vfs: atomic f_pos accesses as per POSIX

Our write() system call has always been atomic in the sense that you get
the expected thread-safe contiguous write, but we haven't actually
guaranteed that concurrent writes are serialized wrt f_pos accesses, so
threads (or processes) that share a file descriptor and use "write()"
concurrently would quite likely overwrite each others data.

This violates POSIX.1-2008/SUSv4 Section XSI 2.9.7 that says:

 "2.9.7 Thread Interactions with Regular File Operations

  All of the following functions shall be atomic with respect to each
  other in the effects specified in POSIX.1-2008 when they operate on
  regular files or symbolic links: [...]"

and one of the effects is the file position update.

This unprotected file position behavior is not new behavior, and nobody
has ever cared.  Until now.  Yongzhi Pan reported unexpected behavior to
Michael Kerrisk that was due to this.

This resolves the issue with a f_pos-specific lock that is taken by
read/write/lseek on file descriptors that may be shared across threads
or processes.

Reported-by: Yongzhi Pan <panyongzhi@gmail.com>
Reported-by: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

4 files changed, 46 insertions, 15 deletions

diff --git a/fs/file_table.c b/fs/file_table.c
index 5fff9030be34..5b24008ea4f6 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -135,6 +135,7 @@ struct file *get_empty_filp(void)
 	atomic_long_set(&f->f_count, 1);
 	rwlock_init(&f->f_owner.lock);
 	spin_lock_init(&f->f_lock);
+	mutex_init(&f->f_pos_lock);
 	eventpoll_init_file(f);
 	/* f->f_version: 0 */
 	return f;
diff --git a/fs/namei.c b/fs/namei.c
index 385f7817bfcc..2f730ef9b4b3 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1884,7 +1884,7 @@ static int path_init(int dfd, const char *name, unsigned int flags,
 
 		nd->path = f.file->f_path;
 		if (flags & LOOKUP_RCU) {
-			if (f.need_put)
+			if (f.flags & FDPUT_FPUT)
 				*fp = f.file;
 			nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq);
 			rcu_read_lock();
diff --git a/fs/open.c b/fs/open.c
index 4b3e1edf2fe4..b9ed8b25c108 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -705,6 +705,10 @@ static int do_dentry_open(struct file *f,
 		return 0;
 	}
 
+	/* POSIX.1-2008/SUSv4 Section XSI 2.9.7 */
+	if (S_ISREG(inode->i_mode))
+		f->f_mode |= FMODE_ATOMIC_POS;
+
 	f->f_op = fops_get(inode->i_fop);
 	if (unlikely(WARN_ON(!f->f_op))) {
 		error = -ENODEV;
diff --git a/fs/read_write.c b/fs/read_write.c
index edc5746a902a..932bb3414a96 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -264,10 +264,36 @@ loff_t vfs_llseek(struct file *file, loff_t offset, int whence)
 }
 EXPORT_SYMBOL(vfs_llseek);
 
+/*
+ * We only lock f_pos if we have threads or if the file might be
+ * shared with another process. In both cases we'll have an elevated
+ * file count (done either by fdget() or by fork()).
+ */
+static inline struct fd fdget_pos(int fd)
+{
+	struct fd f = fdget(fd);
+	struct file *file = f.file;
+
+	if (file && (file->f_mode & FMODE_ATOMIC_POS)) {
+		if (file_count(file) > 1) {
+			f.flags |= FDPUT_POS_UNLOCK;
+			mutex_lock(&file->f_pos_lock);
+		}
+	}
+	return f;
+}
+
+static inline void fdput_pos(struct fd f)
+{
+	if (f.flags & FDPUT_POS_UNLOCK)
+		mutex_unlock(&f.file->f_pos_lock);
+	fdput(f);
+}
+
 SYSCALL_DEFINE3(lseek, unsigned int, fd, off_t, offset, unsigned int, whence)
 {
 	off_t retval;
-	struct fd f = fdget(fd);
+	struct fd f = fdget_pos(fd);
 	if (!f.file)
 		return -EBADF;
 
@@ -278,7 +304,7 @@ SYSCALL_DEFINE3(lseek, unsigned int, fd, off_t, offset, unsigned int, whence)
 		if (res != (loff_t)retval)
 			retval = -EOVERFLOW;	/* LFS: should only happen on 32 bit platforms */
 	}
-	fdput(f);
+	fdput_pos(f);
 	return retval;
 }
 
@@ -498,7 +524,7 @@ static inline void file_pos_write(struct file *file, loff_t pos)
 
