Rollup merge of rust-lang#103660 - ozkanonur:master, r=jyn514

improve `filesearch::get_or_default_sysroot`

`fn get_or_default_sysroot` is now improved and used in `miri` and `clippy`, and tests are still passing as they should. So we no longer need to implement custom workarounds/hacks to find sysroot in tools like miri/clippy.

Resolves rust-lang#98832

re-opened from rust-lang#103581

Cargo.lock

@@ -3646,7 +3646,6 @@ dependencies = [
 name = "rustc_interface"
 version = "0.0.0"
 dependencies = [
- "libc",
  "libloading",
  "rustc-rayon",
  "rustc-rayon-core",
@@ -3689,7 +3688,6 @@ dependencies = [
  "rustc_ty_utils",
  "smallvec",
  "tracing",
- "winapi",
 ]
 
 [[package]]
@@ -4109,6 +4107,7 @@ name = "rustc_session"
 version = "0.0.0"
 dependencies = [
  "getopts",
+ "libc",
  "rustc_ast",
  "rustc_data_structures",
  "rustc_errors",
@@ -4120,7 +4119,9 @@ dependencies = [
  "rustc_serialize",
  "rustc_span",
  "rustc_target",
+ "smallvec",
  "tracing",
+ "winapi",
 ]
 
 [[package]]

compiler/rustc_codegen_ssa/src/back/link.rs

@@ -1153,7 +1153,8 @@ fn link_sanitizer_runtime(sess: &Session, linker: &mut dyn Linker, name: &str) {
         if path.exists() {
             return session_tlib;
         } else {
-            let default_sysroot = filesearch::get_or_default_sysroot();
+            let default_sysroot =
+                filesearch::get_or_default_sysroot().expect("Failed finding sysroot");
             let default_tlib = filesearch::make_target_lib_path(
                 &default_sysroot,
                 sess.opts.target_triple.triple(),

compiler/rustc_interface/Cargo.toml

@@ -48,12 +48,6 @@ rustc_resolve = { path = "../rustc_resolve" }
 rustc_trait_selection = { path = "../rustc_trait_selection" }
 rustc_ty_utils = { path = "../rustc_ty_utils" }
 
-[target.'cfg(unix)'.dependencies]
-libc = "0.2"
-
-[target.'cfg(windows)'.dependencies]
-winapi = { version = "0.3", features = ["libloaderapi"] }
-
 [dev-dependencies]
 rustc_target = { path = "../rustc_target" }
 

compiler/rustc_interface/src/util.rs

@@ -9,6 +9,7 @@ use rustc_session as session;
 use rustc_session::config::CheckCfg;
 use rustc_session::config::{self, CrateType};
 use rustc_session::config::{ErrorOutputType, Input, OutputFilenames};
+use rustc_session::filesearch::sysroot_candidates;
 use rustc_session::lint::{self, BuiltinLintDiagnostics, LintBuffer};
 use rustc_session::parse::CrateConfig;
 use rustc_session::{early_error, filesearch, output, Session};
@@ -78,7 +79,7 @@ pub fn create_session(
 
     let bundle = match rustc_errors::fluent_bundle(
         sopts.maybe_sysroot.clone(),
-        sysroot_candidates(),
+        sysroot_candidates().to_vec(),
         sopts.unstable_opts.translate_lang.clone(),
         sopts.unstable_opts.translate_additional_ftl.as_deref(),
         sopts.unstable_opts.translate_directionality_markers,
@@ -273,100 +274,6 @@ fn get_rustc_path_inner(bin_path: &str) -> Option<PathBuf> {
     })
 }
 
