2007-09-07 14:15:31 +07:00
|
|
|
/*
|
2008-06-10 23:20:58 +07:00
|
|
|
* zfcp device driver
|
2005-04-17 05:20:36 +07:00
|
|
|
*
|
2008-06-10 23:20:58 +07:00
|
|
|
* Error Recovery Procedures (ERP).
|
2007-09-07 14:15:31 +07:00
|
|
|
*
|
2012-07-20 16:15:04 +07:00
|
|
|
* Copyright IBM Corp. 2002, 2010
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
|
|
|
|
2008-12-25 19:39:53 +07:00
|
|
|
#define KMSG_COMPONENT "zfcp"
|
|
|
|
#define pr_fmt(fmt) KMSG_COMPONENT ": " fmt
|
|
|
|
|
2009-08-18 20:43:25 +07:00
|
|
|
#include <linux/kthread.h>
|
2005-04-17 05:20:36 +07:00
|
|
|
#include "zfcp_ext.h"
|
2010-02-17 17:18:50 +07:00
|
|
|
#include "zfcp_reqlist.h"
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
#define ZFCP_MAX_ERPS 3
|
|
|
|
|
|
|
|
enum zfcp_erp_act_flags {
|
|
|
|
ZFCP_STATUS_ERP_TIMEDOUT = 0x10000000,
|
|
|
|
ZFCP_STATUS_ERP_CLOSE_ONLY = 0x01000000,
|
|
|
|
ZFCP_STATUS_ERP_DISMISSING = 0x00100000,
|
|
|
|
ZFCP_STATUS_ERP_DISMISSED = 0x00200000,
|
|
|
|
ZFCP_STATUS_ERP_LOWMEM = 0x00400000,
|
2010-09-08 19:39:54 +07:00
|
|
|
ZFCP_STATUS_ERP_NO_REF = 0x00800000,
|
2008-07-02 15:56:40 +07:00
|
|
|
};
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
enum zfcp_erp_steps {
|
|
|
|
ZFCP_ERP_STEP_UNINITIALIZED = 0x0000,
|
|
|
|
ZFCP_ERP_STEP_FSF_XCONFIG = 0x0001,
|
|
|
|
ZFCP_ERP_STEP_PHYS_PORT_CLOSING = 0x0010,
|
|
|
|
ZFCP_ERP_STEP_PORT_CLOSING = 0x0100,
|
|
|
|
ZFCP_ERP_STEP_PORT_OPENING = 0x0800,
|
2010-09-08 19:39:55 +07:00
|
|
|
ZFCP_ERP_STEP_LUN_CLOSING = 0x1000,
|
|
|
|
ZFCP_ERP_STEP_LUN_OPENING = 0x2000,
|
2008-07-02 15:56:40 +07:00
|
|
|
};
|
|
|
|
|
|
|
|
enum zfcp_erp_act_type {
|
2010-09-08 19:39:55 +07:00
|
|
|
ZFCP_ERP_ACTION_REOPEN_LUN = 1,
|
2008-07-02 15:56:40 +07:00
|
|
|
ZFCP_ERP_ACTION_REOPEN_PORT = 2,
|
|
|
|
ZFCP_ERP_ACTION_REOPEN_PORT_FORCED = 3,
|
|
|
|
ZFCP_ERP_ACTION_REOPEN_ADAPTER = 4,
|
|
|
|
};
|
|
|
|
|
|
|
|
enum zfcp_erp_act_state {
|
|
|
|
ZFCP_ERP_ACTION_RUNNING = 1,
|
|
|
|
ZFCP_ERP_ACTION_READY = 2,
|
|
|
|
};
|
|
|
|
|
|
|
|
enum zfcp_erp_act_result {
|
|
|
|
ZFCP_ERP_SUCCEEDED = 0,
|
|
|
|
ZFCP_ERP_FAILED = 1,
|
|
|
|
ZFCP_ERP_CONTINUES = 2,
|
|
|
|
ZFCP_ERP_EXIT = 3,
|
|
|
|
ZFCP_ERP_DISMISSED = 4,
|
|
|
|
ZFCP_ERP_NOMEM = 5,
|
|
|
|
};
|
|
|
|
|
|
|
|
static void zfcp_erp_adapter_block(struct zfcp_adapter *adapter, int mask)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_clear_adapter_status(adapter,
|
|
|
|
ZFCP_STATUS_COMMON_UNBLOCKED | mask);
|
2006-09-19 03:29:56 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_action_exists(struct zfcp_erp_action *act)
|
2006-09-19 03:29:56 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_erp_action *curr_act;
|
|
|
|
|
|
|
|
list_for_each_entry(curr_act, &act->adapter->erp_running_head, list)
|
|
|
|
if (act == curr_act)
|
|
|
|
return ZFCP_ERP_ACTION_RUNNING;
|
|
|
|
return 0;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_action_ready(struct zfcp_erp_action *act)
|
2006-09-19 03:29:56 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
|
|
|
|
|
|
|
list_move(&act->list, &act->adapter->erp_ready_head);
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("erardy1", act);
|
2009-08-18 20:43:25 +07:00
|
|
|
wake_up(&adapter->erp_ready_wq);
