md/bitmap: fix a might_sleep() warning.

commit 8eb23b9f35aae413140d3fda766a98092c21e9b0
    sched: Debug nested sleeps

causes false-positive warnings in RAID5 code.

This annotation removes them and adds a comment
explaining why there is no real problem.

Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: NeilBrown <neilb@suse.de>

diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index da3604e73e8abafbd8e127dbfc659360d23310e5..1695ee5f3ffc30b883c83c1926745ba22e48a7e1 100644 (file)
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -72,6 +72,19 @@ __acquires(bitmap->lock)
        /* this page has not been allocated yet */
 
        spin_unlock_irq(&bitmap->lock);
+       /* It is possible that this is being called inside a
+        * prepare_to_wait/finish_wait loop from raid5c:make_request().
+        * In general it is not permitted to sleep in that context as it
+        * can cause the loop to spin freely.
+        * That doesn't apply here as we can only reach this point
+        * once with any loop.
+        * When this function completes, either bp[page].map or
+        * bp[page].hijacked.  In either case, this function will
+        * abort before getting to this point again.  So there is
+        * no risk of a free-spin, and so it is safe to assert
+        * that sleeping here is allowed.
+        */
+       sched_annotate_sleep();
        mappage = kzalloc(PAGE_SIZE, GFP_NOIO);
        spin_lock_irq(&bitmap->lock);