Auto merge of rust-lang#35718 - michaelwoerister:incr-comp-dir-locking, r=alexcrichton

Implement synchronization scheme for incr. comp. directory

This PR implements a copy-on-write-based synchronization scheme for the incremental compilation cache directory. For technical details, see the documentation at the beginning of `rustc_incremental/persist/fs.rs`.

The PR contains unit tests for some functions but for testing whether the scheme properly handles races, a more elaborate test setup would be needed. It would probably involve a small tool that allows to manipulate the incremental compilation directory in a controlled way and then letting a compiler instance run against directories in different states. I don't know if it's worth the trouble of adding another test category to `compiletest`, but I'd be happy to do so.

Fixes rust-lang#32754
Fixes rust-lang#34957

Author: bors

mk/crates.mk

@@ -112,7 +112,7 @@ DEPS_rustc := syntax fmt_macros flate arena serialize getopts rbml \
 	      rustc_const_math syntax_pos rustc_errors
 DEPS_rustc_back := std syntax flate log libc
 DEPS_rustc_borrowck := rustc log graphviz syntax syntax_pos rustc_errors rustc_mir
-DEPS_rustc_data_structures := std log serialize
+DEPS_rustc_data_structures := std log serialize libc
 DEPS_rustc_driver := arena flate getopts graphviz libc rustc rustc_back rustc_borrowck \
                      rustc_typeck rustc_mir rustc_resolve log syntax serialize rustc_llvm \
                      rustc_trans rustc_privacy rustc_lint rustc_plugin \
@@ -137,9 +137,8 @@ DEPS_rustc_save_analysis := rustc log syntax syntax_pos serialize
 DEPS_rustc_typeck := rustc syntax syntax_pos rustc_platform_intrinsics rustc_const_math \
                      rustc_const_eval rustc_errors
 
-DEPS_rustdoc := rustc rustc_driver native:hoedown serialize getopts \
-                test rustc_lint rustc_const_eval syntax_pos
-
+DEPS_rustdoc := rustc rustc_driver native:hoedown serialize getopts test \
+                rustc_lint rustc_const_eval syntax_pos rustc_data_structures
 
 TOOL_DEPS_compiletest := test getopts log serialize
 TOOL_DEPS_rustdoc := rustdoc

src/librustc/hir/svh.rs

@@ -17,6 +17,7 @@
 
 use std::fmt;
 use std::hash::{Hash, Hasher};
+use serialize::{Encodable, Decodable, Encoder, Decoder};
 
 #[derive(Copy, Clone, PartialEq, Eq, Debug)]
 pub struct Svh {
@@ -51,3 +52,17 @@ impl fmt::Display for Svh {
         f.pad(&self.to_string())
     }
 }
+
+impl Encodable for Svh {
+    fn encode<S: Encoder>(&self, s: &mut S) -> Result<(), S::Error> {
+        s.emit_u64(self.as_u64().to_le())
+    }
+}
+
+impl Decodable for Svh {
+    fn decode<D: Decoder>(d: &mut D) -> Result<Svh, D::Error> {
+        d.read_u64()
+         .map(u64::from_le)
+         .map(Svh::new)
+    }
+}

src/librustc/session/mod.rs

@@ -33,10 +33,11 @@ use syntax::feature_gate::AttributeType;
 use syntax_pos::{Span, MultiSpan};
 
 use rustc_back::target::Target;
+use rustc_data_structures::flock;
 use llvm;
 
 use std::path::{Path, PathBuf};
-use std::cell::{Cell, RefCell};
+use std::cell::{self, Cell, RefCell};
 use std::collections::{HashMap, HashSet};
 use std::env;
 use std::ffi::CString;
@@ -101,6 +102,8 @@ pub struct Session {
     /// macro name and defintion span in the source crate.
     pub imported_macro_spans: RefCell<HashMap<Span, (String, Span)>>,
 
+    incr_comp_session: RefCell<IncrCompSession>,
+
     next_node_id: Cell<ast::NodeId>,
 }
 