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("erardy2", act);
|
2006-09-19 03:29:56 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_action_dismiss(struct zfcp_erp_action *act)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
act->status |= ZFCP_STATUS_ERP_DISMISSED;
|
|
|
|
if (zfcp_erp_action_exists(act) == ZFCP_ERP_ACTION_RUNNING)
|
|
|
|
zfcp_erp_action_ready(act);
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static void zfcp_erp_action_dismiss_lun(struct scsi_device *sdev)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
|
|
|
|
if (atomic_read(&zfcp_sdev->status) & ZFCP_STATUS_COMMON_ERP_INUSE)
|
|
|
|
zfcp_erp_action_dismiss(&zfcp_sdev->erp_action);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_action_dismiss_port(struct zfcp_port *port)
|
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (atomic_read(&port->status) & ZFCP_STATUS_COMMON_ERP_INUSE)
|
|
|
|
zfcp_erp_action_dismiss(&port->erp_action);
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
else {
|
|
|
|
spin_lock(port->adapter->scsi_host->host_lock);
|
|
|
|
__shost_for_each_device(sdev, port->adapter->scsi_host)
|
2010-09-08 19:39:55 +07:00
|
|
|
if (sdev_to_zfcp(sdev)->port == port)
|
|
|
|
zfcp_erp_action_dismiss_lun(sdev);
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_unlock(port->adapter->scsi_host->host_lock);
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_action_dismiss_adapter(struct zfcp_adapter *adapter)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_port *port;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (atomic_read(&adapter->status) & ZFCP_STATUS_COMMON_ERP_INUSE)
|
|
|
|
zfcp_erp_action_dismiss(&adapter->erp_action);
|
2009-11-24 22:53:58 +07:00
|
|
|
else {
|
|
|
|
read_lock(&adapter->port_list_lock);
|
|
|
|
list_for_each_entry(port, &adapter->port_list, list)
|
2008-07-02 15:56:40 +07:00
|
|
|
zfcp_erp_action_dismiss_port(port);
|
2009-11-24 22:53:58 +07:00
|
|
|
read_unlock(&adapter->port_list_lock);
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_required_act(int want, struct zfcp_adapter *adapter,
|
|
|
|
struct zfcp_port *port,
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int need = want;
|
2010-09-08 19:39:55 +07:00
|
|
|
int l_status, p_status, a_status;
|
|
|
|
struct zfcp_scsi_dev *zfcp_sdev;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
switch (want) {
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_LUN:
|
|
|
|
zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
l_status = atomic_read(&zfcp_sdev->status);
|
|
|
|
if (l_status & ZFCP_STATUS_COMMON_ERP_INUSE)
|
2008-07-02 15:56:40 +07:00
|
|
|
return 0;
|
|
|
|
p_status = atomic_read(&port->status);
|
|
|
|
if (!(p_status & ZFCP_STATUS_COMMON_RUNNING) ||
|
|
|
|
p_status & ZFCP_STATUS_COMMON_ERP_FAILED)
|
|
|
|
return 0;
|
|
|
|
if (!(p_status & ZFCP_STATUS_COMMON_UNBLOCKED))
|
|
|
|
need = ZFCP_ERP_ACTION_REOPEN_PORT;
|
|
|
|
/* fall through */
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED:
|
2010-07-08 14:53:06 +07:00
|
|
|
p_status = atomic_read(&port->status);
|
|
|
|
if (!(p_status & ZFCP_STATUS_COMMON_OPEN))
|
|
|
|
need = ZFCP_ERP_ACTION_REOPEN_PORT;
|
|
|
|
/* fall through */
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT:
