[PATCH 0/6] iscsi fixes for 5.19 or 5.20

[PATCH 0/6] iscsi fixes for 5.19 or 5.20

[Mike Christie]

The following patches are some fixes for qla4xxx and qedi. They were built
over linus's tree, but apply over Martin's staging and queueing branches.
They do not have conflicts with the other iscsi patches that I've ackd on
the list, so they can be applied before or after those patches.

The first patch is trivial and fixes a bug that can only be triggered with
qla4xxx which should be rare. Patches 2 - 6 are more invassive and fix a
regression in qedi where shutdown hangs when you are using that driver for
iscsi boot. I was not sure if this was too much of an edge case and the
pathes were too invassive for 5.19 so the patches do apply over either
of your 5.19 or 5.20 branches.

[PATCH 1/6] scsi: iscsi: Fix HW conn removal use after free

[Mike Christie]

If qla4xxx doesn't remove the connection before the session, the iSCSI
class tries to remove the connection for it. We were doing a
iscsi_put_conn() in the iter function which is not needed and will result
in a use after free because iscsi_remove_conn() will free the connection.

Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 2c0dd64159b0..e6084e158cc0 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -2138,8 +2138,6 @@ static int iscsi_iter_destroy_conn_fn(struct device *dev, void *data)
 		return 0;
 
 	iscsi_remove_conn(iscsi_dev_to_conn(dev));
-	iscsi_put_conn(iscsi_dev_to_conn(dev));
-
 	return 0;
 }
 
-- 
2.25.1

[PATCH 2/6] scsi: iscsi: Allow iscsi_if_stop_conn to be called from kernel

[Mike Christie]

iscsi_if_stop_conn is only called from the userspace interface but in the
next patch we will want to call it from the kernel interface to allow
drivers like qedi to remove sessions from inside the kernel during
shutdown. This removes the iscsi_uevent code from iscsi_if_stop_conn so we
can call it in a new helper in the next patch.

Signed-off-by: Mike Christie <michael.christie@oracle.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index e6084e158cc0..0d83b6360b8a 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -2257,16 +2257,8 @@ static void iscsi_if_disconnect_bound_ep(struct iscsi_cls_conn *conn,
 	}
 }
 
-static int iscsi_if_stop_conn(struct iscsi_transport *transport,
-			      struct iscsi_uevent *ev)
+static int iscsi_if_stop_conn(struct iscsi_cls_conn *conn, int flag)
 {
-	int flag = ev->u.stop_conn.flag;
-	struct iscsi_cls_conn *conn;
-
-	conn = iscsi_conn_lookup(ev->u.stop_conn.sid, ev->u.stop_conn.cid);
-	if (!conn)
-		return -EINVAL;
-
 	ISCSI_DBG_TRANS_CONN(conn, "iscsi if conn stop.\n");
 	/*
 	 * If this is a termination we have to call stop_conn with that flag
@@ -3713,7 +3705,12 @@ static int iscsi_if_transport_conn(struct iscsi_transport *transport,
 	case ISCSI_UEVENT_DESTROY_CONN:
 		return iscsi_if_destroy_conn(transport, ev);
 	case ISCSI_UEVENT_STOP_CONN:
-		return iscsi_if_stop_conn(transport, ev);
+		conn = iscsi_conn_lookup(ev->u.stop_conn.sid,
+					 ev->u.stop_conn.cid);
+		if (!conn)
+			return -EINVAL;
+
+		return iscsi_if_stop_conn(conn, ev->u.stop_conn.flag);
 	}
 
 	/*
-- 
2.25.1

[PATCH 3/6] scsi: iscsi: Cleanup bound endpoints during shutdown.

[Mike Christie]

In the next patch we allow drivers to drive session removal during
shutdown. In this case iscsid will not be running, so we need to detect
bound endpoints and disconnect them. This moves the bound ep check so we
now always check.

