer to the function.
+            return Some(CodePtr(fn_ptr));
+        }
+        None
     }
 }