On Tue, 2011-10-18 at 06:57 -0400, Christoph Hellwig wrote:
> plain text document attachment (target-optimize-single-sgls)
> If we only have a single task per command (which at least in my testing
> is the by far most common case) we do not have to allocate a new per-task
> S/G list but can reuse the one from the command.
>
> Signed-off-by: Christoph Hellwig <hch@xxxxxx>
>
A very nice optimization for the single task case, but you would be
surprised how many uber fast flash backends end up using max_sectors=128
and use task_count > 1 for normal large block operation.. :)
> Index: lio-core/drivers/target/target_core_transport.c
> ===================================================================
> --- lio-core.orig/drivers/target/target_core_transport.c 2011-10-18 12:28:49.444351865 +0200
> +++ lio-core/drivers/target/target_core_transport.c 2011-10-18 12:28:53.465359432 +0200
> @@ -3531,7 +3531,8 @@ static void transport_free_dev_tasks(str
> */
> del_timer_sync(&task->task_timer);
>
> - kfree(task->task_sg);
> + if (task->task_sg != cmd->t_data_sg)
> + kfree(task->task_sg);
>
> list_del(&task->t_list);
>
Folding in the following change to your original patch to address a
double memory free with BIDI operation.
Thanks Christoph!
--nab
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index dd49f44..467e9f43 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -3553,7 +3553,8 @@ static void transport_free_dev_tasks(struct se_cmd *cmd)
*/
del_timer_sync(&task->task_timer);
- if (task->task_sg != cmd->t_data_sg)
+ if (task->task_sg != cmd->t_data_sg &&
+ task->task_sg != cmd->t_bidi_data_sg)
kfree(task->task_sg);
list_del(&task->t_list);
--
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
