lkml.org 
[lkml]   [2019]   [Aug]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH 5.2 081/131] dax: Fix missed wakeup in put_unlocked_entry()
Date
From: Jan Kara <jack@suse.cz>

commit 61c30c98ef17e5a330d7bb8494b78b3d6dffe9b8 upstream.

The condition checking whether put_unlocked_entry() needs to wake up
following waiter got broken by commit 23c84eb78375 ("dax: Fix missed
wakeup with PMD faults"). We need to wake the waiter whenever the passed
entry is valid (i.e., non-NULL and not special conflict entry). This
could lead to processes never being woken up when waiting for entry
lock. Fix the condition.

Cc: <stable@vger.kernel.org>
Link: http://lore.kernel.org/r/20190729120228.GC17833@quack2.suse.cz
Fixes: 23c84eb78375 ("dax: Fix missed wakeup with PMD faults")
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
fs/dax.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

--- a/fs/dax.c
+++ b/fs/dax.c
@@ -267,7 +267,7 @@ static void wait_entry_unlocked(struct x
static void put_unlocked_entry(struct xa_state *xas, void *entry)
{
/* If we were the only waiter woken, wake the next one */
- if (entry && dax_is_conflict(entry))
+ if (entry && !dax_is_conflict(entry))
dax_wake_entry(xas, entry, false);
}


\
 
 \ /
  Last update: 2019-08-05 15:25    [W:0.764 / U:0.180 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site