 SYSCALL_DEFINE3(read, unsigned int, fd, char __user *, buf, size_t, count)
 {
-	struct fd f = fdget(fd);
+	struct fd f = fdget_pos(fd);
 	ssize_t ret = -EBADF;
 
 	if (f.file) {
@@ -506,7 +532,7 @@ SYSCALL_DEFINE3(read, unsigned int, fd, char __user *, buf, size_t, count)
 		ret = vfs_read(f.file, buf, count, &pos);
 		if (ret >= 0)
 			file_pos_write(f.file, pos);
-		fdput(f);
+		fdput_pos(f);
 	}
 	return ret;
 }
@@ -514,7 +540,7 @@ SYSCALL_DEFINE3(read, unsigned int, fd, char __user *, buf, size_t, count)
 SYSCALL_DEFINE3(write, unsigned int, fd, const char __user *, buf,
 		size_t, count)
 {
-	struct fd f = fdget(fd);
+	struct fd f = fdget_pos(fd);
 	ssize_t ret = -EBADF;
 
 	if (f.file) {
@@ -522,7 +548,7 @@ SYSCALL_DEFINE3(write, unsigned int, fd, const char __user *, buf,
 		ret = vfs_write(f.file, buf, count, &pos);
 		if (ret >= 0)
 			file_pos_write(f.file, pos);
-		fdput(f);
+		fdput_pos(f);
 	}
 
 	return ret;
@@ -797,7 +823,7 @@ EXPORT_SYMBOL(vfs_writev);
 SYSCALL_DEFINE3(readv, unsigned long, fd, const struct iovec __user *, vec,
 		unsigned long, vlen)
 {
-	struct fd f = fdget(fd);
+	struct fd f = fdget_pos(fd);
 	ssize_t ret = -EBADF;
 
 	if (f.file) {
@@ -805,7 +831,7 @@ SYSCALL_DEFINE3(readv, unsigned long, fd, const struct iovec __user *, vec,
 		ret = vfs_readv(f.file, vec, vlen, &pos);
 		if (ret >= 0)
 			file_pos_write(f.file, pos);
-		fdput(f);
+		fdput_pos(f);
 	}
 
 	if (ret > 0)
@@ -817,7 +843,7 @@ SYSCALL_DEFINE3(readv, unsigned long, fd, const struct iovec __user *, vec,
 SYSCALL_DEFINE3(writev, unsigned long, fd, const struct iovec __user *, vec,
 		unsigned long, vlen)
 {
-	struct fd f = fdget(fd);
+	struct fd f = fdget_pos(fd);
 	ssize_t ret = -EBADF;
 
 	if (f.file) {
@@ -825,7 +851,7 @@ SYSCALL_DEFINE3(writev, unsigned long, fd, const struct iovec __user *, vec,
 		ret = vfs_writev(f.file, vec, vlen, &pos);
 		if (ret >= 0)
 			file_pos_write(f.file, pos);
-		fdput(f);
+		fdput_pos(f);
 	}
 
 	if (ret > 0)
@@ -968,7 +994,7 @@ COMPAT_SYSCALL_DEFINE3(readv, compat_ulong_t, fd,
 		const struct compat_iovec __user *,vec,
 		compat_ulong_t, vlen)
 {
-	struct fd f = fdget(fd);
+	struct fd f = fdget_pos(fd);
 	ssize_t ret;
 	loff_t pos;
 
@@ -978,7 +1004,7 @@ COMPAT_SYSCALL_DEFINE3(readv, compat_ulong_t, fd,
 	ret = compat_readv(f.file, vec, vlen, &pos);
 	if (ret >= 0)
 		f.file->f_pos = pos;
-	fdput(f);
+	fdput_pos(f);
 	return ret;
 }
 
@@ -1035,7 +1061,7 @@ COMPAT_SYSCALL_DEFINE3(writev, compat_ulong_t, fd,
 		const struct compat_iovec __user *, vec,
 		compat_ulong_t, vlen)
 {
-	struct fd f = fdget(fd);
+	struct fd f = fdget_pos(fd);
 	ssize_t ret;
 	loff_t pos;
 
@@ -1045,7 +1071,7 @@ COMPAT_SYSCALL_DEFINE3(writev, compat_ulong_t, fd,
 	ret = compat_writev(f.file, vec, vlen, &pos);
 	if (ret >= 0)
 		f.file->f_pos = pos;
-	fdput(f);
+	fdput_pos(f);
 	return ret;
 }