-fn sysroot_candidates() -> Vec<PathBuf> {
-    let target = session::config::host_triple();
-    let mut sysroot_candidates = vec![filesearch::get_or_default_sysroot()];
-    let path = current_dll_path().and_then(|s| s.canonicalize().ok());
-    if let Some(dll) = path {
-        // use `parent` twice to chop off the file name and then also the
-        // directory containing the dll which should be either `lib` or `bin`.
-        if let Some(path) = dll.parent().and_then(|p| p.parent()) {
-            // The original `path` pointed at the `rustc_driver` crate's dll.
-            // Now that dll should only be in one of two locations. The first is
-            // in the compiler's libdir, for example `$sysroot/lib/*.dll`. The
-            // other is the target's libdir, for example
-            // `$sysroot/lib/rustlib/$target/lib/*.dll`.
-            //
-            // We don't know which, so let's assume that if our `path` above
-            // ends in `$target` we *could* be in the target libdir, and always
-            // assume that we may be in the main libdir.
-            sysroot_candidates.push(path.to_owned());
-
-            if path.ends_with(target) {
-                sysroot_candidates.extend(
-                    path.parent() // chop off `$target`
-                        .and_then(|p| p.parent()) // chop off `rustlib`
-                        .and_then(|p| p.parent()) // chop off `lib`
-                        .map(|s| s.to_owned()),
-                );
-            }
-        }
-    }
-
-    return sysroot_candidates;
-
-    #[cfg(unix)]
-    fn current_dll_path() -> Option<PathBuf> {
-        use std::ffi::{CStr, OsStr};
-        use std::os::unix::prelude::*;
-
-        unsafe {
-            let addr = current_dll_path as usize as *mut _;
-            let mut info = mem::zeroed();
-            if libc::dladdr(addr, &mut info) == 0 {
-                info!("dladdr failed");
-                return None;
-            }
-            if info.dli_fname.is_null() {
-                info!("dladdr returned null pointer");
-                return None;
-            }
-            let bytes = CStr::from_ptr(info.dli_fname).to_bytes();
-            let os = OsStr::from_bytes(bytes);
-            Some(PathBuf::from(os))
-        }
-    }
-
-    #[cfg(windows)]
-    fn current_dll_path() -> Option<PathBuf> {
-        use std::ffi::OsString;
-        use std::io;
-        use std::os::windows::prelude::*;
-        use std::ptr;
-
-        use winapi::um::libloaderapi::{
-            GetModuleFileNameW, GetModuleHandleExW, GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS,
-        };
-
-        unsafe {
-            let mut module = ptr::null_mut();
-            let r = GetModuleHandleExW(
-                GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS,
-                current_dll_path as usize as *mut _,
-                &mut module,
-            );
-            if r == 0 {
-                info!("GetModuleHandleExW failed: {}", io::Error::last_os_error());
-                return None;
-            }
-            let mut space = Vec::with_capacity(1024);
-            let r = GetModuleFileNameW(module, space.as_mut_ptr(), space.capacity() as u32);
-            if r == 0 {
-                info!("GetModuleFileNameW failed: {}", io::Error::last_os_error());
-                return None;
-            }
-            let r = r as usize;
-            if r >= space.capacity() {
-                info!("our buffer was too small? {}", io::Error::last_os_error());
-                return None;
-            }
-            space.set_len(r);
-            let os = OsString::from_wide(&space);
-            Some(PathBuf::from(os))
-        }
-    }
-}
-
 fn get_codegen_sysroot(maybe_sysroot: &Option<PathBuf>, backend_name: &str) -> MakeBackendFn {
     // For now we only allow this function to be called once as it'll dlopen a
     // few things, which seems to work best if we only do that once. In

compiler/rustc_session/Cargo.toml

@@ -17,3 +17,10 @@ rustc_span = { path = "../rustc_span" }
 rustc_fs_util = { path = "../rustc_fs_util" }
 rustc_ast = { path = "../rustc_ast" }
 rustc_lint_defs = { path = "../rustc_lint_defs" }
+smallvec = "1.8.1"
+
+[target.'cfg(unix)'.dependencies]
+libc = "0.2"
+
+[target.'cfg(windows)'.dependencies]
+winapi = { version = "0.3", features = ["libloaderapi"] }

