Move FASYNC bit handling to f_op->fasync()

Removing the BKL from FASYNC handling ran into the challenge of keeping the
setting of the FASYNC bit in filp->f_flags atomic with regard to calls to
the underlying fasync() function.  Andi Kleen suggested moving the handling
of that bit into fasync(); this patch does exactly that.  As a result, we
have a couple of internal API changes: fasync() must now manage the FASYNC
bit, and it will be called without the BKL held.

As it happens, every fasync() implementation in the kernel with one
exception calls fasync_helper().  So, if we make fasync_helper() set the
FASYNC bit, we can avoid making any changes to the other fasync()
functions - as long as those functions, themselves, have proper locking.
Most fasync() implementations do nothing but call fasync_helper() - which
has its own lock - so they are easily verified as correct.  The BKL had
already been pushed down into the rest.

The networking code has its own version of fasync_helper(), so that code
has been augmented with explicit FASYNC bit handling.

Cc: Al Viro <viro@ZenIV.linux.org.uk>
Cc: David Miller <davem@davemloft.net>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>

1 files changed, 16 insertions, 13 deletions

diff --git a/fs/fcntl.c b/fs/fcntl.c
index 04df8570a2d2..431bb6459273 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -141,7 +141,7 @@ SYSCALL_DEFINE1(dup, unsigned int, fildes)
         return ret;
 }
 
-#define SETFL_MASK (O_APPEND | O_NONBLOCK | O_NDELAY | FASYNC | O_DIRECT | O_NOATIME)
+#define SETFL_MASK (O_APPEND | O_NONBLOCK | O_NDELAY | O_DIRECT | O_NOATIME)
 
 static int setfl(int fd, struct file * filp, unsigned long arg)
 {
@@ -177,23 +177,19 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
                 return error;
 
         /*
-         * We still need a lock here for now to keep multiple FASYNC calls
-         * from racing with each other.
+         * ->fasync() is responsible for setting the FASYNC bit.
          */
-        lock_kernel();
-        if ((arg ^ filp->f_flags) & FASYNC) {
-                if (filp->f_op && filp->f_op->fasync) {
-                        error = filp->f_op->fasync(fd, filp, (arg & FASYNC) != 0);
-                        if (error < 0)
-                                goto out;
-                }
+        if (((arg ^ filp->f_flags) & FASYNC) && filp->f_op &&
+                        filp->f_op->fasync) {
+                error = filp->f_op->fasync(fd, filp, (arg & FASYNC) != 0);
+                if (error < 0)
+                        goto out;
         }
-
         spin_lock(&filp->f_lock);
         filp->f_flags = (arg & SETFL_MASK) | (filp->f_flags & ~SETFL_MASK);
         spin_unlock(&filp->f_lock);
+
  out:
-        unlock_kernel();
         return error;
 }
 
@@ -518,7 +514,7 @@ static DEFINE_RWLOCK(fasync_lock);
 static struct kmem_cache *fasync_cache __read_mostly;
 
 /*
- * fasync_helper() is used by some character device drivers (mainly mice)
+ * fasync_helper() is used by almost all character device drivers
  * to set up the fasync queue. It returns negative on error, 0 if it did
  * no changes and positive if it added/deleted the entry.
  */
@@ -557,6 +553,13 @@ int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fap
                 result = 1;
         }
 out:
+        /* Fix up FASYNC bit while still holding fasync_lock */
+        spin_lock(&filp->f_lock);
+        if (on)
+                filp->f_flags |= FASYNC;
+        else
+                filp->f_flags &= ~FASYNC;
+        spin_unlock(&filp->f_lock);
         write_unlock_irq(&fasync_lock);
         return result;
 }