|
2008-07-02 15:56:40 +07:00
|
|
|
p_status = atomic_read(&port->status);
|
|
|
|
if (p_status & ZFCP_STATUS_COMMON_ERP_INUSE)
|
|
|
|
return 0;
|
|
|
|
a_status = atomic_read(&adapter->status);
|
|
|
|
if (!(a_status & ZFCP_STATUS_COMMON_RUNNING) ||
|
|
|
|
a_status & ZFCP_STATUS_COMMON_ERP_FAILED)
|
|
|
|
return 0;
|
2010-11-17 20:23:42 +07:00
|
|
|
if (p_status & ZFCP_STATUS_COMMON_NOESC)
|
|
|
|
return need;
|
2008-07-02 15:56:40 +07:00
|
|
|
if (!(a_status & ZFCP_STATUS_COMMON_UNBLOCKED))
|
|
|
|
need = ZFCP_ERP_ACTION_REOPEN_ADAPTER;
|
|
|
|
/* fall through */
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_ADAPTER:
|
|
|
|
a_status = atomic_read(&adapter->status);
|
|
|
|
if (a_status & ZFCP_STATUS_COMMON_ERP_INUSE)
|
|
|
|
return 0;
|
2009-08-18 20:43:27 +07:00
|
|
|
if (!(a_status & ZFCP_STATUS_COMMON_RUNNING) &&
|
|
|
|
!(a_status & ZFCP_STATUS_COMMON_OPEN))
|
|
|
|
return 0; /* shutdown requested for closed adapter */
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
return need;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:54 +07:00
|
|
|
static struct zfcp_erp_action *zfcp_erp_setup_act(int need, u32 act_status,
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_adapter *adapter,
|
|
|
|
struct zfcp_port *port,
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_erp_action *erp_action;
|
2010-09-08 19:39:55 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
switch (need) {
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_LUN:
|
|
|
|
zfcp_sdev = sdev_to_zfcp(sdev);
|
2010-09-08 19:39:54 +07:00
|
|
|
if (!(act_status & ZFCP_STATUS_ERP_NO_REF))
|
2010-09-08 19:39:55 +07:00
|
|
|
if (scsi_device_get(sdev))
|
2010-09-08 19:39:54 +07:00
|
|
|
return NULL;
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(ZFCP_STATUS_COMMON_ERP_INUSE,
|
2010-09-08 19:39:55 +07:00
|
|
|
&zfcp_sdev->status);
|
|
|
|
erp_action = &zfcp_sdev->erp_action;
|
2010-11-17 20:23:43 +07:00
|
|
|
memset(erp_action, 0, sizeof(struct zfcp_erp_action));
|
|
|
|
erp_action->port = port;
|
|
|
|
erp_action->sdev = sdev;
|
2010-09-08 19:39:55 +07:00
|
|
|
if (!(atomic_read(&zfcp_sdev->status) &
|
|
|
|
ZFCP_STATUS_COMMON_RUNNING))
|
2010-09-08 19:39:54 +07:00
|
|
|
act_status |= ZFCP_STATUS_ERP_CLOSE_ONLY;
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT:
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED:
|
2010-02-17 17:18:56 +07:00
|
|
|
if (!get_device(&port->dev))
|
2009-11-24 22:54:05 +07:00
|
|
|
return NULL;
|
2008-07-02 15:56:40 +07:00
|
|
|
zfcp_erp_action_dismiss_port(port);
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(ZFCP_STATUS_COMMON_ERP_INUSE, &port->status);
|
2008-07-02 15:56:40 +07:00
|
|
|
erp_action = &port->erp_action;
|
2010-11-17 20:23:43 +07:00
|
|
|
memset(erp_action, 0, sizeof(struct zfcp_erp_action));
|
|
|
|
erp_action->port = port;
|
2008-07-02 15:56:40 +07:00
|
|
|
if (!(atomic_read(&port->status) & ZFCP_STATUS_COMMON_RUNNING))
|
2010-09-08 19:39:54 +07:00
|
|
|
act_status |= ZFCP_STATUS_ERP_CLOSE_ONLY;
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_ADAPTER:
|
2009-11-24 22:53:59 +07:00
|
|
|
kref_get(&adapter->ref);
|
2008-07-02 15:56:40 +07:00
|
|
|
zfcp_erp_action_dismiss_adapter(adapter);
