batman-adv/patches/0008-batman-adv-wait-for-rtnl-in-batadv_store_mesh_iface-.patch

   1 From 96cd7725540f4dccdd6fbb4fde59243e1cc1ad80 Mon Sep 17 00:00:00 2001
   2 From: Matthias Schiffer <mschiffer@universe-factory.net>
   3 Date: Tue, 28 May 2013 17:32:32 +0200
   4 Subject: [PATCH 08/10] batman-adv: wait for rtnl in batadv_store_mesh_iface
   5  instead of failing if it is taken
   6
   7 The rtnl_lock in batadv_store_mesh_iface has been converted to a rtnl_trylock
   8 some time ago to avoid a possible deadlock between rtnl and s_active on removal
   9 of the sysfs nodes.
  10
  11 The behaviour introduced by that was quite confusing as it could lead to the
  12 sysfs store to fail, making batman-adv setup scripts unreliable. As recently the
  13 sysfs removal was postponed to a worker not running with the rtnl taken, the
  14 deadlock can't occur any more and it is safe to change the trylock back to a
  15 lock to make the sysfs store reliable again.
  16
  17 Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
  18 Reviewed-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de>
  19 Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
  20 ---
  21  sysfs.c |    5 +----
  22  1 file changed, 1 insertion(+), 4 deletions(-)
  23
  24 diff --git a/sysfs.c b/sysfs.c
  25 index 15a22ef..929e304 100644
  26 --- a/sysfs.c
  27 +++ b/sysfs.c
  28 @@ -582,10 +582,7 @@ static ssize_t batadv_store_mesh_iface(struct kobject *kobj,
  29             (strncmp(hard_iface->soft_iface->name, buff, IFNAMSIZ) == 0))
  30                 goto out;
  31
  32 -       if (!rtnl_trylock()) {
  33 -               ret = -ERESTARTSYS;
  34 -               goto out;
  35 -       }
  36 +       rtnl_lock();
  37
  38         if (status_tmp == BATADV_IF_NOT_IN_USE) {
  39                 batadv_hardif_disable_interface(hard_iface,
  40 --
  41 1.7.10.4
  42