[PATCH 5/9] target: Skip ->queue_data_in() callbacks for COMPARE_AND_WRITE

[Nicholas A. Bellinger]

From: Nicholas Bellinger <nab@xxxxxxxxxxxxx>

COMPARE_AND_WRITE uses cmd->t_bidi_data_sg for it's READ payload
and cmd->data_direction == DMA_TO_DEVICE, but the payload is
only used for internal comparision, and never actually sent over
the wire.

So, check for this special case in to avoid TFO->queue_data_in()
within transport_complete_qf() + target_complete_ok_work() code.

Cc: Christoph Hellwig <hch@xxxxxx>
Cc: Hannes Reinecke <hare@xxxxxxx>
Cc: Martin Petersen <martin.petersen@xxxxxxxxxx>
Cc: Chris Mason <chris.mason@xxxxxxxxxxxx>
Cc: James Bottomley <JBottomley@xxxxxxxxxxxxx>
Cc: Nicholas Bellinger <nab@xxxxxxxxxxxxxxx>
Signed-off-by: Nicholas Bellinger <nab@xxxxxxxxxxxxx>
---
 drivers/target/target_core_transport.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index f7dc479..60d1336 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -1832,7 +1832,8 @@ static void transport_complete_qf(struct se_cmd *cmd)
 		ret = cmd->se_tfo->queue_data_in(cmd);
 		break;
 	case DMA_TO_DEVICE:
-		if (cmd->t_bidi_data_sg) {
+		if (cmd->t_bidi_data_sg &&
+		    cmd->t_task_cdb[0] != COMPARE_AND_WRITE) {
 			ret = cmd->se_tfo->queue_data_in(cmd);
 			if (ret < 0)
 				break;
@@ -1947,7 +1948,8 @@ static void target_complete_ok_work(struct work_struct *work)
 		/*
 		 * Check if we need to send READ payload for BIDI-COMMAND
 		 */
-		if (cmd->t_bidi_data_sg) {
+		if (cmd->t_bidi_data_sg &&
+		    cmd->t_task_cdb[0] != COMPARE_AND_WRITE) {
 			spin_lock(&cmd->se_lun->lun_sep_lock);
 			if (cmd->se_lun->lun_sep) {
 				cmd->se_lun->lun_sep->sep_stats.tx_data_octets +=
-- 
1.7.10.4

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/