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(ZFCP_STATUS_COMMON_ERP_INUSE, &adapter->status);
|
2008-07-02 15:56:40 +07:00
|
|
|
erp_action = &adapter->erp_action;
|
2010-11-17 20:23:43 +07:00
|
|
|
memset(erp_action, 0, sizeof(struct zfcp_erp_action));
|
2008-07-02 15:56:40 +07:00
|
|
|
if (!(atomic_read(&adapter->status) &
|
|
|
|
ZFCP_STATUS_COMMON_RUNNING))
|
2010-09-08 19:39:54 +07:00
|
|
|
act_status |= ZFCP_STATUS_ERP_CLOSE_ONLY;
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
|
|
|
|
|
|
|
default:
|
|
|
|
return NULL;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
erp_action->adapter = adapter;
|
|
|
|
erp_action->action = need;
|
2010-09-08 19:39:54 +07:00
|
|
|
erp_action->status = act_status;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
return erp_action;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_action_enqueue(int want, struct zfcp_adapter *adapter,
|
|
|
|
struct zfcp_port *port,
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev,
|
2010-12-02 21:16:16 +07:00
|
|
|
char *id, u32 act_status)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int retval = 1, need;
|
2010-12-02 21:16:12 +07:00
|
|
|
struct zfcp_erp_action *act;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-08-18 20:43:25 +07:00
|
|
|
if (!adapter->erp_thread)
|
2008-07-02 15:56:40 +07:00
|
|
|
return -EIO;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
need = zfcp_erp_required_act(want, adapter, port, sdev);
|
2008-07-02 15:56:40 +07:00
|
|
|
if (!need)
|
2005-04-17 05:20:36 +07:00
|
|
|
goto out;
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
act = zfcp_erp_setup_act(need, act_status, adapter, port, sdev);
|
2008-07-02 15:56:40 +07:00
|
|
|
if (!act)
|
|
|
|
goto out;
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(ZFCP_STATUS_ADAPTER_ERP_PENDING, &adapter->status);
|
2008-07-02 15:56:40 +07:00
|
|
|
++adapter->erp_total_count;
|
|
|
|
list_add_tail(&act->list, &adapter->erp_ready_head);
|
2009-08-18 20:43:25 +07:00
|
|
|
wake_up(&adapter->erp_ready_wq);
|
2008-07-02 15:56:40 +07:00
|
|
|
retval = 0;
|
2005-04-17 05:20:36 +07:00
|
|
|
out:
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_trig(id, adapter, port, sdev, want, need);
|
2005-04-17 05:20:36 +07:00
|
|
|
return retval;
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int _zfcp_erp_adapter_reopen(struct zfcp_adapter *adapter,
|
2010-12-02 21:16:16 +07:00
|
|
|
int clear_mask, char *id)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
|
|
|
zfcp_erp_adapter_block(adapter, clear_mask);
|
2009-03-02 19:09:08 +07:00
|
|
|
zfcp_scsi_schedule_rports_block(adapter);
|
2008-07-02 15:56:40 +07:00
|
|
|
|
|
|
|
/* ensure propagation of failed status to new devices */
|
|
|
|
if (atomic_read(&adapter->status) & ZFCP_STATUS_COMMON_ERP_FAILED) {
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_set_adapter_status(adapter,
|
|
|
|
ZFCP_STATUS_COMMON_ERP_FAILED);
|
2008-07-02 15:56:40 +07:00
|
|
|
return -EIO;
|
|
|
|
}
|
|
|
|
return zfcp_erp_action_enqueue(ZFCP_ERP_ACTION_REOPEN_ADAPTER,
|
2010-12-02 21:16:16 +07:00
|
|
|
adapter, NULL, NULL, id, 0);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* zfcp_erp_adapter_reopen - Reopen adapter.