compiler/rustc_session/src/config.rs

@@ -2431,7 +2431,7 @@ pub fn build_session_options(matches: &getopts::Matches) -> Options {
     let sysroot = match &sysroot_opt {
         Some(s) => s,
         None => {
-            tmp_buf = crate::filesearch::get_or_default_sysroot();
+            tmp_buf = crate::filesearch::get_or_default_sysroot().expect("Failed finding sysroot");
             &tmp_buf
         }
     };

compiler/rustc_session/src/filesearch.rs

@@ -1,12 +1,12 @@
 //! A module for searching for libraries
 
+use smallvec::{smallvec, SmallVec};
 use std::env;
 use std::fs;
 use std::iter::FromIterator;
 use std::path::{Path, PathBuf};
 
 use crate::search_paths::{PathKind, SearchPath};
-use rustc_fs_util::fix_windows_verbatim_for_gcc;
 
 #[derive(Copy, Clone)]
 pub enum FileMatch {
@@ -62,29 +62,126 @@ pub fn make_target_lib_path(sysroot: &Path, target_triple: &str) -> PathBuf {
     PathBuf::from_iter([sysroot, Path::new(&rustlib_path), Path::new("lib")])
 }
 
-/// This function checks if sysroot is found using env::args().next(), and if it
-/// is not found, uses env::current_exe() to imply sysroot.
-pub fn get_or_default_sysroot() -> PathBuf {
-    // Follow symlinks.  If the resolved path is relative, make it absolute.
-    fn canonicalize(path: PathBuf) -> PathBuf {
-        let path = fs::canonicalize(&path).unwrap_or(path);
-        // See comments on this target function, but the gist is that
-        // gcc chokes on verbatim paths which fs::canonicalize generates
-        // so we try to avoid those kinds of paths.
-        fix_windows_verbatim_for_gcc(&path)
+#[cfg(unix)]
+fn current_dll_path() -> Result<PathBuf, String> {
+    use std::ffi::{CStr, OsStr};
+    use std::os::unix::prelude::*;
+
+    unsafe {
+        let addr = current_dll_path as usize as *mut _;
+        let mut info = std::mem::zeroed();
+        if libc::dladdr(addr, &mut info) == 0 {
+            return Err("dladdr failed".into());
+        }
+        if info.dli_fname.is_null() {
+            return Err("dladdr returned null pointer".into());
+        }
+        let bytes = CStr::from_ptr(info.dli_fname).to_bytes();
+        let os = OsStr::from_bytes(bytes);
+        Ok(PathBuf::from(os))
     }
+}
 
