From: Nicholas Bellinger <nab@xxxxxxxxxxxxxxx>
After the v4.2+ RCU conversion to se_node_acl->lun_entry_hlist,
a BUG_ON() was added in core_enable_device_list_for_node() to
detect when the passed *lun does not match the existing
orig->se_lun pointer reference.
However, this scenario can occur happen when a dynamically
generated NodeACL is being converted to an explicit NodeACL,
when the explicit NodeACL contains a different LUN mapping
than the default provided by the WWN endpoint.
So instead of triggering BUG_ON(), go ahead and fail instead
following the original pre RCU conversion logic.
Reported-by: Benjamin ESTRABAUD <ben.estrabaud@xxxxxxxxxx>
Cc: Benjamin ESTRABAUD <ben.estrabaud@xxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx # 4.2+
Signed-off-by: Nicholas Bellinger <nab@xxxxxxxxxxxxxxx>
---
drivers/target/target_core_device.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c
index 1ebd13e..23e89af 100644
--- a/drivers/target/target_core_device.c
+++ b/drivers/target/target_core_device.c
@@ -345,14 +345,22 @@ int core_enable_device_list_for_node(
lockdep_is_held(&nacl->lun_entry_mutex));
if (orig_lun != lun) {
- pr_err("Existing orig->se_lun doesn't match new lun"
- " for dynamic -> explicit NodeACL conversion:"
- " %s\n", nacl->initiatorname);
+ pr_warn_ratelimited("Existing orig->se_lun doesn't match"
+ " new lun for dynamic -> explicit NodeACL"
+ " conversion: %s\n", nacl->initiatorname);
+ mutex_unlock(&nacl->lun_entry_mutex);
+ kfree(new);
+ return -EINVAL;
+ }
+ if (orig->se_lun_acl != NULL) {
+ pr_warn_ratelimited("Detected existing explicit"
+ " se_lun_acl->se_lun_group reference for %s"
+ " mapped_lun: %llu, ignoring\n",
+ nacl->initiatorname, mapped_lun);
mutex_unlock(&nacl->lun_entry_mutex);
kfree(new);
return -EINVAL;
}
- BUG_ON(orig->se_lun_acl != NULL);
rcu_assign_pointer(new->se_lun, lun);
rcu_assign_pointer(new->se_lun_acl, lun_acl);
--
1.9.1