|
|
|
|
* @adapter: Adapter to reopen.
|
|
|
|
* @clear: Status flags to clear.
|
|
|
|
* @id: Id for debug trace event.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2010-12-02 21:16:16 +07:00
|
|
|
void zfcp_erp_adapter_reopen(struct zfcp_adapter *adapter, int clear, char *id)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
|
2009-11-24 22:53:58 +07:00
|
|
|
zfcp_erp_adapter_block(adapter, clear);
|
|
|
|
zfcp_scsi_schedule_rports_block(adapter);
|
|
|
|
|
|
|
|
write_lock_irqsave(&adapter->erp_lock, flags);
|
|
|
|
if (atomic_read(&adapter->status) & ZFCP_STATUS_COMMON_ERP_FAILED)
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_set_adapter_status(adapter,
|
|
|
|
ZFCP_STATUS_COMMON_ERP_FAILED);
|
2009-11-24 22:53:58 +07:00
|
|
|
else
|
|
|
|
zfcp_erp_action_enqueue(ZFCP_ERP_ACTION_REOPEN_ADAPTER, adapter,
|
2010-12-02 21:16:16 +07:00
|
|
|
NULL, NULL, id, 0);
|
2009-11-24 22:53:58 +07:00
|
|
|
write_unlock_irqrestore(&adapter->erp_lock, flags);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
|
|
|
* zfcp_erp_adapter_shutdown - Shutdown adapter.
|
|
|
|
* @adapter: Adapter to shut down.
|
|
|
|
* @clear: Status flags to clear.
|
|
|
|
* @id: Id for debug trace event.
|
|
|
|
*/
|
|
|
|
void zfcp_erp_adapter_shutdown(struct zfcp_adapter *adapter, int clear,
|
2010-12-02 21:16:16 +07:00
|
|
|
char *id)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
|
|
|
int flags = ZFCP_STATUS_COMMON_RUNNING | ZFCP_STATUS_COMMON_ERP_FAILED;
|
2010-12-02 21:16:16 +07:00
|
|
|
zfcp_erp_adapter_reopen(adapter, clear | flags, id);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
|
|
|
* zfcp_erp_port_shutdown - Shutdown port
|
|
|
|
* @port: Port to shut down.
|
|
|
|
* @clear: Status flags to clear.
|
|
|
|
* @id: Id for debug trace event.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2010-12-02 21:16:16 +07:00
|
|
|
void zfcp_erp_port_shutdown(struct zfcp_port *port, int clear, char *id)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int flags = ZFCP_STATUS_COMMON_RUNNING | ZFCP_STATUS_COMMON_ERP_FAILED;
|
2010-12-02 21:16:16 +07:00
|
|
|
zfcp_erp_port_reopen(port, clear | flags, id);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
static void zfcp_erp_port_block(struct zfcp_port *port, int clear)
|
|
|
|
{
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_clear_port_status(port,
|
|
|
|
ZFCP_STATUS_COMMON_UNBLOCKED | clear);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
|
2010-12-02 21:16:16 +07:00
|
|
|
static void _zfcp_erp_port_forced_reopen(struct zfcp_port *port, int clear,
|
|
|
|
char *id)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
|
|
|
zfcp_erp_port_block(port, clear);
|
2009-03-02 19:09:08 +07:00
|
|
|
zfcp_scsi_schedule_rport_block(port);
|
2008-07-02 15:56:40 +07:00
|
|
|
|
|
|
|
if (atomic_read(&port->status) & ZFCP_STATUS_COMMON_ERP_FAILED)
|
|
|
|
return;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
zfcp_erp_action_enqueue(ZFCP_ERP_ACTION_REOPEN_PORT_FORCED,
|
2010-12-02 21:16:16 +07:00
|
|
|
port->adapter, port, NULL, id, 0);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* zfcp_erp_port_forced_reopen - Forced close of port and open again
|
|
|
|
* @port: Port to force close and to reopen.
