LU-6529 ldlm: cancel aged locks for LRUR

It doesn't make sense to keep the very aged lock even with the
LRUR policy. This patch decreased the default ns_max_age from 10
hours to 65 minutes and changed LRUR policy to cancel very aged
locks.

Signed-off-by: Niu Yawei <yawei.niu@intel.com>
Change-Id: Iabf6f50a7837821afcf2a75b6c17c7484b34667f
Reviewed-on: http://review.whamcloud.com/14856
Tested-by: Jenkins
Reviewed-by: Jinshan Xiong <jinshan.xiong@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

diff --git a/lustre/include/lustre_dlm.h b/lustre/include/lustre_dlm.h
index 1935e2f..192bb84 100644 (file)
--- a/lustre/include/lustre_dlm.h
+++ b/lustre/include/lustre_dlm.h
@@ -63,7 +63,7 @@ struct obd_device;
 #define OBD_LDLM_DEVICENAME  "ldlm"
 
 #define LDLM_DEFAULT_LRU_SIZE (100 * num_online_cpus())
-#define LDLM_DEFAULT_MAX_ALIVE (cfs_time_seconds(36000))
+#define LDLM_DEFAULT_MAX_ALIVE (cfs_time_seconds(3900)) /* 65 min */
 #define LDLM_CTIME_AGE_LIMIT (10)
 #define LDLM_DEFAULT_PARALLEL_AST_LIMIT 1024
 

diff --git a/lustre/ldlm/ldlm_request.c b/lustre/ldlm/ldlm_request.c
index 54ecbdd..684062f 100644 (file)
--- a/lustre/ldlm/ldlm_request.c
+++ b/lustre/ldlm/ldlm_request.c
@@ -1480,6 +1480,12 @@ static ldlm_policy_res_t ldlm_cancel_lrur_policy(struct ldlm_namespace *ns,
        if (count && added >= count)
                return LDLM_POLICY_KEEP_LOCK;
 
+       /* Despite of the LV, It doesn't make sense to keep the lock which
+        * is unused for ns_max_age time. */
+       if (cfs_time_after(cfs_time_current(),
+                          cfs_time_add(lock->l_last_used, ns->ns_max_age)))
+               return LDLM_POLICY_CANCEL_LOCK;
+
        slv = ldlm_pool_get_slv(pl);
        lvf = ldlm_pool_get_lvf(pl);
        la = cfs_duration_sec(cfs_time_sub(cur,

diff --git a/lustre/tests/sanity.sh b/lustre/tests/sanity.sh
index 7bb9c3a..e7747e5 100644 (file)
--- a/lustre/tests/sanity.sh
+++ b/lustre/tests/sanity.sh
@@ -8395,6 +8395,38 @@ test_124b() {
 }
 run_test 124b "lru resize (performance test) ======================="
 
+test_124c() {
+       [ $PARALLEL == "yes" ] && skip "skip parallel run" && return
+       [ -z "$($LCTL get_param -n mdc.*.connect_flags | grep lru_resize)" ] &&
+               skip "no lru resize on server" && return 0
+
+       # cache ununsed locks on client
+       local nr=100
+       cancel_lru_locks mdc
+       test_mkdir -p $DIR/$tdir || error "failed to create $DIR/$tdir"
+       createmany -o $DIR/$tdir/f $nr ||
+               error "failed to create $nr files in $DIR/$tdir"
+       ls -l $DIR/$tdir > /dev/null
+
+       local nsdir="ldlm.namespaces.*-MDT0000-mdc-*"
+       local unused=$($LCTL get_param -n $nsdir.lock_unused_count)
+       local max_age=$($LCTL get_param -n $nsdir.lru_max_age)
+       local recalc_p=$($LCTL get_param -n $nsdir.pool.recalc_period)
+       echo "unused=$unused, max_age=$max_age, recalc_p=$recalc_p"
+
+       # set lru_max_age to 1 sec
+       $LCTL set_param $nsdir.lru_max_age=1000 # jiffies
+       echo "sleep $((recalc_p * 2)) seconds..."
+       sleep $((recalc_p * 2))
+
+       local remaining=$($LCTL get_param -n $nsdir.lock_unused_count)
+       # restore lru_max_age
+       $LCTL set_param -n $nsdir.lru_max_age $max_age
+       [ $remaining -eq 0 ] || error "$remaining locks are not canceled"
+       unlinkmany $DIR/$tdir/f $nr
+}
+run_test 124c "LRUR cancel very aged locks"
+
 test_125() { # 13358
        [ -z "$(lctl get_param -n llite.*.client_type | grep local)" ] && skip "must run as local client" && return
        [ -z "$(lctl get_param -n mdc.*-mdc-*.connect_flags | grep acl)" ] && skip "must have acl enabled" && return