if blktrace program segfault it will not be able
to call BLKTRACETEARDOWN. Now if we run the blktrace
again that would result in a failure to create the
block/<device> debugfs directory.This will result
in blk_remove_root() to be called which will set
blk_tree_root to NULL. But the  debugfs block dir
still exist because it contain subdirectory.
Now if we try to fix it using BLKTRACETEARDOWN
it won't work because blk_tree_root is NULL.
Fix the same.
Tested as below
root@qemu-image:/home/kvaneesh/blktrace# ./blktrace  -d /dev/hdc
Segmentation fault
root@qemu-image:/home/kvaneesh/blktrace# ./blktrace  -d /dev/hdc
BLKTRACESETUP: No such file or directory
Failed to start trace on /dev/hdc
root@qemu-image:/home/kvaneesh/blktrace# ./blktrace  -k /dev/hdc
root@qemu-image:/home/kvaneesh/blktrace# ./blktrace  -d /dev/hdc
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
 static struct dentry *blk_create_tree(const char *blk_name)
 {
        struct dentry *dir = NULL;
+       int created = 0;
 
        mutex_lock(&blk_tree_mutex);
 
                blk_tree_root = debugfs_create_dir("block", NULL);
                if (!blk_tree_root)
                        goto err;
+               created = 1;
        }
 
        dir = debugfs_create_dir(blk_name, blk_tree_root);
        if (dir)
                root_users++;
-       else
-               blk_remove_root();
+       else {
+               /* Delete root only if we created it */
+               if (created)
+                       blk_remove_root();
+       }
 
 err:
        mutex_unlock(&blk_tree_mutex);