[clang][deps] Cache `VFS::getRealPath()` (#68645)

This PR starts caching calls to
`DependencyScanningWorkerFilesystem::getRealPath()` that we use whenever
we canonicalize module map path. In the case of the real VFS, this
functions performs an expensive syscall that we'd like to do as rarely
as possible.

This PR keeps the real path out of `CachedFileSystemEntry`, since that's
**immutable**; populating the real path on creation of this data
structure (every stat/open) would be expensive.

1 files changed, 98 insertions, 11 deletions

diff --git a/clang/lib/Tooling/DependencyScanning/DependencyScanningFilesystem.cpp b/clang/lib/Tooling/DependencyScanning/DependencyScanningFilesystem.cpp
index c66780d..84185c9 100644
--- a/clang/lib/Tooling/DependencyScanning/DependencyScanningFilesystem.cpp
+++ b/clang/lib/Tooling/DependencyScanning/DependencyScanningFilesystem.cpp
@@ -113,8 +113,8 @@ DependencyScanningFilesystemSharedCache::CacheShard::findEntryByFilename(
     StringRef Filename) const {
   assert(llvm::sys::path::is_absolute_gnu(Filename));
   std::lock_guard<std::mutex> LockGuard(CacheLock);
-  auto It = EntriesByFilename.find(Filename);
-  return It == EntriesByFilename.end() ? nullptr : It->getValue();
+  auto It = CacheByFilename.find(Filename);
+  return It == CacheByFilename.end() ? nullptr : It->getValue().first;
 }
 
 const CachedFileSystemEntry *
@@ -130,11 +130,16 @@ DependencyScanningFilesystemSharedCache::CacheShard::
     getOrEmplaceEntryForFilename(StringRef Filename,
                                  llvm::ErrorOr<llvm::vfs::Status> Stat) {
   std::lock_guard<std::mutex> LockGuard(CacheLock);
-  auto Insertion = EntriesByFilename.insert({Filename, nullptr});
-  if (Insertion.second)
-    Insertion.first->second =
+  auto [It, Inserted] = CacheByFilename.insert({Filename, {nullptr, nullptr}});
+  auto &[CachedEntry, CachedRealPath] = It->getValue();
+  if (!CachedEntry) {
+    // The entry is not present in the shared cache. Either the cache doesn't
+    // know about the file at all, or it only knows about its real path.
+    assert((Inserted || CachedRealPath) && "existing file with empty pair");
+    CachedEntry =
         new (EntryStorage.Allocate()) CachedFileSystemEntry(std::move(Stat));
-  return *Insertion.first->second;
+  }
+  return *CachedEntry;
 }
 
 const CachedFileSystemEntry &
@@ -142,16 +147,17 @@ DependencyScanningFilesystemSharedCache::CacheShard::getOrEmplaceEntryForUID(
     llvm::sys::fs::UniqueID UID, llvm::vfs::Status Stat,
     std::unique_ptr<llvm::MemoryBuffer> Contents) {
   std::lock_guard<std::mutex> LockGuard(CacheLock);
-  auto Insertion = EntriesByUID.insert({UID, nullptr});
-  if (Insertion.second) {
+  auto [It, Inserted] = EntriesByUID.insert({UID, nullptr});
+  auto &CachedEntry = It->getSecond();
+  if (Inserted) {
     CachedFileContents *StoredContents = nullptr;
     if (Contents)
       StoredContents = new (ContentsStorage.Allocate())
           CachedFileContents(std::move(Contents));
-    Insertion.first->second = new (EntryStorage.Allocate())
+    CachedEntry = new (EntryStorage.Allocate())
         CachedFileSystemEntry(std::move(Stat), StoredContents);
   }
-  return *Insertion.first->second;
+  return *CachedEntry;
 }
 
