Check for existence before permissions

Return EEXIST when overwritting a file as long as the caller has exec
permission on the parent directory, even if the caller doesn't have
write permission.

Also reordered the mount write check, which happens before permission
is checked.

Closes #5164

PiperOrigin-RevId: 351868123

5 files changed, 91 insertions, 57 deletions

diff --git a/pkg/sentry/fsimpl/gofer/filesystem.go b/pkg/sentry/fsimpl/gofer/filesystem.go
index df27554d3..91d5dc174 100644
--- a/pkg/sentry/fsimpl/gofer/filesystem.go
+++ b/pkg/sentry/fsimpl/gofer/filesystem.go
@@ -407,33 +407,44 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 	if err != nil {
 		return err
 	}
-	if err := parent.checkPermissions(rp.Credentials(), vfs.MayWrite|vfs.MayExec); err != nil {
+
+	// Order of checks is important. First check if parent directory can be
+	// executed, then check for existence, and lastly check if mount is writable.
+	if err := parent.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 		return err
 	}
 	name := rp.Component()
 	if name == "." || name == ".." {
 		return syserror.EEXIST
 	}
-	if len(name) > maxFilenameLen {
-		return syserror.ENAMETOOLONG
-	}
 	if parent.isDeleted() {
 		return syserror.ENOENT
 	}
+
+	parent.dirMu.Lock()
+	defer parent.dirMu.Unlock()
+
+	child, err := fs.getChildLocked(ctx, rp.VirtualFilesystem(), parent, name, &ds)
+	switch {
+	case err != nil && err != syserror.ENOENT:
+		return err
+	case child != nil:
+		return syserror.EEXIST
+	}
+
 	mnt := rp.Mount()
 	if err := mnt.CheckBeginWrite(); err != nil {
 		return err
 	}
 	defer mnt.EndWrite()
-	parent.dirMu.Lock()
-	defer parent.dirMu.Unlock()
+
+	if err := parent.checkPermissions(rp.Credentials(), vfs.MayWrite); err != nil {
+		return err
+	}
+	if !dir && rp.MustBeDir() {
+		return syserror.ENOENT
+	}
 	if parent.isSynthetic() {
-		if child := parent.children[name]; child != nil {
-			return syserror.EEXIST
-		}
-		if !dir && rp.MustBeDir() {
-			return syserror.ENOENT
-		}
 		if createInSyntheticDir == nil {
 			return syserror.EPERM
 		}
@@ -449,47 +460,20 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 		parent.watches.Notify(ctx, name, uint32(ev), 0, vfs.InodeEvent, false /* unlinked */)
 		return nil
 	}
-	if fs.opts.interop == InteropModeShared {
-		if child := parent.children[name]; child != nil && child.isSynthetic() {
-			return syserror.EEXIST
-		}
-		if !dir && rp.MustBeDir() {
-			return syserror.ENOENT
-		}
-		// The existence of a non-synthetic dentry at name would be inconclusive
-		// because the file it represents may have been deleted from the remote
-		// filesystem, so we would need to make an RPC to revalidate the dentry.