Signed-off-by: Mike Christie <michael.christie@oracle.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 0d83b6360b8a..e4614dede7e9 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -2260,6 +2260,16 @@ static void iscsi_if_disconnect_bound_ep(struct iscsi_cls_conn *conn,
 static int iscsi_if_stop_conn(struct iscsi_cls_conn *conn, int flag)
 {
 	ISCSI_DBG_TRANS_CONN(conn, "iscsi if conn stop.\n");
+	/*
+	 * For offload, iscsid may not know about the ep like when iscsid is
+	 * restarted or for kernel based session shutdown iscsid is not even
+	 * up. For these cases, we do the disconnect now.
+	 */
+	mutex_lock(&conn->ep_mutex);
+	if (conn->ep)
+		iscsi_if_disconnect_bound_ep(conn, conn->ep, true);
+	mutex_unlock(&conn->ep_mutex);
+
 	/*
 	 * If this is a termination we have to call stop_conn with that flag
 	 * so the correct states get set. If we haven't run the work yet try to
@@ -2269,16 +2279,6 @@ static int iscsi_if_stop_conn(struct iscsi_cls_conn *conn, int flag)
 		cancel_work_sync(&conn->cleanup_work);
 		iscsi_stop_conn(conn, flag);
 	} else {
-		/*
-		 * For offload, when iscsid is restarted it won't know about
-		 * existing endpoints so it can't do a ep_disconnect. We clean
-		 * it up here for userspace.
-		 */
-		mutex_lock(&conn->ep_mutex);
-		if (conn->ep)
-			iscsi_if_disconnect_bound_ep(conn, conn->ep, true);
-		mutex_unlock(&conn->ep_mutex);
-
 		/*
 		 * Figure out if it was the kernel or userspace initiating this.
 		 */
-- 
2.25.1

[PATCH 4/6] scsi: iscsi: Add helper to remove a session from the kernel

[Mike Christie]

During qedi shutdown we need to stop the iSCSI layer from sending new nops
as pings and from responding to target ones and make sure there is no
running connection cleanups. Commit d1f2ce77638d ("scsi: qedi: Fix host
removal with running sessions") converted the driver to use the libicsi
helper to drive session removal, so the above issues could be handled. The
problem is that during system shutdown iscsid will not be running so when
we try to remove the root session we will hang waiting for userspace to
reply.

This patch adds a helper that will drive the destruction of sessions like
these during system shutdown.

Signed-off-by: Mike Christie <michael.christie@oracle.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 49 +++++++++++++++++++++++++++++
 include/scsi/scsi_transport_iscsi.h |  1 +
 2 files changed, 50 insertions(+)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index e4614dede7e9..b67a4a938cd1 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -2334,6 +2334,55 @@ static void iscsi_cleanup_conn_work_fn(struct work_struct *work)
 	ISCSI_DBG_TRANS_CONN(conn, "cleanup done.\n");
 }
 