-    // Use env::current_exe() to get the path of the executable following
-    // symlinks/canonicalizing components.
-    fn from_current_exe() -> PathBuf {
-        match env::current_exe() {
-            Ok(exe) => {
-                let mut p = canonicalize(exe);
-                p.pop();
-                p.pop();
-                p
+#[cfg(windows)]
+fn current_dll_path() -> Result<PathBuf, String> {
+    use std::ffi::OsString;
+    use std::io;
+    use std::os::windows::prelude::*;
+    use std::ptr;
+
+    use winapi::um::libloaderapi::{
+        GetModuleFileNameW, GetModuleHandleExW, GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS,
+    };
+
+    unsafe {
+        let mut module = ptr::null_mut();
+        let r = GetModuleHandleExW(
+            GET_MODULE_HANDLE_EX_FLAG_FROM_ADDRESS,
+            current_dll_path as usize as *mut _,
+            &mut module,
+        );
+        if r == 0 {
+            return Err(format!("GetModuleHandleExW failed: {}", io::Error::last_os_error()));
+        }
+        let mut space = Vec::with_capacity(1024);
+        let r = GetModuleFileNameW(module, space.as_mut_ptr(), space.capacity() as u32);
+        if r == 0 {
+            return Err(format!("GetModuleFileNameW failed: {}", io::Error::last_os_error()));
+        }
+        let r = r as usize;
+        if r >= space.capacity() {
+            return Err(format!("our buffer was too small? {}", io::Error::last_os_error()));
+        }
+        space.set_len(r);
+        let os = OsString::from_wide(&space);
+        Ok(PathBuf::from(os))
+    }
+}
+
+pub fn sysroot_candidates() -> SmallVec<[PathBuf; 2]> {
+    let target = crate::config::host_triple();
+    let mut sysroot_candidates: SmallVec<[PathBuf; 2]> =
+        smallvec![get_or_default_sysroot().expect("Failed finding sysroot")];
+    let path = current_dll_path().and_then(|s| Ok(s.canonicalize().map_err(|e| e.to_string())?));
+    if let Ok(dll) = path {
+        // use `parent` twice to chop off the file name and then also the
+        // directory containing the dll which should be either `lib` or `bin`.
+        if let Some(path) = dll.parent().and_then(|p| p.parent()) {
+            // The original `path` pointed at the `rustc_driver` crate's dll.
+            // Now that dll should only be in one of two locations. The first is
+            // in the compiler's libdir, for example `$sysroot/lib/*.dll`. The
+            // other is the target's libdir, for example
+            // `$sysroot/lib/rustlib/$target/lib/*.dll`.
+            //
+            // We don't know which, so let's assume that if our `path` above
+            // ends in `$target` we *could* be in the target libdir, and always
+            // assume that we may be in the main libdir.
+            sysroot_candidates.push(path.to_owned());
+
+            if path.ends_with(target) {
+                sysroot_candidates.extend(
+                    path.parent() // chop off `$target`
+                        .and_then(|p| p.parent()) // chop off `rustlib`
+                        .and_then(|p| p.parent()) // chop off `lib`
+                        .map(|s| s.to_owned()),
+                );
             }
-            Err(e) => panic!("failed to get current_exe: {e}"),
+        }
+    }
+
+    return sysroot_candidates;
+}
+
+/// This function checks if sysroot is found using env::args().next(), and if it
+/// is not found, finds sysroot from current rustc_driver dll.
+pub fn get_or_default_sysroot() -> Result<PathBuf, String> {
+    fn default_from_rustc_driver_dll() -> Result<PathBuf, String> {
+        let dll =
+            current_dll_path().and_then(|s| Ok(s.canonicalize().map_err(|e| e.to_string())?))?;
+
+        // `dll` will be in one of the following two:
+        // - compiler's libdir: $sysroot/lib/*.dll
+        // - target's libdir: $sysroot/lib/rustlib/$target/lib/*.dll
+        //
+        // use `parent` twice to chop off the file name and then also the
+        // directory containing the dll
+        let dir = dll.parent().and_then(|p| p.parent()).ok_or(format!(
+            "Could not move 2 levels upper using `parent()` on {}",
+            dll.display()
+        ))?;
+
+        // if `dir` points target's dir, move up to the sysroot
+        if dir.ends_with(crate::config::host_triple()) {
+            dir.parent() // chop off `$target`
+                .and_then(|p| p.parent()) // chop off `rustlib`
+                .and_then(|p| p.parent()) // chop off `lib`
+                .map(|s| s.to_owned())
+                .ok_or(format!(
+                    "Could not move 3 levels upper using `parent()` on {}",
+                    dir.display()
+                ))
+        } else {
+            Ok(dir.to_owned())
         }
     }
 
@@ -118,7 +215,5 @@ pub fn get_or_default_sysroot() -> PathBuf {
         }
     }
 
-    // Check if sysroot is found using env::args().next(), and if is not found,
-    // use env::current_exe() to imply sysroot.
-    from_env_args_next().unwrap_or_else(from_current_exe)
+    Ok(from_env_args_next().unwrap_or(default_from_rustc_driver_dll()?))
 }

compiler/rustc_session/src/session.rs

@@ -1280,7 +1280,7 @@ pub fn build_session(
 
     let sysroot = match &sopts.maybe_sysroot {
         Some(sysroot) => sysroot.clone(),
-        None => filesearch::get_or_default_sysroot(),
+        None => filesearch::get_or_default_sysroot().expect("Failed finding sysroot"),
     };
 
     let target_cfg = config::build_target_config(&sopts, target_override, &sysroot);