-		// Just attempt the file creation RPC instead. If a file does exist, the
-		// RPC will fail with EEXIST like we would have. If the RPC succeeds, and a
-		// stale dentry exists, the dentry will fail revalidation next time it's
-		// used.
-		if err := createInRemoteDir(parent, name, &ds); err != nil {
-			return err
-		}
-		ev := linux.IN_CREATE
-		if dir {
-			ev |= linux.IN_ISDIR
-		}
-		parent.watches.Notify(ctx, name, uint32(ev), 0, vfs.InodeEvent, false /* unlinked */)
-		return nil
-	}
-	if child := parent.children[name]; child != nil {
-		return syserror.EEXIST
-	}
-	if !dir && rp.MustBeDir() {
-		return syserror.ENOENT
-	}
-	// No cached dentry exists; however, there might still be an existing file
-	// at name. As above, we attempt the file creation RPC anyway.
+	// No cached dentry exists; however, in InteropModeShared there might still be
+	// an existing file at name. Just attempt the file creation RPC anyways. If a
+	// file does exist, the RPC will fail with EEXIST like we would have.
 	if err := createInRemoteDir(parent, name, &ds); err != nil {
 		return err
 	}
-	if child, ok := parent.children[name]; ok && child == nil {
-		// Delete the now-stale negative dentry.
-		delete(parent.children, name)
+	if fs.opts.interop != InteropModeShared {
+		if child, ok := parent.children[name]; ok && child == nil {
+			// Delete the now-stale negative dentry.
+			delete(parent.children, name)
+		}
+		parent.touchCMtime()
+		parent.dirents = nil
 	}
-	parent.touchCMtime()
-	parent.dirents = nil
 	ev := linux.IN_CREATE
 	if dir {
 		ev |= linux.IN_ISDIR
diff --git a/pkg/sentry/fsimpl/kernfs/filesystem.go b/pkg/sentry/fsimpl/kernfs/filesystem.go
index e77523f22..a7a553619 100644
--- a/pkg/sentry/fsimpl/kernfs/filesystem.go
+++ b/pkg/sentry/fsimpl/kernfs/filesystem.go
@@ -208,7 +208,9 @@ func (fs *Filesystem) walkParentDirLocked(ctx context.Context, rp *vfs.Resolving
 // * Filesystem.mu must be locked for at least reading.
 // * isDir(parentInode) == true.
 func checkCreateLocked(ctx context.Context, creds *auth.Credentials, name string, parent *Dentry) error {
-	if err := parent.inode.CheckPermissions(ctx, creds, vfs.MayWrite|vfs.MayExec); err != nil {
+	// Order of checks is important. First check if parent directory can be
+	// executed, then check for existence, and lastly check if mount is writable.
+	if err := parent.inode.CheckPermissions(ctx, creds, vfs.MayExec); err != nil {
 		return err
 	}
 	if name == "." || name == ".." {
@@ -223,6 +225,9 @@ func checkCreateLocked(ctx context.Context, creds *auth.Credentials, name string
 	if parent.VFSDentry().IsDead() {
 		return syserror.ENOENT
 	}
+	if err := parent.inode.CheckPermissions(ctx, creds, vfs.MayWrite); err != nil {
+		return err
+	}
 	return nil
 }
 
