public inbox for linux-scsi@vger.kernel.org
 help / color / mirror / Atom feed
From: Mike Christie <michael.christie@oracle.com>
To: mlombard@redhat.com, martin.petersen@oracle.com,
	mgurtovoy@nvidia.com, sagi@grimberg.me, d.bogdanov@yadro.com,
	linux-scsi@vger.kernel.org, target-devel@vger.kernel.org
Cc: Mike Christie <michael.christie@oracle.com>
Subject: [PATCH 17/18] scsi: target: Remove sess_cmd_lock
Date: Thu,  9 Mar 2023 16:33:11 -0600	[thread overview]
Message-ID: <20230309223312.94595-18-michael.christie@oracle.com> (raw)
In-Reply-To: <20230309223312.94595-1-michael.christie@oracle.com>

We used to ue the sess_cmd_lock to grab a ref to the se_cmd and add it to
the se_cmd_list if the session was not being stopped. And if the cmd was
being completed while __target_check_io_state was being run then we held
the lock while taking a ref because we were looping over the se_cmd_list
and because we didn't remove the cmd from the list until after the last
ref was released in target_release_cmd_kref.

In:

Commit 6f55b06f9b07 ("scsi: target: Drop sess_cmd_lock from I/O path")

we switched from the sess list and lock use to the per cpu session
cmd_count, so the lock was no longer needed, but it didn't get removed
from __target_check_io_state. This patch removes the last target uses of
sess_cmd_lock.

Signed-off-by: Mike Christie <michael.christie@oracle.com>
---
 drivers/target/target_core_tmr.c       | 11 -----------
 drivers/target/target_core_transport.c |  1 -
 include/target/target_core_base.h      |  1 -
 3 files changed, 13 deletions(-)

diff --git a/drivers/target/target_core_tmr.c b/drivers/target/target_core_tmr.c
index 88d2a7839876..6fa037ffc119 100644
--- a/drivers/target/target_core_tmr.c
+++ b/drivers/target/target_core_tmr.c
@@ -71,10 +71,6 @@ static int target_check_cdb_and_preempt(struct list_head *list,
 static bool __target_check_io_state(struct se_cmd *se_cmd,
 				    struct se_session *tmr_sess, bool tas)
 {
-	struct se_session *sess = se_cmd->se_sess;
-
-	lockdep_assert_held(&sess->sess_cmd_lock);
-
 	/*
 	 * If command already reached CMD_T_COMPLETE state within
 	 * target_complete_cmd() or CMD_T_FABRIC_STOP due to shutdown,
@@ -137,9 +133,7 @@ void core_tmr_abort_task(
 			pr_err("ABORT_TASK: Found referenced %s task_tag: %llu\n",
 			       se_cmd->se_tfo->fabric_name, ref_tag);
 
-			spin_lock(&se_sess->sess_cmd_lock);
 			rc = __target_check_io_state(se_cmd, se_sess, 0);
-			spin_unlock(&se_sess->sess_cmd_lock);
 			if (!rc)
 				continue;
 
@@ -206,10 +200,7 @@ static void core_tmr_drain_tmr_list(
 		if (WARN_ON_ONCE(!sess))
 			continue;
 
-		spin_lock(&sess->sess_cmd_lock);
 		rc = __target_check_io_state(cmd, sess, 0);
-		spin_unlock(&sess->sess_cmd_lock);
-
 		if (!rc) {
 			printk("LUN_RESET TMR: non-zero kref_get_unless_zero\n");
 			continue;
@@ -310,9 +301,7 @@ static void core_tmr_drain_state_list(
 			if (WARN_ON_ONCE(!sess))
 				continue;
 
-			spin_lock(&sess->sess_cmd_lock);
 			rc = __target_check_io_state(cmd, tmr_sess, tas);
-			spin_unlock(&sess->sess_cmd_lock);
 			if (!rc)
 				continue;
 
