| Commit message (Expand) | Author | Age | Files | Lines |
* | md-cluster: make resync lock also could be interruptted | Guoqing Jiang | 2016-09-21 | 1 | -2/+3 |
* | md-cluster: introduce dlm_lock_sync_interruptible to fix tasks hang | Guoqing Jiang | 2016-09-21 | 1 | -1/+39 |
* | md-cluster: convert the completion to wait queue | Guoqing Jiang | 2016-09-21 | 1 | -5/+9 |
* | md-cluster: protect md_find_rdev_nr_rcu with rcu lock | Guoqing Jiang | 2016-09-21 | 1 | -4/+8 |
* | md-cluster: remove some unnecessary dlm_unlock_sync | Guoqing Jiang | 2016-09-21 | 1 | -5/+1 |
* | md-cluster: use FORCEUNLOCK in lockres_free | Guoqing Jiang | 2016-09-21 | 1 | -15/+11 |
* | md-cluster: fix error return code in join() | Wei Yongjun | 2016-08-24 | 1 | -3/+9 |
* | md-cluster: check the return value of process_recvd_msg | Guoqing Jiang | 2016-05-09 | 1 | -3/+10 |
* | md-cluster: gather resync infos and enable recv_thread after bitmap is ready | Guoqing Jiang | 2016-05-09 | 1 | -6/+24 |
* | md-cluster: sync bitmap when node received RESYNCING msg | Guoqing Jiang | 2016-05-04 | 1 | -0/+27 |
* | md-cluster: fix locking when node joins cluster during message broadcast | Guoqing Jiang | 2016-05-04 | 1 | -3/+10 |
* | md-cluster: unregister thread if err happened | Guoqing Jiang | 2016-05-04 | 1 | -0/+2 |
* | md-cluster: wake up thread to continue recovery | Guoqing Jiang | 2016-05-04 | 1 | -3/+6 |
* | md-cluster: change resync lock from asynchronous to synchronous | Guoqing Jiang | 2016-05-04 | 1 | -2/+0 |
* | md-cluster: fix missing memory free | Shaohua Li | 2016-01-25 | 1 | -1/+5 |
* | md-cluster: update comments for MD_CLUSTER_SEND_LOCKED_ALREADY | Guoqing Jiang | 2016-01-06 | 1 | -5/+18 |
* | md-cluster: Protect communication with mutexes | Guoqing Jiang | 2016-01-06 | 1 | -10/+63 |
* | md-cluster: Defer MD reloading to mddev->thread | Guoqing Jiang | 2016-01-06 | 1 | -1/+3 |
* | md-cluster: append some actions when change bitmap from clustered to none | Guoqing Jiang | 2016-01-06 | 1 | -0/+57 |
* | md-cluster: Fix the remove sequence with the new MD reload code | Goldwyn Rodrigues | 2016-01-06 | 1 | -1/+1 |
* | md-cluster: remove a disk asynchronously from cluster environment | Guoqing Jiang | 2016-01-06 | 1 | -2/+5 |
* | md-cluster: Avoid the resync ping-pong | Goldwyn Rodrigues | 2016-01-06 | 1 | -0/+8 |
* | md-cluster: remove mddev arg from add_resync_info() | NeilBrown | 2015-10-24 | 1 | -3/+3 |
* | md-cluster: don't cast void pointers when assigning them. | NeilBrown | 2015-10-24 | 1 | -2/+2 |
* | md-cluster: discard unused sb_mutex. | NeilBrown | 2015-10-24 | 1 | -2/+0 |
* | md-cluster: Fix warnings when build with CF=-D__CHECK_ENDIAN__ | Guoqing Jiang | 2015-10-24 | 1 | -10/+11 |
* | md-cluster: metadata_update_finish: consistently use cmsg.raid_slot as le32 | NeilBrown | 2015-10-16 | 1 | -4/+5 |
* | md-cluster: Add 'SUSE' as author for md-cluster.c | Guoqing Jiang | 2015-10-12 | 1 | -0/+1 |
* | md-cluster: zero cmsg before it was sent | Guoqing Jiang | 2015-10-12 | 1 | -3/+3 |
* | md-cluster: make sure the node do not receive it's own msg | Guoqing Jiang | 2015-10-12 | 1 | -0/+3 |
* | md-cluster: remove unnecessary setting for slot | Guoqing Jiang | 2015-10-12 | 1 | -2/+0 |
* | md-cluster: make other members of cluster_msg is handled by little endian funcs | Guoqing Jiang | 2015-10-12 | 1 | -10/+14 |
* | md-cluster: Do not printk() every received message | Goldwyn Rodrigues | 2015-10-12 | 1 | -14/+0 |
* | md-cluster: Fix adding of new disk with new reload code | Goldwyn Rodrigues | 2015-10-12 | 1 | -10/+25 |
* | md-cluster: Perform resync/recovery under a DLM lock | Goldwyn Rodrigues | 2015-10-12 | 1 | -3/+26 |
* | md-cluster: Improve md_reload_sb to be less error prone | Goldwyn Rodrigues | 2015-10-12 | 1 | -11/+16 |
* | md-cluster: Wake up suspended process | Goldwyn Rodrigues | 2015-10-12 | 1 | -2/+5 |
* | md-cluster: send BITMAP_NEEDS_SYNC when node is leaving cluster | Guoqing Jiang | 2015-10-12 | 1 | -0/+20 |
* | md-cluster: Use a small window for resync | Goldwyn Rodrigues | 2015-10-12 | 1 | -36/+5 |
* | md-cluster: complete all write requests before adding suspend_info | Goldwyn Rodrigues | 2015-10-12 | 1 | -2/+5 |
* | md-cluster: remove inappropriate try_module_get from join() | NeilBrown | 2015-08-31 | 1 | -4/+0 |
* | md-cluster: Read the disk bitmap sb and check if it needs recovery | Guoqing Jiang | 2015-08-31 | 1 | -1/+15 |
* | md-cluster: only call complete(&cinfo->completion) when node join cluster | Guoqing Jiang | 2015-08-31 | 1 | -1/+10 |
* | md-cluster: add missed lockres_free | Guoqing Jiang | 2015-08-31 | 1 | -1/+3 |
* | md-cluster: remove the unused sb_lock | Guoqing Jiang | 2015-08-31 | 1 | -9/+0 |
* | md-cluster: init suspend_list and suspend_lock early in join | Guoqing Jiang | 2015-08-31 | 1 | -3/+2 |
* | md-cluster: add the error check if failed to get dlm lock | Guoqing Jiang | 2015-08-31 | 1 | -6/+35 |
* | md-cluster: init completion within lockres_init | Guoqing Jiang | 2015-08-31 | 1 | -2/+1 |
* | md-cluster: fix deadlock issue on message lock | Guoqing Jiang | 2015-08-31 | 1 | -7/+7 |
* | md-cluster: transfer the resync ownership to another node | Guoqing Jiang | 2015-08-31 | 1 | -0/+15 |