staging: lustre: fix bug with LL_MRF_RETURN in loop_make_request
authorGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tue, 1 Oct 2013 03:46:36 +0000 (20:46 -0700)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Tue, 1 Oct 2013 03:46:36 +0000 (20:46 -0700)
Turns out that LL_MRF_RETURN is "void" so loop_make_request() would fall
through to the error path for a successful call.  Either this never
matters, or the error path really doesn't do anything.

Fix it up to do what the code is written to look like it is doing, if
that really is what it should be doing is another story...

Cc: Peng Tao <tao.peng@emc.com>
Cc: Zhao Hongjiang <zhaohongjiang@huawei.com>
Cc: Andreas Dilger <andreas.dilger@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/staging/lustre/lustre/include/linux/lustre_compat25.h
drivers/staging/lustre/lustre/llite/lloop.c

index 1644d40b26823968c8000357f057016adaf50d45..359c6c1ba68d013589b92765c3bb40e48044b07b 100644 (file)
@@ -232,8 +232,6 @@ static inline int ll_namei_to_lookup_intent_flag(int flag)
        return flag;
 }
 
-# define LL_MRF_RETURN(rc)
-
 #include <linux/fs.h>
 
 # define ll_umode_t    umode_t
index 66ed25e0efa8c897a05eb1708641ec41c14866f5..e2421ea6135292aa37af132259356602770dcb67 100644 (file)
@@ -365,10 +365,9 @@ static void loop_make_request(struct request_queue *q, struct bio *old_bio)
                goto err;
        }
        loop_add_bio(lo, old_bio);
-       LL_MRF_RETURN(0);
+       return;
 err:
        cfs_bio_io_error(old_bio, old_bio->bi_size);
-       LL_MRF_RETURN(0);
 }