dlm: fix race between remove and lookup

It was possible for a remove message on an old
rsb to be sent after a lookup message on a new
rsb, where the rsbs were for the same resource
name.  This could lead to a missing directory
entry for the new rsb.

It is fixed by keeping a copy of the resource
name being removed until after the remove has
been sent.  A lookup checks if this in-progress
remove matches the name it is looking up.

Signed-off-by: David Teigland <teigland@redhat.com>

1 files changed, 13 insertions, 0 deletions

diff --git a/fs/dlm/dlm_internal.h b/fs/dlm/dlm_internal.h
index a5f82d5b3946..9d3e485f88c8 100644
--- a/fs/dlm/dlm_internal.h
+++ b/fs/dlm/dlm_internal.h
@@ -498,6 +498,13 @@ struct rcom_lock {
         char                    rl_lvb[0];
 };
 
+/*
+ * The max number of resources per rsbtbl bucket that shrink will attempt
+ * to remove in each iteration.
+ */
+
+#define DLM_REMOVE_NAMES_MAX 8
+
 struct dlm_ls {
         struct list_head        ls_list;        /* list of lockspaces */
         dlm_lockspace_t         *ls_local_handle;
@@ -531,6 +538,12 @@ struct dlm_ls {
         int                     ls_new_rsb_count;
         struct list_head        ls_new_rsb;     /* new rsb structs */
 
+        spinlock_t              ls_remove_spin;
+        char                    ls_remove_name[DLM_RESNAME_MAXLEN+1];
+        char                    *ls_remove_names[DLM_REMOVE_NAMES_MAX];
+        int                     ls_remove_len;
+        int                     ls_remove_lens[DLM_REMOVE_NAMES_MAX];
+
         struct list_head        ls_nodes;       /* current nodes in ls */
         struct list_head        ls_nodes_gone;  /* dead node list, recovery */
         int                     ls_num_nodes;   /* number of nodes in ls */