[PATCH] relax nr_to_write requirements

Relax the requirements on the writeback_mapping a_op.

This function is passed the number of pages which it should write.  The
current fs-writeback.c code will get confused if the address_space
writes back more pages than it was asked to.

With this change the address_space may write more pages than required
if that is convenient.  Extent-based fileystems may wish to do this.

Documentation/filesystems/Locking

@@ -179,11 +179,12 @@ existing instances of this method ->sync_page() itself doesn't look
 well-defined...
 
 	->writeback_mapping() is used for periodic writeback and for
-systemcall-initiated sync operations. The address_space should start
-I/O against at least *nr_to_write pages.  *nr_to_write must be decremented
-for each page which is written.  *nr_to_write must not go negative (this
-will be relaxed later).  If nr_to_write is NULL, all dirty pages must
-be written.
+systemcall-initiated sync operations.  The address_space should start
+I/O against at least *nr_to_write pages.  *nr_to_write must be
+decremented for each page which is written.  The address_space
+implementation may write more (or less) pages than *nr_to_write asks
+for, but it should try to be reasonably close.  If nr_to_write is NULL,
+all dirty pages must be written.
 
 	->vm_writeback() is called from the VM.  The address_space should
 start I/O against at least *nr_to_write pages, including the passed page. As

fs/fs-writeback.c

@@ -257,7 +257,7 @@ static void sync_sb_inodes(struct super_block *sb, int sync_mode,
 		if (current_is_pdflush())
 			writeback_release(bdi);
 
-		if (nr_to_write && *nr_to_write == 0)
+		if (nr_to_write && *nr_to_write <= 0)
 			break;
 	}
 out:
@@ -301,7 +301,7 @@ void writeback_unlocked_inodes(int *nr_to_write, int sync_mode,
 					older_than_this);
 			spin_lock(&sb_lock);
 		}
-		if (nr_to_write && *nr_to_write == 0)
+		if (nr_to_write && *nr_to_write <= 0)
 			break;
 	}
 	spin_unlock(&sb_lock);