@@ -331,6 +334,76 @@ impl Session {
             &self.opts.search_paths,
             kind)
     }
+
+    pub fn init_incr_comp_session(&self,
+                                  session_dir: PathBuf,
+                                  lock_file: flock::Lock) {
+        let mut incr_comp_session = self.incr_comp_session.borrow_mut();
+
+        if let IncrCompSession::NotInitialized = *incr_comp_session { } else {
+            bug!("Trying to initialize IncrCompSession `{:?}`", *incr_comp_session)
+        }
+
+        *incr_comp_session = IncrCompSession::Active {
+            session_directory: session_dir,
+            lock_file: lock_file,
+        };
+    }
+
+    pub fn finalize_incr_comp_session(&self, new_directory_path: PathBuf) {
+        let mut incr_comp_session = self.incr_comp_session.borrow_mut();
+
+        if let IncrCompSession::Active { .. } = *incr_comp_session { } else {
+            bug!("Trying to finalize IncrCompSession `{:?}`", *incr_comp_session)
+        }
+
+        // Note: This will also drop the lock file, thus unlocking the directory
+        *incr_comp_session = IncrCompSession::Finalized {
+            session_directory: new_directory_path,
+        };
+    }
+
+    pub fn mark_incr_comp_session_as_invalid(&self) {
+        let mut incr_comp_session = self.incr_comp_session.borrow_mut();
+
+        let session_directory = match *incr_comp_session {
+            IncrCompSession::Active { ref session_directory, .. } => {
+                session_directory.clone()
+            }
+            _ => bug!("Trying to invalidate IncrCompSession `{:?}`",
+                      *incr_comp_session),
+        };
+
+        // Note: This will also drop the lock file, thus unlocking the directory
+        *incr_comp_session = IncrCompSession::InvalidBecauseOfErrors {
+            session_directory: session_directory
+        };
+    }
+
+    pub fn incr_comp_session_dir(&self) -> cell::Ref<PathBuf> {
+        let incr_comp_session = self.incr_comp_session.borrow();
+        cell::Ref::map(incr_comp_session, |incr_comp_session| {
+            match *incr_comp_session {
+                IncrCompSession::NotInitialized => {
+                    bug!("Trying to get session directory from IncrCompSession `{:?}`",
+                        *incr_comp_session)
+                }
+                IncrCompSession::Active { ref session_directory, .. } |
+                IncrCompSession::Finalized { ref session_directory } |
+                IncrCompSession::InvalidBecauseOfErrors { ref session_directory } => {
+                    session_directory
+                }
+            }
+        })
+    }
+
+    pub fn incr_comp_session_dir_opt(&self) -> Option<cell::Ref<PathBuf>> {
+        if self.opts.incremental.is_some() {
+            Some(self.incr_comp_session_dir())
+        } else {
+            None
+        }
+    }
 }
 
 pub fn build_session(sopts: config::Options,
@@ -446,13 +519,39 @@ pub fn build_session_(sopts: config::Options,
         injected_panic_runtime: Cell::new(None),
         available_macros: RefCell::new(HashSet::new()),
         imported_macro_spans: RefCell::new(HashMap::new()),
+        incr_comp_session: RefCell::new(IncrCompSession::NotInitialized),
     };
 
     init_llvm(&sess);
 
     sess
 }
 
+/// Holds data on the current incremental compilation session, if there is one.
+#[derive(Debug)]
+pub enum IncrCompSession {
+    // This is the state the session will be in until the incr. comp. dir is
+    // needed.
+    NotInitialized,
+    // This is the state during which the session directory is private and can
+    // be modified.
+    Active {
+        session_directory: PathBuf,
+        lock_file: flock::Lock,
+    },
+    // This is the state after the session directory has been finalized. In this
+    // state, the contents of the directory must not be modified any more.
+    Finalized {
+        session_directory: PathBuf,
+    },
+    // This is an error state that is reached when some compilation error has
+    // occurred. It indicates that the contents of the session directory must
+    // not be used, since they might be invalid.
+    InvalidBecauseOfErrors {
+        session_directory: PathBuf,
+    }
+}
+
 fn init_llvm(sess: &Session) {
     unsafe {
         // Before we touch LLVM, make sure that multithreading is enabled.

src/librustc/util/fs.rs

@@ -56,14 +56,49 @@ pub fn fix_windows_verbatim_for_gcc(p: &Path) -> PathBuf {
     }
 }
 
+pub enum LinkOrCopy {
+    Link,
+    Copy
+}
+
 /// Copy `p` into `q`, preferring to use hard-linking if possible. If
 /// `q` already exists, it is removed first.
-pub fn link_or_copy<P: AsRef<Path>, Q: AsRef<Path>>(p: P, q: Q) -> io::Result<()> {
+/// The result indicates which of the two operations has been performed.
+pub fn link_or_copy<P: AsRef<Path>, Q: AsRef<Path>>(p: P, q: Q) -> io::Result<LinkOrCopy> {
     let p = p.as_ref();
     let q = q.as_ref();
     if q.exists() {
         try!(fs::remove_file(&q));
     }
-    fs::hard_link(p, q)
-        .or_else(|_| fs::copy(p, q).map(|_| ()))
+
+    match fs::hard_link(p, q) {
+        Ok(()) => Ok(LinkOrCopy::Link),
+        Err(_) => {
+            match fs::copy(p, q) {
+                Ok(_) => Ok(LinkOrCopy::Copy),
+                Err(e) => Err(e)
+            }
+        }
+    }
+}
+
+// Like std::fs::create_dir_all, except handles concurrent calls among multiple
+// threads or processes.
+pub fn create_dir_racy(path: &Path) -> io::Result<()> {
+    match fs::create_dir(path) {
+        Ok(()) => return Ok(()),
+        Err(ref e) if e.kind() == io::ErrorKind::AlreadyExists => return Ok(()),
+        Err(ref e) if e.kind() == io::ErrorKind::NotFound => {}
+        Err(e) => return Err(e),
+    }
+    match path.parent() {
+        Some(p) => try!(create_dir_racy(p)),
+        None => return Err(io::Error::new(io::ErrorKind::Other,
+                                          "failed to create whole tree")),
+    }
+    match fs::create_dir(path) {
+        Ok(()) => Ok(()),
+        Err(ref e) if e.kind() == io::ErrorKind::AlreadyExists => Ok(()),
+        Err(e) => Err(e),
+    }
 }