 const CachedFileSystemEntry &
@@ -159,7 +165,40 @@ DependencyScanningFilesystemSharedCache::CacheShard::
     getOrInsertEntryForFilename(StringRef Filename,
                                 const CachedFileSystemEntry &Entry) {
   std::lock_guard<std::mutex> LockGuard(CacheLock);
-  return *EntriesByFilename.insert({Filename, &Entry}).first->getValue();
+  auto [It, Inserted] = CacheByFilename.insert({Filename, {&Entry, nullptr}});
+  auto &[CachedEntry, CachedRealPath] = It->getValue();
+  if (!Inserted || !CachedEntry)
+    CachedEntry = &Entry;
+  return *CachedEntry;
+}
+
+const CachedRealPath *
+DependencyScanningFilesystemSharedCache::CacheShard::findRealPathByFilename(
+    StringRef Filename) const {
+  assert(llvm::sys::path::is_absolute_gnu(Filename));
+  std::lock_guard<std::mutex> LockGuard(CacheLock);
+  auto It = CacheByFilename.find(Filename);
+  return It == CacheByFilename.end() ? nullptr : It->getValue().second;
+}
+
+const CachedRealPath &DependencyScanningFilesystemSharedCache::CacheShard::
+    getOrEmplaceRealPathForFilename(StringRef Filename,
+                                    llvm::ErrorOr<llvm::StringRef> RealPath) {
+  std::lock_guard<std::mutex> LockGuard(CacheLock);
+
+  const CachedRealPath *&StoredRealPath = CacheByFilename[Filename].second;
+  if (!StoredRealPath) {
+    auto OwnedRealPath = [&]() -> CachedRealPath {
+      if (!RealPath)
+        return RealPath.getError();
+      return RealPath->str();
+    }();
+
+    StoredRealPath = new (RealPathStorage.Allocate())
+        CachedRealPath(std::move(OwnedRealPath));
+  }
+
+  return *StoredRealPath;
 }
 
 static bool shouldCacheStatFailures(StringRef Filename) {
@@ -321,6 +360,54 @@ DependencyScanningWorkerFilesystem::openFileForRead(const Twine &Path) {
   return DepScanFile::create(Result.get());
 }
 
+std::error_code
+DependencyScanningWorkerFilesystem::getRealPath(const Twine &Path,
+                                                SmallVectorImpl<char> &Output) {
+  SmallString<256> OwnedFilename;
+  StringRef OriginalFilename = Path.toStringRef(OwnedFilename);
+
+  SmallString<256> PathBuf;
+  auto FilenameForLookup = tryGetFilenameForLookup(OriginalFilename, PathBuf);
+  if (!FilenameForLookup)
+    return FilenameForLookup.getError();
+
+  auto HandleCachedRealPath =
+      [&Output](const CachedRealPath &RealPath) -> std::error_code {
+    if (!RealPath)
+      return RealPath.getError();
+    Output.assign(RealPath->begin(), RealPath->end());
+    return {};
+  };
+
+  // If we already have the result in local cache, no work required.
+  if (const auto *RealPath =
+          LocalCache.findRealPathByFilename(*FilenameForLookup))
+    return HandleCachedRealPath(*RealPath);
+
+  // If we have the result in the shared cache, cache it locally.
+  auto &Shard = SharedCache.getShardForFilename(*FilenameForLookup);
+  if (const auto *ShardRealPath =
+          Shard.findRealPathByFilename(*FilenameForLookup)) {
+    const auto &RealPath = LocalCache.insertRealPathForFilename(
+        *FilenameForLookup, *ShardRealPath);
+    return HandleCachedRealPath(RealPath);
+  }
+
+  // If we don't know the real path, compute it...
+  std::error_code EC = getUnderlyingFS().getRealPath(OriginalFilename, Output);
+  llvm::ErrorOr<llvm::StringRef> ComputedRealPath = EC;
+  if (!EC)
+    ComputedRealPath = StringRef{Output.data(), Output.size()};
+
+  // ...and try to write it into the shared cache. In case some other thread won
+  // this race and already wrote its own result there, just adopt it. Write
+  // whatever is in the shared cache into the local one.
+  const auto &RealPath = Shard.getOrEmplaceRealPathForFilename(
+      *FilenameForLookup, ComputedRealPath);
+  return HandleCachedRealPath(
+      LocalCache.insertRealPathForFilename(*FilenameForLookup, RealPath));
+}
+
 std::error_code DependencyScanningWorkerFilesystem::setCurrentWorkingDirectory(
     const Twine &Path) {
   std::error_code EC = ProxyFileSystem::setCurrentWorkingDirectory(Path);