+static int iscsi_iter_force_destroy_conn_fn(struct device *dev, void *data)
+{
+	struct iscsi_transport *transport;
+	struct iscsi_cls_conn *conn;
+
+	if (!iscsi_is_conn_dev(dev))
+		return 0;
+
+	conn = iscsi_dev_to_conn(dev);
+	transport = conn->transport;
+
+	if (READ_ONCE(conn->state) != ISCSI_CONN_DOWN)
+		iscsi_if_stop_conn(conn, STOP_CONN_TERM);
+
+	transport->destroy_conn(conn);
+	return 0;
+}
+
+/**
+ * iscsi_force_destroy_session - destroy a session from the kernel
+ * @session: session to destroy
+ *
+ * Force the destruction of a session from the kernel. This should only be
+ * used when userspace is no longer running during system shutdown.
+ */
+void iscsi_force_destroy_session(struct iscsi_cls_session *session)
+{
+	struct iscsi_transport *transport = session->transport;
+	unsigned long flags;
+
+	WARN_ON_ONCE(system_state == SYSTEM_RUNNING);
+
+	spin_lock_irqsave(&sesslock, flags);
+	if (list_empty(&session->sess_list)) {
+		spin_unlock_irqrestore(&sesslock, flags);
+		/*
+		 * Conn/ep is already freed. Session is being torn down via
+		 * async path. For shutdown we don't care about it so return.
+		 */
+		return;
+	}
+	spin_unlock_irqrestore(&sesslock, flags);
+
+	device_for_each_child(&session->dev, NULL,
+			      iscsi_iter_force_destroy_conn_fn);
+	transport->destroy_session(session);
+}
+EXPORT_SYMBOL_GPL(iscsi_force_destroy_session);
+
 void iscsi_free_session(struct iscsi_cls_session *session)
 {
 	ISCSI_DBG_TRANS_SESSION(session, "Freeing session\n");
diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h
index 9acb8422f680..d6eab7cb221a 100644
--- a/include/scsi/scsi_transport_iscsi.h
+++ b/include/scsi/scsi_transport_iscsi.h
@@ -442,6 +442,7 @@ extern struct iscsi_cls_session *iscsi_create_session(struct Scsi_Host *shost,
 						struct iscsi_transport *t,
 						int dd_size,
 						unsigned int target_id);
+extern void iscsi_force_destroy_session(struct iscsi_cls_session *session);
 extern void iscsi_remove_session(struct iscsi_cls_session *session);
 extern void iscsi_free_session(struct iscsi_cls_session *session);
 extern struct iscsi_cls_conn *iscsi_alloc_conn(struct iscsi_cls_session *sess,
-- 
2.25.1

[PATCH 5/6] scsi: qedi: Use QEDI_MODE_NORMAL for error handling

[Mike Christie]

When handling errors that lead to host removal use QEDI_MODE_NORMAL. There
is currently no difference in behavior between NORMAL and SHUTDOWN, but in
a subsequent commit we will want to know when we are called from the
pci_driver shutdown callout vs remove/err_handler so we know when userspace
is up.

Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
---
 drivers/scsi/qedi/qedi_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/qedi/qedi_main.c b/drivers/scsi/qedi/qedi_main.c
index 83ffba7f51da..deebe62e2b41 100644
--- a/drivers/scsi/qedi/qedi_main.c
+++ b/drivers/scsi/qedi/qedi_main.c
@@ -2491,7 +2491,7 @@ static void qedi_board_disable_work(struct work_struct *work)
 	if (test_and_set_bit(QEDI_IN_SHUTDOWN, &qedi->flags))
 		return;
 
-	__qedi_remove(qedi->pdev, QEDI_MODE_SHUTDOWN);
+	__qedi_remove(qedi->pdev, QEDI_MODE_NORMAL);
 }
 
 static void qedi_shutdown(struct pci_dev *pdev)
-- 
2.25.1

[PATCH 6/6] scsi: iscsi: Fix session removal on shutdown

[Mike Christie]

When the system is shutting down, iscsid is not running so we will not get
a response to the ISCSI_ERR_INVALID_HOST error event. The system shutdown
will then hang waiting on userspace to remove the session.

This has libiscsi force the destruction of the session from the kernel when
iscsi_host_remove() is called from a driver's shutdown callout.

This fixes a regression added in qedi boot with commit d1f2ce77638d ("scsi:
qedi: Fix host removal with running sessions") which made qedi use the
common session removal function that waits on userspace instead of rolling
its own kernel based removal.

Fixes: d1f2ce77638d ("scsi: qedi: Fix host removal with running sessions")
Reviewed-by: Lee Duncan <lduncan@suse.com>
Signed-off-by: Mike Christie <michael.christie@oracle.com>
---
 drivers/infiniband/ulp/iser/iscsi_iser.c | 4 ++--
 drivers/scsi/be2iscsi/be_main.c          | 2 +-
 drivers/scsi/bnx2i/bnx2i_iscsi.c         | 2 +-
 drivers/scsi/cxgbi/libcxgbi.c            | 2 +-
 drivers/scsi/iscsi_tcp.c                 | 4 ++--
 drivers/scsi/libiscsi.c                  | 9 +++++++--
 drivers/scsi/qedi/qedi_main.c            | 9 ++++++---
 include/scsi/libiscsi.h                  | 2 +-
 8 files changed, 21 insertions(+), 13 deletions(-)

diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c
index 321949a570ed..620ae5b2d80d 100644
--- a/drivers/infiniband/ulp/iser/iscsi_iser.c
+++ b/drivers/infiniband/ulp/iser/iscsi_iser.c
@@ -568,7 +568,7 @@ static void iscsi_iser_session_destroy(struct iscsi_cls_session *cls_session)
 	struct Scsi_Host *shost = iscsi_session_to_shost(cls_session);
 
 	iscsi_session_teardown(cls_session);
-	iscsi_host_remove(shost);
+	iscsi_host_remove(shost, false);
 	iscsi_host_free(shost);
 }
 
@@ -685,7 +685,7 @@ iscsi_iser_session_create(struct iscsi_endpoint *ep,
 	return cls_session;
 
 remove_host:
-	iscsi_host_remove(shost);
+	iscsi_host_remove(shost, false);
 free_host:
 	iscsi_host_free(shost);
 	return NULL;
diff --git a/drivers/scsi/be2iscsi/be_main.c b/drivers/scsi/be2iscsi/be_main.c
index 3bb0adefbe06..02026476c39c 100644
--- a/drivers/scsi/be2iscsi/be_main.c
+++ b/drivers/scsi/be2iscsi/be_main.c
@@ -5745,7 +5745,7 @@ static void beiscsi_remove(struct pci_dev *pcidev)
 	cancel_work_sync(&phba->sess_work);
 
 	beiscsi_iface_destroy_default(phba);
-	iscsi_host_remove(phba->shost);
+	iscsi_host_remove(phba->shost, false);
 	beiscsi_disable_port(phba, 1);
 
 	/* after cancelling boot_work */
diff --git a/drivers/scsi/bnx2i/bnx2i_iscsi.c b/drivers/scsi/bnx2i/bnx2i_iscsi.c
index 15fbd09baa94..a3c800e04a2e 100644
--- a/drivers/scsi/bnx2i/bnx2i_iscsi.c
+++ b/drivers/scsi/bnx2i/bnx2i_iscsi.c
@@ -909,7 +909,7 @@ void bnx2i_free_hba(struct bnx2i_hba *hba)
 {
 	struct Scsi_Host *shost = hba->shost;
 
-	iscsi_host_remove(shost);
+	iscsi_host_remove(shost, false);
 	INIT_LIST_HEAD(&hba->ep_ofld_list);
 	INIT_LIST_HEAD(&hba->ep_active_list);
 	INIT_LIST_HEAD(&hba->ep_destroy_list);
diff --git a/drivers/scsi/cxgbi/libcxgbi.c b/drivers/scsi/cxgbi/libcxgbi.c
index 4365d52c6430..32abdf0fa9aa 100644
--- a/drivers/scsi/cxgbi/libcxgbi.c
+++ b/drivers/scsi/cxgbi/libcxgbi.c
@@ -328,7 +328,7 @@ void cxgbi_hbas_remove(struct cxgbi_device *cdev)
 		chba = cdev->hbas[i];
 		if (chba) {
 			cdev->hbas[i] = NULL;
-			iscsi_host_remove(chba->shost);
+			iscsi_host_remove(chba->shost, false);
 			pci_dev_put(cdev->pdev);
 			iscsi_host_free(chba->shost);
 		}
diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 9fee70d6434a..52c6f70d60ec 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -898,7 +898,7 @@ iscsi_sw_tcp_session_create(struct iscsi_endpoint *ep, uint16_t cmds_max,
 remove_session:
 	iscsi_session_teardown(cls_session);
 remove_host:
-	iscsi_host_remove(shost);
+	iscsi_host_remove(shost, false);
 free_host:
 	iscsi_host_free(shost);
 	return NULL;
@@ -915,7 +915,7 @@ static void iscsi_sw_tcp_session_destroy(struct iscsi_cls_session *cls_session)
 	iscsi_tcp_r2tpool_free(cls_session->dd_data);
 	iscsi_session_teardown(cls_session);
 
