[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20110601080337.029731823@blue.kroah.org>
Date: Wed, 01 Jun 2011 17:00:10 +0900
From: Greg KH <gregkh@...e.de>
To: linux-kernel@...r.kernel.org, stable@...nel.org
Cc: stable-review@...nel.org, torvalds@...ux-foundation.org,
akpm@...ux-foundation.org, alan@...rguk.ukuu.org.uk,
Nicholas Bellinger <nab@...ux-iscsi.org>,
James Bottomley <jbottomley@...allels.com>,
Greg Kroah-Hartman <gregkh@...e.de>
Subject: [074/146] [SCSI] target: Fix bug with task_sg chained
2.6.38-stable review patch. If anyone has any objections, please let us know.
------------------
transport_free_dev_tasks release
From: Nicholas Bellinger <nab@...ux-iscsi.org>
commit f436677262a5b524ac87675014c6d4e8ee153029 upstream.
This patch addresses a bug in the target core release path for HW
operation where transport_free_dev_tasks() was incorrectly being called
from transport_lun_remove_cmd() while releasing a se_cmd reference and
calling struct target_core_fabric_ops->queue_data_in().
This would result in a OOPs with HW target mode when the release of
se_task->task_sg[] would happen before pci_unmap_sg() can be called in
HW target mode fabric module code. This patch addresses the issue by
moving transport_free_dev_tasks() from transport_lun_remove_cmd() into
transport_generic_free_cmd(), and adding TRANSPORT_FREE_CMD_INTR and
transport_generic_free_cmd_intr() to allow se_cmd descriptor release
to happen fromfrom within transport_processing_thread() process context
when release of se_cmd is not possible from HW interrupt context.
Signed-off-by: Nicholas Bellinger <nab@...ux-iscsi.org>
Signed-off-by: James Bottomley <jbottomley@...allels.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@...e.de>
---
drivers/target/target_core_transport.c | 13 ++++++++++++-
include/target/target_core_base.h | 1 +
include/target/target_core_transport.h | 1 +
3 files changed, 14 insertions(+), 1 deletion(-)
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -765,7 +765,6 @@ static void transport_lun_remove_cmd(str
transport_all_task_dev_remove_state(cmd);
spin_unlock_irqrestore(&T_TASK(cmd)->t_state_lock, flags);
- transport_free_dev_tasks(cmd);
check_lun:
spin_lock_irqsave(&lun->lun_cmd_lock, flags);
@@ -2062,6 +2061,13 @@ int transport_generic_handle_tmr(
}
EXPORT_SYMBOL(transport_generic_handle_tmr);
+void transport_generic_free_cmd_intr(
+ struct se_cmd *cmd)
+{
+ transport_add_cmd_to_queue(cmd, TRANSPORT_FREE_CMD_INTR);
+}
+EXPORT_SYMBOL(transport_generic_free_cmd_intr);
+
static int transport_stop_tasks_for_cmd(struct se_cmd *cmd)
{
struct se_task *task, *task_tmp;
@@ -5302,6 +5308,8 @@ void transport_generic_free_cmd(
if (wait_for_tasks && cmd->transport_wait_for_tasks)
cmd->transport_wait_for_tasks(cmd, 0, 0);
+ transport_free_dev_tasks(cmd);
+
transport_generic_remove(cmd, release_to_pool,
session_reinstatement);
}
@@ -6142,6 +6150,9 @@ get_cmd:
case TRANSPORT_REMOVE:
transport_generic_remove(cmd, 1, 0);
break;
+ case TRANSPORT_FREE_CMD_INTR:
+ transport_generic_free_cmd(cmd, 0, 1, 0);
+ break;
case TRANSPORT_PROCESS_TMR:
transport_generic_do_tmr(cmd);
break;
--- a/include/target/target_core_base.h
+++ b/include/target/target_core_base.h
@@ -98,6 +98,7 @@ enum transport_state_table {
TRANSPORT_REMOVE = 14,
TRANSPORT_FREE = 15,
TRANSPORT_NEW_CMD_MAP = 16,
+ TRANSPORT_FREE_CMD_INTR = 17,
};
/* Used for struct se_cmd->se_cmd_flags */
--- a/include/target/target_core_transport.h
+++ b/include/target/target_core_transport.h
@@ -170,6 +170,7 @@ extern int transport_generic_handle_cdb_
extern int transport_generic_handle_data(struct se_cmd *);
extern void transport_new_cmd_failure(struct se_cmd *);
extern int transport_generic_handle_tmr(struct se_cmd *);
+extern void transport_generic_free_cmd_intr(struct se_cmd *);
extern void __transport_stop_task_timer(struct se_task *, unsigned long *);
extern unsigned char transport_asciihex_to_binaryhex(unsigned char val[2]);
extern int transport_generic_map_mem_to_cmd(struct se_cmd *cmd, struct scatterlist *, u32,
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists