LU-5331 obdclass: serialize lu_site purge

Umount process relies on lu_site_purge(-1) to purge all
objects before umount, however, if there happen to have a
cache shrinker which calls lu_site_purge(nr) in parallel,
some objects may still being freed by cache shrinker even
after the lu_site_purge(-1) called by umount done.

This can be simply fixed by serializing purge threads,
since it doesn't make any sense to have them in parallel.

Signed-off-by: Niu Yawei <yawei.niu@intel.com>
Change-Id: Iec5d7fd1d5714ad0e4bd12a50a837858a60b2b1c
Reviewed-on: http://review.whamcloud.com/11099
Tested-by: Jenkins
Tested-by: Maloo <hpdd-maloo@intel.com>
Reviewed-by: Lai Siyao <lai.siyao@intel.com>
Reviewed-by: Mike Pershin <mike.pershin@intel.com>
Reviewed-by: Oleg Drokin <oleg.drokin@intel.com>

diff --git a/lustre/include/lu_object.h b/lustre/include/lu_object.h
index d3b234d..96b6586 100644 (file)
--- a/lustre/include/lu_object.h
+++ b/lustre/include/lu_object.h
@@ -626,7 +626,10 @@ struct lu_site {
          **/
         cfs_list_t                ls_ld_linkage;
        spinlock_t              ls_ld_lock;
-
+       /**
+        * Lock to serialize site purge.
+        */
+       struct mutex            ls_purge_mutex;
        /**
         * lu_site stats
         */

diff --git a/lustre/obdclass/lu_object.c b/lustre/obdclass/lu_object.c
index 618c5bc..86ccdc4 100644 (file)
--- a/lustre/obdclass/lu_object.c
+++ b/lustre/obdclass/lu_object.c
@@ -367,6 +367,11 @@ int lu_site_purge(const struct lu_env *env, struct lu_site *s, int nr)
         start = s->ls_purge_start;
         bnr = (nr == ~0) ? -1 : nr / CFS_HASH_NBKT(s->ls_obj_hash) + 1;
  again:
+       /*
+        * It doesn't make any sense to make purge threads parallel, that can
+        * only bring troubles to us. See LU-5331.
+        */
+       mutex_lock(&s->ls_purge_mutex);
         did_sth = 0;
         cfs_hash_for_each_bucket(s->ls_obj_hash, &bd, i) {
                 if (i < start)
@@ -411,6 +416,7 @@ int lu_site_purge(const struct lu_env *env, struct lu_site *s, int nr)
                 if (nr == 0)
                         break;
         }
+       mutex_unlock(&s->ls_purge_mutex);
 
         if (nr != 0 && did_sth && start != 0) {
                 start = 0; /* restart from the first bucket */
@@ -1095,6 +1101,7 @@ int lu_site_init(struct lu_site *s, struct lu_device *top)
        ENTRY;
 
        memset(s, 0, sizeof *s);
+       mutex_init(&s->ls_purge_mutex);
        bits = lu_htable_order(top);
        snprintf(name, 16, "lu_site_%s", top->ld_type->ldt_name);
        for (bits = min(max(LU_SITE_BITS_MIN, bits), LU_SITE_BITS_MAX);