> - if (sess->sess_tearing_down || se_cmd->cmd_wait_set) {
> + if (sess->sess_tearing_down) {
> pr_debug("Attempted to abort io tag: %llu already shutdown,"
> " skipping\n", se_cmd->tag);
> spin_unlock(&se_cmd->t_state_lock);
> @@ -247,7 +247,7 @@ static void core_tmr_drain_tmr_list(
> spin_unlock(&sess->sess_cmd_lock);
> continue;
> }
> - if (sess->sess_tearing_down || cmd->cmd_wait_set) {
> + if (sess->sess_tearing_down) {
If se_cmd->cmd_wait_set is set, sess->sess_tearing_down must always
be set as well, so these are a trivially correct cleanup. Maybe
split this into a separate prep patch?
> init_completion(&cmd->finished);
> spin_lock_init(&cmd->t_state_lock);
> kref_init(&cmd->cmd_kref);
> @@ -2538,16 +2537,6 @@ int transport_generic_free_cmd(struct se_cmd *cmd, int wait_for_tasks)
> if (cmd->se_lun)
> transport_lun_remove_cmd(cmd);
> }
> - /*
> - * If the task has been internally aborted due to TMR ABORT_TASK
> - * or LUN_RESET, target_core_tmr.c is responsible for performing
> - * the remaining calls to target_put_sess_cmd(), and not the
> - * callers of this function.
> - */
> - if (aborted) {
> - pr_debug("Detected CMD_T_ABORTED for ITT: %llu\n", cmd->tag);
> - wait_for_completion(&cmd->cmd_wait_comp);
> - }
Previously this waited for the command to be freed from the TMR code,
which provided a synchronization point. I don't think it's useful,
but it should be mentioned in the changelog.
Except for that the patch looks fine to me:
Reviewed-by: Christoph Hellwig <hch@xxxxxx>
--
To unsubscribe from this list: send the line "unsubscribe target-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
