[LU-14981] ldlm_handle_cp_callback() RETURN() among GOTO(out, ...)s.

• Type: Bug
• Resolution: Not a Bug
• Priority: Minor
• Severity: 3
• Assignee: WC Triage
• Reporter: John Hammond

Description

7fff052c930da4822c3b2a13d130da7473a20a58 (https://review.whamcloud.com/19898 LU-7791 ldlm: signal vs CP callback race) adds to ldlm_handle_cp_callback() a RETURN() among many GOTO(out, ...)s. As a result there is no wake_up(&lock->l_waitq).

        if (!ldlm_res_eq(&dlm_req->lock_desc.l_resource.lr_name,
                         &lock->l_resource->lr_name)) {
                ldlm_resource_unlink_lock(lock);
                unlock_res_and_lock(lock);
                rc = ldlm_lock_change_resource(ns, lock,
                                &dlm_req->lock_desc.l_resource.lr_name);
		if (rc < 0) {
                        LDLM_ERROR(lock, "Failed to allocate resource");
                        GOTO(out, rc);
                }
                LDLM_DEBUG(lock, "completion AST, new resource");
                lock_res_and_lock(lock);
        }

        if (ldlm_is_failed(lock)) {
		unlock_res_and_lock(lock);
                LDLM_LOCK_RELEASE(lock);
                RETURN(-EINVAL);
        }

        if (ldlm_is_destroyed(lock) ||
            ldlm_is_granted(lock)) {
                /* b=11300: the lock has already been granted */
                unlock_res_and_lock(lock);
                LDLM_DEBUG(lock, "Double grant race happened");
                GOTO(out, rc = 0);
        }

...
out:
        if (rc < 0) {
                lock_res_and_lock(lock);
		ldlm_set_failed(lock);
                unlock_res_and_lock(lock);
                wake_up(&lock->l_waitq);
        }
        LDLM_LOCK_RELEASE(lock);

        return 0;
}

Is this correct?