iscsi-target: Fix mutex_trylock usage in iscsit_increment_maxcmdsn
Linux-MTD Mailing List
linux-mtd at lists.infradead.org
Fri Nov 22 17:59:08 EST 2013
Gitweb: http://git.infradead.org/?p=mtd-2.6.git;a=commit;h=5e8e6b4b3adebf01a9d97056cbbfd8c44330df99
Commit: 5e8e6b4b3adebf01a9d97056cbbfd8c44330df99
Parent: 0f59cc3fc1c04413eb1368b57f5e806ebd731ef8
Author: Nicholas Bellinger <nab at linux-iscsi.org>
AuthorDate: Tue Nov 12 17:54:56 2013 -0800
Committer: Nicholas Bellinger <nab at linux-iscsi.org>
CommitDate: Tue Nov 12 17:54:56 2013 -0800
iscsi-target: Fix mutex_trylock usage in iscsit_increment_maxcmdsn
This patch fixes a >= v3.10 regression bug with mutex_trylock() usage
within iscsit_increment_maxcmdsn(), that was originally added to allow
for a special case where ->cmdsn_mutex was already held from the
iscsit_execute_cmd() exception path for ib_isert.
When !mutex_trylock() was occuring under contention during normal RX/TX
process context codepaths, the bug was manifesting itself as the following
protocol error:
Received CmdSN: 0x000fcbb7 is greater than MaxCmdSN: 0x000fcbb6, protocol error.
Received CmdSN: 0x000fcbb8 is greater than MaxCmdSN: 0x000fcbb6, protocol error.
This patch simply avoids the direct ib_isert callback in lio_queue_status()
for the special iscsi_execute_cmd() exception cases, that allows the problematic
mutex_trylock() usage in iscsit_increment_maxcmdsn() to go away.
Reported-by: Moussa Ba <moussaba at micron.com>
Tested-by: Moussa Ba <moussaba at micron.com>
Cc: <stable at vger.kernel.org> # v3.10+
Signed-off-by: Nicholas Bellinger <nab at linux-iscsi.org>
---
drivers/target/iscsi/iscsi_target_configfs.c | 5 +++++
drivers/target/iscsi/iscsi_target_device.c | 6 +-----
2 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/target/iscsi/iscsi_target_configfs.c b/drivers/target/iscsi/iscsi_target_configfs.c
index 1eec37c..fde3624 100644
--- a/drivers/target/iscsi/iscsi_target_configfs.c
+++ b/drivers/target/iscsi/iscsi_target_configfs.c
@@ -1790,6 +1790,11 @@ static int lio_queue_status(struct se_cmd *se_cmd)
struct iscsi_cmd *cmd = container_of(se_cmd, struct iscsi_cmd, se_cmd);
cmd->i_state = ISTATE_SEND_STATUS;
+
+ if (cmd->se_cmd.scsi_status || cmd->sense_reason) {
+ iscsit_add_cmd_to_response_queue(cmd, cmd->conn, cmd->i_state);
+ return 0;
+ }
cmd->conn->conn_transport->iscsit_queue_status(cmd->conn, cmd);
return 0;
diff --git a/drivers/target/iscsi/iscsi_target_device.c b/drivers/target/iscsi/iscsi_target_device.c
index 6c7a510..7087c73 100644
--- a/drivers/target/iscsi/iscsi_target_device.c
+++ b/drivers/target/iscsi/iscsi_target_device.c
@@ -58,11 +58,7 @@ void iscsit_increment_maxcmdsn(struct iscsi_cmd *cmd, struct iscsi_session *sess
cmd->maxcmdsn_inc = 1;
- if (!mutex_trylock(&sess->cmdsn_mutex)) {
- sess->max_cmd_sn += 1;
- pr_debug("Updated MaxCmdSN to 0x%08x\n", sess->max_cmd_sn);
- return;
- }
+ mutex_lock(&sess->cmdsn_mutex);
sess->max_cmd_sn += 1;
pr_debug("Updated MaxCmdSN to 0x%08x\n", sess->max_cmd_sn);
mutex_unlock(&sess->cmdsn_mutex);
More information about the linux-mtd-cvs
mailing list