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index 692104325b38..c260cb60f5cb 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -277,7 +277,6 @@ void transport_init_session(struct se_session *se_sess)
 {
 	INIT_LIST_HEAD(&se_sess->sess_list);
 	INIT_LIST_HEAD(&se_sess->sess_acl_list);
-	spin_lock_init(&se_sess->sess_cmd_lock);
 }
 EXPORT_SYMBOL(transport_init_session);
 
diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h
index 0a5b51f8e5e8..845db96f50dd 100644
--- a/include/target/target_core_base.h
+++ b/include/target/target_core_base.h
@@ -636,7 +636,6 @@ struct se_session {
 	void			*fabric_sess_ptr;
 	struct list_head	sess_list;
 	struct list_head	sess_acl_list;
-	spinlock_t		sess_cmd_lock;
 	void			*sess_cmd_map;
 	struct sbitmap_queue	sess_tag_pool;
 	struct target_cmd_counter *cmd_cnt;
-- 
2.31.1


  parent reply	other threads:[~2023-03-09 22:34 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-09 22:32 [PATCH 00/18] target: TMF and recovery fixes Mike Christie
2023-03-09 22:32 ` [PATCH 01/18] scsi: target: Move sess cmd counter to new struct Mike Christie
2023-03-09 22:32 ` [PATCH 02/18] scsi: target: Move cmd counter allocation Mike Christie
2023-03-09 22:32 ` [PATCH 03/18] scsi: target: Pass in cmd counter to use during cmd setup Mike Christie
2023-03-09 22:32 ` [PATCH 04/18] scsi: target: iscsit/isert: Alloc per conn cmd counter Mike Christie
2023-03-09 22:32 ` [PATCH 05/18] scsi: target: iscsit: stop/wait on cmds during conn close Mike Christie
2023-03-09 22:33 ` [PATCH 06/18] scsi: target: Drop t_state_lock use in compare_and_write_post Mike Christie
2023-03-09 22:33 ` [PATCH 07/18] scsi: target: Treat CMD_T_FABRIC_STOP like CMD_T_STOP Mike Christie
2023-03-15 10:47   ` Dmitry Bogdanov
2023-03-15 22:54     ` Mike Christie
2023-03-16  0:01       ` michael.christie
2023-03-09 22:33 ` [PATCH 08/18] scsi: target: iscsit: Add helper to check when cmd has failed Mike Christie
2023-03-09 22:33 ` [PATCH 09/18] scsi: target: iscsit: Cleanup isert commands at conn closure Mike Christie
2023-03-09 22:33 ` [PATCH 10/18] IB/isert: Fix hang in target_wait_for_cmds Mike Christie
2023-03-09 22:33 ` [PATCH 11/18] IB/isert: Fix use after free during conn cleanup Mike Christie
2023-03-15 15:21   ` Sagi Grimberg
2023-03-09 22:33 ` [PATCH 12/18] scsi: target: iscsit: free cmds before session free Mike Christie
2023-03-09 22:33 ` [PATCH 13/18] scsi: target: Fix multiple LUN_RESET handling Mike Christie
2023-03-15 16:13   ` Dmitry Bogdanov
2023-03-15 16:44     ` Mike Christie
2023-03-15 19:11       ` Dmitry Bogdanov
2023-03-15 21:42         ` Mike Christie
2023-03-16 10:39           ` Dmitry Bogdanov
2023-03-16 16:03             ` Mike Christie
2023-03-16 16:07             ` Mike Christie
2023-03-09 22:33 ` [PATCH 14/18] scsi: target: Don't set CMD_T_FABRIC_STOP for aborted tasks Mike Christie
2023-03-09 22:33 ` [PATCH 15/18] scsi: target: iscsit: Fix TAS handling during conn cleanup Mike Christie
2023-03-09 22:33 ` [PATCH 16/18] scsi: target: drop tas arg from __transport_wait_for_tasks Mike Christie
2023-03-09 22:33 ` Mike Christie [this message]
2023-03-09 22:33 ` [PATCH 18/18] scsi: target: Move tag pr_debug to before we do a put on the cmd Mike Christie

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230309223312.94595-18-michael.christie@oracle.com \
    --to=michael.christie@oracle.com \
    --cc=d.bogdanov@yadro.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=mgurtovoy@nvidia.com \
    --cc=mlombard@redhat.com \
    --cc=sagi@grimberg.me \
    --cc=target-devel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox