diff options
author | Xavier Hernandez <xhernandez@datalab.es> | 2015-05-20 15:17:35 +0200 |
---|---|---|
committer | Pranith Kumar Karampuri <pkarampu@redhat.com> | 2015-05-27 03:25:47 -0700 |
commit | 3b666b40efbed157e8c5991f29b345d93b28c659 (patch) | |
tree | a6fb9a20bed31bbcb0e5dd2025e51d7f4ea6e257 /xlators/cluster/ec/src/ec-common.h | |
parent | 5513144feb5b062b733d7514adf194429e31666f (diff) |
cluster/ec: Forced unlock when lock contention is detected
EC uses an eager lock mechanism to optimize multiple read/write
requests on the same entry or inode. This increases performance
but can have adverse results when other clients try to access the
same entry/inode.
To solve this, this patch adds a functionality to detect when this
happens and force an earlier release to not block other clients.
The method consists on requesting GF_GLUSTERFS_INODELK_COUNT and
GF_GLUSTERFS_ENTRYLK_COUNT for all fops that take a lock. When this
count is greater than one, the lock is marked to be released. All
fops already waiting for this lock will be executed normally before
releasing the lock, but new requests that also require it will be
blocked and restarted after the lock has been released and reacquired
again.
Another problem was that some operations did correctly lock the
parent of an entry when needed, but got the size and version xattrs
from the entry instead of the parent.
This patch solves this problem by binding all queries of size and
version to each lock and replacing all entrylk calls by inodelk ones
to remove concurrent updates on directory metadata. This also allows
rename to correctly update source and destination directories.
Change-Id: I2df0b22bc6f407d49f3cbf0733b0720015bacfbd
BUG: 1165041
Signed-off-by: Xavier Hernandez <xhernandez@datalab.es>
Reviewed-on: http://review.gluster.org/10852
Tested-by: NetBSD Build System
Tested-by: Gluster Build System <jenkins@build.gluster.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Diffstat (limited to 'xlators/cluster/ec/src/ec-common.h')
-rw-r--r-- | xlators/cluster/ec/src/ec-common.h | 32 |
1 files changed, 19 insertions, 13 deletions
diff --git a/xlators/cluster/ec/src/ec-common.h b/xlators/cluster/ec/src/ec-common.h index 78cf261feeb..c0db0218699 100644 --- a/xlators/cluster/ec/src/ec-common.h +++ b/xlators/cluster/ec/src/ec-common.h @@ -38,19 +38,20 @@ typedef enum { #define EC_MINIMUM_MIN -2 #define EC_MINIMUM_ALL -3 -#define EC_LOCK_ENTRY 0 -#define EC_LOCK_INODE 1 +#define EC_UPDATE_DATA 1 +#define EC_UPDATE_META 2 +#define EC_QUERY_INFO 4 +#define EC_INODE_SIZE 8 #define EC_STATE_START 0 #define EC_STATE_END 0 #define EC_STATE_INIT 1 #define EC_STATE_LOCK 2 -#define EC_STATE_GET_SIZE_AND_VERSION 3 -#define EC_STATE_DISPATCH 4 -#define EC_STATE_PREPARE_ANSWER 5 -#define EC_STATE_REPORT 6 -#define EC_STATE_LOCK_REUSE 7 -#define EC_STATE_UNLOCK 8 +#define EC_STATE_DISPATCH 3 +#define EC_STATE_PREPARE_ANSWER 4 +#define EC_STATE_REPORT 5 +#define EC_STATE_LOCK_REUSE 6 +#define EC_STATE_UNLOCK 7 #define EC_STATE_DELAYED_START 100 @@ -84,16 +85,21 @@ void ec_update_bad(ec_fop_data_t * fop, uintptr_t good); void ec_fop_set_error(ec_fop_data_t * fop, int32_t error); -void ec_lock_prepare_inode(ec_fop_data_t *fop, loc_t *loc, int32_t update); -void ec_lock_prepare_entry(ec_fop_data_t *fop, loc_t *loc, int32_t update); -void ec_lock_prepare_fd(ec_fop_data_t *fop, fd_t *fd, int32_t update); +void ec_lock_prepare_inode(ec_fop_data_t *fop, loc_t *loc, uint32_t flags); +void ec_lock_prepare_parent_inode(ec_fop_data_t *fop, loc_t *loc, + uint32_t flags); +void ec_lock_prepare_fd(ec_fop_data_t *fop, fd_t *fd, uint32_t flags); void ec_lock(ec_fop_data_t * fop); void ec_lock_reuse(ec_fop_data_t *fop); void ec_unlock(ec_fop_data_t * fop); void ec_unlock_force(ec_fop_data_t *fop); -void ec_get_size_version(ec_fop_data_t * fop); -void ec_prepare_update(ec_fop_data_t *fop); +gf_boolean_t ec_get_inode_size(ec_fop_data_t *fop, inode_t *inode, + uint64_t *size); +gf_boolean_t ec_set_inode_size(ec_fop_data_t *fop, inode_t *inode, + uint64_t size); +void ec_clear_inode_info(ec_fop_data_t *fop, inode_t *inode); + void ec_flush_size_version(ec_fop_data_t * fop); void ec_dispatch_all(ec_fop_data_t * fop); |