Commit b246d3d1 authored by Xue jiufei's avatar Xue jiufei Committed by Linus Torvalds

ocfs2: fix a deadlock while o2net_wq doing direct memory reclaim

Fix a deadlock problem caused by direct memory reclaim in o2net_wq.  The
situation is as follows:

1) Receive a connect message from another node, node queues a
   work_struct o2net_listen_work.

2) o2net_wq processes this work and call the following functions:

o2net_wq
-> o2net_accept_one
  -> sock_create_lite
    -> sock_alloc()
      -> kmem_cache_alloc with GFP_KERNEL
        -> ____cache_alloc_node
          ->__alloc_pages_nodemask
            -> do_try_to_free_pages
              -> shrink_slab
                -> evict
                  -> ocfs2_evict_inode
                    -> ocfs2_drop_lock
                      -> dlmunlock
                        -> o2net_send_message_vec

   then o2net_wq wait for the unlock reply from master.

3) tcp layer received the reply, call o2net_data_ready() and queue
   sc_rx_work, waiting o2net_wq to process this work.

4) o2net_wq is a single thread workqueue, it process the work one by
   one.  Right now it is still doing o2net_listen_work and cannot handle
   sc_rx_work.  so we deadlock.

Junxiao Bi's patch "mm: clear __GFP_FS when PF_MEMALLOC_NOIO is set"
(http://ozlabs.org/~akpm/mmots/broken-out/mm-clear-__gfp_fs-when-pf_memalloc_noio-is-set.patch)
clears __GFP_FS in memalloc_noio_flags() besides __GFP_IO.  We use
memalloc_noio_save() to set process flag PF_MEMALLOC_NOIO so that all
allocations done by this process are done as if GFP_NOIO was specified.
We are not reentering filesystem while doing memory reclaim.
Signed-off-by: default avatarjoyce.xue <xuejiufei@huawei.com>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 934f3072
...@@ -1601,7 +1601,15 @@ static void o2net_start_connect(struct work_struct *work) ...@@ -1601,7 +1601,15 @@ static void o2net_start_connect(struct work_struct *work)
struct sockaddr_in myaddr = {0, }, remoteaddr = {0, }; struct sockaddr_in myaddr = {0, }, remoteaddr = {0, };
int ret = 0, stop; int ret = 0, stop;
unsigned int timeout; unsigned int timeout;
unsigned int noio_flag;
/*
* sock_create allocates the sock with GFP_KERNEL. We must set
* per-process flag PF_MEMALLOC_NOIO so that all allocations done
* by this process are done as if GFP_NOIO was specified. So we
* are not reentering filesystem while doing memory reclaim.
*/
noio_flag = memalloc_noio_save();
/* if we're greater we initiate tx, otherwise we accept */ /* if we're greater we initiate tx, otherwise we accept */
if (o2nm_this_node() <= o2net_num_from_nn(nn)) if (o2nm_this_node() <= o2net_num_from_nn(nn))
goto out; goto out;
...@@ -1710,6 +1718,7 @@ static void o2net_start_connect(struct work_struct *work) ...@@ -1710,6 +1718,7 @@ static void o2net_start_connect(struct work_struct *work)
if (mynode) if (mynode)
o2nm_node_put(mynode); o2nm_node_put(mynode);
memalloc_noio_restore(noio_flag);
return; return;
} }
...@@ -1836,6 +1845,15 @@ static int o2net_accept_one(struct socket *sock, int *more) ...@@ -1836,6 +1845,15 @@ static int o2net_accept_one(struct socket *sock, int *more)
struct o2nm_node *local_node = NULL; struct o2nm_node *local_node = NULL;
struct o2net_sock_container *sc = NULL; struct o2net_sock_container *sc = NULL;
struct o2net_node *nn; struct o2net_node *nn;
unsigned int noio_flag;
/*
* sock_create_lite allocates the sock with GFP_KERNEL. We must set
* per-process flag PF_MEMALLOC_NOIO so that all allocations done
* by this process are done as if GFP_NOIO was specified. So we
* are not reentering filesystem while doing memory reclaim.
*/
noio_flag = memalloc_noio_save();
BUG_ON(sock == NULL); BUG_ON(sock == NULL);
*more = 0; *more = 0;
...@@ -1952,6 +1970,8 @@ static int o2net_accept_one(struct socket *sock, int *more) ...@@ -1952,6 +1970,8 @@ static int o2net_accept_one(struct socket *sock, int *more)
o2nm_node_put(local_node); o2nm_node_put(local_node);
if (sc) if (sc)
sc_put(sc); sc_put(sc);
memalloc_noio_restore(noio_flag);
return ret; return ret;
} }
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment