From: Xue jiufei Date: Thu, 9 Oct 2014 22:28:26 +0000 (-0700) Subject: ocfs2: fix a deadlock while o2net_wq doing direct memory reclaim X-Git-Tag: fixes-against-v3.18-rc2~137^2~56 X-Git-Url: http://git.openpandora.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=b246d3d11e9c04f76a4fd6aae9c61da82bba0afb;p=pandora-kernel.git 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: joyce.xue Cc: Junxiao Bi Cc: Joel Becker Cc: Mark Fasheh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Reading git-diff-tree failed