diff --git a/pkg/sentry/fsimpl/overlay/filesystem.go b/pkg/sentry/fsimpl/overlay/filesystem.go
index d55bdc97f..e46f593c7 100644
--- a/pkg/sentry/fsimpl/overlay/filesystem.go
+++ b/pkg/sentry/fsimpl/overlay/filesystem.go
@@ -480,9 +480,6 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 	if err != nil {
 		return err
 	}
-	if err := parent.checkPermissions(rp.Credentials(), vfs.MayWrite|vfs.MayExec); err != nil {
-		return err
-	}
 	name := rp.Component()
 	if name == "." || name == ".." {
 		return syserror.EEXIST
@@ -490,11 +487,11 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 	if parent.vfsd.IsDead() {
 		return syserror.ENOENT
 	}
-	mnt := rp.Mount()
-	if err := mnt.CheckBeginWrite(); err != nil {
+
+	if err := parent.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 		return err
 	}
-	defer mnt.EndWrite()
+
 	parent.dirMu.Lock()
 	defer parent.dirMu.Unlock()
 
@@ -514,6 +511,14 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 		return syserror.ENOENT
 	}
 
+	mnt := rp.Mount()
+	if err := mnt.CheckBeginWrite(); err != nil {
+		return err
+	}
+	defer mnt.EndWrite()
+	if err := parent.checkPermissions(rp.Credentials(), vfs.MayWrite|vfs.MayExec); err != nil {
+		return err
+	}
 	// Ensure that the parent directory is copied-up so that we can create the
 	// new file in the upper layer.
 	if err := parent.copyUpLocked(ctx); err != nil {
diff --git a/pkg/sentry/fsimpl/tmpfs/filesystem.go b/pkg/sentry/fsimpl/tmpfs/filesystem.go
index 9296db2fb..453e41d11 100644
--- a/pkg/sentry/fsimpl/tmpfs/filesystem.go
+++ b/pkg/sentry/fsimpl/tmpfs/filesystem.go
@@ -153,7 +153,10 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 	if err != nil {
 		return err
 	}
-	if err := parentDir.inode.checkPermissions(rp.Credentials(), vfs.MayWrite|vfs.MayExec); err != nil {
+
+	// Order of checks is important. First check if parent directory can be
+	// executed, then check for existence, and lastly check if mount is writable.
+	if err := parentDir.inode.checkPermissions(rp.Credentials(), vfs.MayExec); err != nil {
 		return err
 	}
 	name := rp.Component()
@@ -179,6 +182,10 @@ func (fs *filesystem) doCreateAt(ctx context.Context, rp *vfs.ResolvingPath, dir
 		return err
 	}
 	defer mnt.EndWrite()
+
+	if err := parentDir.inode.checkPermissions(rp.Credentials(), vfs.MayWrite); err != nil {
+		return err
+	}
 	if err := create(parentDir, name); err != nil {
 		return err
 	}
diff --git a/test/syscalls/linux/mkdir.cc b/test/syscalls/linux/mkdir.cc
index 27758203d..11fbfa5c5 100644
--- a/test/syscalls/linux/mkdir.cc
+++ b/test/syscalls/linux/mkdir.cc
@@ -82,6 +82,39 @@ TEST_F(MkdirTest, FailsOnDirWithoutWritePerms) {
               SyscallFailsWithErrno(EACCES));
 }
 
+TEST_F(MkdirTest, DirAlreadyExists) {
+  // Drop capabilities that allow us to override file and directory permissions.
+  ASSERT_NO_ERRNO(SetCapability(CAP_DAC_OVERRIDE, false));
+  ASSERT_NO_ERRNO(SetCapability(CAP_DAC_READ_SEARCH, false));
+
+  ASSERT_THAT(mkdir(dirname_.c_str(), 0777), SyscallSucceeds());
+  auto dir = JoinPath(dirname_.c_str(), "foo");
+  EXPECT_THAT(mkdir(dir.c_str(), 0777), SyscallSucceeds());
+
+  struct {
+    int mode;
+    int err;
+  } tests[] = {
+      {.mode = 0000, .err = EACCES},  // No perm
+      {.mode = 0100, .err = EEXIST},  // Exec only
+      {.mode = 0200, .err = EACCES},  // Write only
+      {.mode = 0300, .err = EEXIST},  // Write+exec
+      {.mode = 0400, .err = EACCES},  // Read only
+      {.mode = 0500, .err = EEXIST},  // Read+exec
+      {.mode = 0600, .err = EACCES},  // Read+write
+      {.mode = 0700, .err = EEXIST},  // All
+  };
+  for (const auto& t : tests) {
+    printf("mode: 0%o\n", t.mode);
+    EXPECT_THAT(chmod(dirname_.c_str(), t.mode), SyscallSucceeds());
+    EXPECT_THAT(mkdir(dir.c_str(), 0777), SyscallFailsWithErrno(t.err));
+  }
+
+  // Clean up.
+  EXPECT_THAT(chmod(dirname_.c_str(), 0777), SyscallSucceeds());
+  ASSERT_THAT(rmdir(dir.c_str()), SyscallSucceeds());
+}
+
 TEST_F(MkdirTest, MkdirAtEmptyPath) {
   ASSERT_THAT(mkdir(dirname_.c_str(), 0777), SyscallSucceeds());
   auto fd =