-	iscsi_host_remove(shost);
+	iscsi_host_remove(shost, false);
 	iscsi_host_free(shost);
 }
 
diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c
index 797abf4f5399..3ddb701cd29c 100644
--- a/drivers/scsi/libiscsi.c
+++ b/drivers/scsi/libiscsi.c
@@ -2828,11 +2828,12 @@ static void iscsi_notify_host_removed(struct iscsi_cls_session *cls_session)
 /**
  * iscsi_host_remove - remove host and sessions
  * @shost: scsi host
+ * @is_shutdown: true if called from a driver shutdown callout
  *
  * If there are any sessions left, this will initiate the removal and wait
  * for the completion.
  */
-void iscsi_host_remove(struct Scsi_Host *shost)
+void iscsi_host_remove(struct Scsi_Host *shost, bool is_shutdown)
 {
 	struct iscsi_host *ihost = shost_priv(shost);
 	unsigned long flags;
@@ -2841,7 +2842,11 @@ void iscsi_host_remove(struct Scsi_Host *shost)
 	ihost->state = ISCSI_HOST_REMOVED;
 	spin_unlock_irqrestore(&ihost->lock, flags);
 
-	iscsi_host_for_each_session(shost, iscsi_notify_host_removed);
+	if (!is_shutdown)
+		iscsi_host_for_each_session(shost, iscsi_notify_host_removed);
+	else
+		iscsi_host_for_each_session(shost, iscsi_force_destroy_session);
+
 	wait_event_interruptible(ihost->session_removal_wq,
 				 ihost->num_sessions == 0);
 	if (signal_pending(current))
diff --git a/drivers/scsi/qedi/qedi_main.c b/drivers/scsi/qedi/qedi_main.c
index deebe62e2b41..cecfb2cb4c7b 100644
--- a/drivers/scsi/qedi/qedi_main.c
+++ b/drivers/scsi/qedi/qedi_main.c
@@ -2414,9 +2414,12 @@ static void __qedi_remove(struct pci_dev *pdev, int mode)
 	int rval;
 	u16 retry = 10;
 
-	if (mode == QEDI_MODE_NORMAL || mode == QEDI_MODE_SHUTDOWN) {
-		iscsi_host_remove(qedi->shost);
+	if (mode == QEDI_MODE_NORMAL)
+		iscsi_host_remove(qedi->shost, false);
+	else if (mode == QEDI_MODE_SHUTDOWN)
+		iscsi_host_remove(qedi->shost, true);
 
+	if (mode == QEDI_MODE_NORMAL || mode == QEDI_MODE_SHUTDOWN) {
 		if (qedi->tmf_thread) {
 			destroy_workqueue(qedi->tmf_thread);
 			qedi->tmf_thread = NULL;
@@ -2791,7 +2794,7 @@ static int __qedi_probe(struct pci_dev *pdev, int mode)
 #ifdef CONFIG_DEBUG_FS
 	qedi_dbg_host_exit(&qedi->dbg_ctx);
 #endif
-	iscsi_host_remove(qedi->shost);
+	iscsi_host_remove(qedi->shost, false);
 stop_iscsi_func:
 	qedi_ops->stop(qedi->cdev);
 stop_slowpath:
diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h
index c0703cd20a99..9758a4a9923f 100644
--- a/include/scsi/libiscsi.h
+++ b/include/scsi/libiscsi.h
@@ -411,7 +411,7 @@ extern int iscsi_host_add(struct Scsi_Host *shost, struct device *pdev);
 extern struct Scsi_Host *iscsi_host_alloc(struct scsi_host_template *sht,
 					  int dd_data_size,
 					  bool xmit_can_sleep);
-extern void iscsi_host_remove(struct Scsi_Host *shost);
+extern void iscsi_host_remove(struct Scsi_Host *shost, bool is_shutdown);
 extern void iscsi_host_free(struct Scsi_Host *shost);
 extern int iscsi_target_alloc(struct scsi_target *starget);
 extern int iscsi_host_get_max_scsi_cmds(struct Scsi_Host *shost,
-- 
2.25.1

