get rid of the magic around f_count in aio

__aio_put_req() plays sick games with file refcount.  What
it wants is fput() from atomic context; it's almost always
done with f_count > 1, so they only have to deal with delayed
work in rare cases when their reference happens to be the
last one.  Current code decrements f_count and if it hasn't
hit 0, everything is fine.  Otherwise it keeps a pointer
to struct file (with zero f_count!) around and has delayed
work do __fput() on it.

Better way to do it: use atomic_long_add_unless( , -1, 1)
instead of !atomic_long_dec_and_test().  IOW, decrement it
only if it's not the last reference, leave refcount alone
if it was.  And use normal fput() in delayed work.

I've made that atomic_long_add_unless call a new helper -
fput_atomic().  Drops a reference to file if it's safe to
do in atomic (i.e. if that's not the last one), tells if
it had been able to do that.  aio.c converted to it, __fput()
use is gone.  req->ki_file *always* contributes to refcount
now.  And __fput() became static.

Signed-off-by: Al Viro <[email protected]>

fs/aio.c

@@ -527,7 +527,7 @@ static void aio_fput_routine(struct work_struct *data)
 
 		/* Complete the fput(s) */
 		if (req->ki_filp != NULL)
-			__fput(req->ki_filp);
+			fput(req->ki_filp);
 
 		/* Link the iocb into the context's free list */
 		spin_lock_irq(&ctx->ctx_lock);
@@ -560,11 +560,11 @@ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req)
 
 	/*
 	 * Try to optimize the aio and eventfd file* puts, by avoiding to
-	 * schedule work in case it is not __fput() time. In normal cases,
+	 * schedule work in case it is not final fput() time. In normal cases,
 	 * we would not be holding the last reference to the file*, so
 	 * this function will be executed w/out any aio kthread wakeup.
 	 */
-	if (unlikely(atomic_long_dec_and_test(&req->ki_filp->f_count))) {
+	if (unlikely(!fput_atomic(req->ki_filp))) {
 		get_ioctx(ctx);
 		spin_lock(&fput_lock);
 		list_add(&req->ki_list, &fput_head);

fs/file_table.c

@@ -194,14 +194,6 @@ struct file *alloc_file(struct path *path, fmode_t mode,
 }
 EXPORT_SYMBOL(alloc_file);
 
-void fput(struct file *file)
-{
-	if (atomic_long_dec_and_test(&file->f_count))
-		__fput(file);
-}
-
-EXPORT_SYMBOL(fput);
-
 /**
  * drop_file_write_access - give up ability to write to a file
  * @file: the file to which we will stop writing
@@ -227,10 +219,9 @@ void drop_file_write_access(struct file *file)
 }
 EXPORT_SYMBOL_GPL(drop_file_write_access);
 
-/* __fput is called from task context when aio completion releases the last
- * last use of a struct file *.  Do not use otherwise.
+/* the real guts of fput() - releasing the last reference to file
  */
-void __fput(struct file *file)
+static void __fput(struct file *file)
 {
 	struct dentry *dentry = file->f_path.dentry;
 	struct vfsmount *mnt = file->f_path.mnt;
@@ -268,6 +259,14 @@ void __fput(struct file *file)
 	mntput(mnt);
 }
 
+void fput(struct file *file)
+{
+	if (atomic_long_dec_and_test(&file->f_count))
+		__fput(file);
+}
+
+EXPORT_SYMBOL(fput);
+
 struct file *fget(unsigned int fd)
 {
 	struct file *file;

include/linux/file.h

@@ -11,7 +11,6 @@
 
 struct file;
 
-extern void __fput(struct file *);
 extern void fput(struct file *);
 extern void drop_file_write_access(struct file *file);
 

include/linux/fs.h

@@ -954,6 +954,7 @@ extern spinlock_t files_lock;
 #define file_list_unlock() spin_unlock(&files_lock);
 
 #define get_file(x)	atomic_long_inc(&(x)->f_count)
+#define fput_atomic(x)	atomic_long_add_unless(&(x)->f_count, -1, 1)
 #define file_count(x)	atomic_long_read(&(x)->f_count)
 
 #ifdef CONFIG_DEBUG_WRITECOUNT