From: Roland Kammerer roland.kammerer@linbit.com
[ Upstream commit d29e89e34952a9ad02c77109c71a80043544296e ]
So far there was the possibility that we called genlmsg_new(GFP_NOIO)/mutex_lock() while holding an rcu_read_lock().
This included cases like:
drbd_sync_handshake (acquire the RCU lock) drbd_asb_recover_1p drbd_khelper drbd_bcast_event genlmsg_new(GFP_NOIO) --> may sleep
drbd_sync_handshake (acquire the RCU lock) drbd_asb_recover_1p drbd_khelper notify_helper genlmsg_new(GFP_NOIO) --> may sleep
drbd_sync_handshake (acquire the RCU lock) drbd_asb_recover_1p drbd_khelper notify_helper mutex_lock --> may sleep
While using GFP_ATOMIC whould have been possible in the first two cases, the real fix is to narrow the rcu_read_lock.
Reported-by: Jia-Ju Bai baijiaju1990@163.com Reviewed-by: Lars Ellenberg lars.ellenberg@linbit.com Signed-off-by: Roland Kammerer roland.kammerer@linbit.com Signed-off-by: Jens Axboe axboe@kernel.dk Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/block/drbd/drbd_receiver.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c index 6960fb064731..4f18ec5198fb 100644 --- a/drivers/block/drbd/drbd_receiver.c +++ b/drivers/block/drbd/drbd_receiver.c @@ -3125,7 +3125,7 @@ static enum drbd_conns drbd_sync_handshake(struct drbd_peer_device *peer_device, enum drbd_conns rv = C_MASK; enum drbd_disk_state mydisk; struct net_conf *nc; - int hg, rule_nr, rr_conflict, tentative; + int hg, rule_nr, rr_conflict, tentative, always_asbp;
mydisk = device->state.disk; if (mydisk == D_NEGOTIATING) @@ -3167,8 +3167,12 @@ static enum drbd_conns drbd_sync_handshake(struct drbd_peer_device *peer_device,
rcu_read_lock(); nc = rcu_dereference(peer_device->connection->net_conf); + always_asbp = nc->always_asbp; + rr_conflict = nc->rr_conflict; + tentative = nc->tentative; + rcu_read_unlock();
- if (hg == 100 || (hg == -100 && nc->always_asbp)) { + if (hg == 100 || (hg == -100 && always_asbp)) { int pcount = (device->state.role == R_PRIMARY) + (peer_role == R_PRIMARY); int forced = (hg == -100); @@ -3207,9 +3211,6 @@ static enum drbd_conns drbd_sync_handshake(struct drbd_peer_device *peer_device, "Sync from %s node\n", (hg < 0) ? "peer" : "this"); } - rr_conflict = nc->rr_conflict; - tentative = nc->tentative; - rcu_read_unlock();
if (hg == -100) { /* FIXME this log message is not correct if we end up here