RE: [EXT] [PATCH 0/6] iscsi fixes for 5.19 or 5.20

[Nilesh Javali]

> -----Original Message-----
> From: Mike Christie <michael.christie@oracle.com>
> Sent: Friday, June 17, 2022 3:58 AM
> To: lduncan@suse.com; cleech@redhat.com; Nilesh Javali
> <njavali@marvell.com>; Manish Rangankar <mrangankar@marvell.com>;
> GR-QLogic-Storage-Upstream <GR-QLogic-Storage-
> Upstream@marvell.com>; martin.petersen@oracle.com; linux-
> scsi@vger.kernel.org; jejb@linux.ibm.com
> Subject: [EXT] [PATCH 0/6] iscsi fixes for 5.19 or 5.20
> 
> External Email
> 
> ----------------------------------------------------------------------
> The following patches are some fixes for qla4xxx and qedi. They were built
> over linus's tree, but apply over Martin's staging and queueing branches.
> They do not have conflicts with the other iscsi patches that I've ackd on
> the list, so they can be applied before or after those patches.
> 
> The first patch is trivial and fixes a bug that can only be triggered with
> qla4xxx which should be rare. Patches 2 - 6 are more invassive and fix a
> regression in qedi where shutdown hangs when you are using that driver for
> iscsi boot. I was not sure if this was too much of an edge case and the
> pathes were too invassive for 5.19 so the patches do apply over either
> of your 5.19 or 5.20 branches.
> 

Thanks for the patches.
The series looks good.

Reviewed-by: Nilesh Javali <njavali@marvell.com>
Tested-by: Nilesh Javali <njavali@marvell.com>

Re: [PATCH 0/6] iscsi fixes for 5.19 or 5.20

[Martin K. Petersen]

Mike,

> The following patches are some fixes for qla4xxx and qedi. They were built
> over linus's tree, but apply over Martin's staging and queueing branches.
> They do not have conflicts with the other iscsi patches that I've ackd on
> the list, so they can be applied before or after those patches.

Applied to 5.20/scsi-staging, thanks!

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: [PATCH 0/6] iscsi fixes for 5.19 or 5.20

[Martin K. Petersen]

On Thu, 16 Jun 2022 17:27:32 -0500, Mike Christie wrote:

> The following patches are some fixes for qla4xxx and qedi. They were built
> over linus's tree, but apply over Martin's staging and queueing branches.
> They do not have conflicts with the other iscsi patches that I've ackd on
> the list, so they can be applied before or after those patches.
> 
> The first patch is trivial and fixes a bug that can only be triggered with
> qla4xxx which should be rare. Patches 2 - 6 are more invassive and fix a
> regression in qedi where shutdown hangs when you are using that driver for
> iscsi boot. I was not sure if this was too much of an edge case and the
> pathes were too invassive for 5.19 so the patches do apply over either
> of your 5.19 or 5.20 branches.
> 
> [...]

Applied to 5.20/scsi-queue, thanks!

[1/6] scsi: iscsi: Fix HW conn removal use after free
      https://git.kernel.org/mkp/scsi/c/c577ab7ba5f3
[2/6] scsi: iscsi: Allow iscsi_if_stop_conn to be called from kernel
      https://git.kernel.org/mkp/scsi/c/3328333b47f4
[3/6] scsi: iscsi: Cleanup bound endpoints during shutdown.
      https://git.kernel.org/mkp/scsi/c/da2f132d00d9
[4/6] scsi: iscsi: Add helper to remove a session from the kernel
      https://git.kernel.org/mkp/scsi/c/bb42856bfd54
[5/6] scsi: qedi: Use QEDI_MODE_NORMAL for error handling
      https://git.kernel.org/mkp/scsi/c/7bf01eb0d4f9
[6/6] scsi: iscsi: Fix session removal on shutdown
      https://git.kernel.org/mkp/scsi/c/31500e902759

-- 
Martin K. Petersen	Oracle Linux Engineering