|
2010-12-02 21:16:16 +07:00
|
|
|
* @clear: Status flags to clear.
|
2008-07-02 15:56:40 +07:00
|
|
|
* @id: Id for debug trace event.
|
|
|
|
*/
|
2010-12-02 21:16:16 +07:00
|
|
|
void zfcp_erp_port_forced_reopen(struct zfcp_port *port, int clear, char *id)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
struct zfcp_adapter *adapter = port->adapter;
|
|
|
|
|
2009-11-24 22:53:58 +07:00
|
|
|
write_lock_irqsave(&adapter->erp_lock, flags);
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_port_forced_reopen(port, clear, id);
|
2009-11-24 22:53:58 +07:00
|
|
|
write_unlock_irqrestore(&adapter->erp_lock, flags);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
|
2010-12-02 21:16:16 +07:00
|
|
|
static int _zfcp_erp_port_reopen(struct zfcp_port *port, int clear, char *id)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
|
|
|
zfcp_erp_port_block(port, clear);
|
2009-03-02 19:09:08 +07:00
|
|
|
zfcp_scsi_schedule_rport_block(port);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (atomic_read(&port->status) & ZFCP_STATUS_COMMON_ERP_FAILED) {
|
2005-04-17 05:20:36 +07:00
|
|
|
/* ensure propagation of failed status to new devices */
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_set_port_status(port, ZFCP_STATUS_COMMON_ERP_FAILED);
|
2008-07-02 15:56:40 +07:00
|
|
|
return -EIO;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
return zfcp_erp_action_enqueue(ZFCP_ERP_ACTION_REOPEN_PORT,
|
2010-12-02 21:16:16 +07:00
|
|
|
port->adapter, port, NULL, id, 0);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2008-07-02 15:56:40 +07:00
|
|
|
* zfcp_erp_port_reopen - trigger remote port recovery
|
|
|
|
* @port: port to recover
|
|
|
|
* @clear_mask: flags in port status to be cleared
|
2010-12-02 21:16:16 +07:00
|
|
|
* @id: Id for debug trace event.
|
2005-04-17 05:20:36 +07:00
|
|
|
*
|
2008-07-02 15:56:40 +07:00
|
|
|
* Returns 0 if recovery has been triggered, < 0 if not.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2010-12-02 21:16:16 +07:00
|
|
|
int zfcp_erp_port_reopen(struct zfcp_port *port, int clear, char *id)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int retval;
|
2009-11-24 22:53:58 +07:00
|
|
|
unsigned long flags;
|
2005-04-17 05:20:36 +07:00
|
|
|
struct zfcp_adapter *adapter = port->adapter;
|
|
|
|
|
2009-11-24 22:53:58 +07:00
|
|
|
write_lock_irqsave(&adapter->erp_lock, flags);
|
2010-12-02 21:16:16 +07:00
|
|
|
retval = _zfcp_erp_port_reopen(port, clear, id);
|
2009-11-24 22:53:58 +07:00
|
|
|
write_unlock_irqrestore(&adapter->erp_lock, flags);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
|
|
|
return retval;
|
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static void zfcp_erp_lun_block(struct scsi_device *sdev, int clear_mask)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_clear_lun_status(sdev,
|
|
|
|
ZFCP_STATUS_COMMON_UNBLOCKED | clear_mask);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static void _zfcp_erp_lun_reopen(struct scsi_device *sdev, int clear, char *id,
|
2010-12-02 21:16:16 +07:00
|
|
|
u32 act_status)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
struct zfcp_adapter *adapter = zfcp_sdev->port->adapter;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
zfcp_erp_lun_block(sdev, clear);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
if (atomic_read(&zfcp_sdev->status) & ZFCP_STATUS_COMMON_ERP_FAILED)
|
2008-07-02 15:56:40 +07:00
|
|
|
return;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
zfcp_erp_action_enqueue(ZFCP_ERP_ACTION_REOPEN_LUN, adapter,
|
2010-12-02 21:16:16 +07:00
|
|
|
zfcp_sdev->port, sdev, id, act_status);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2010-09-08 19:39:55 +07:00
|
|
|
* zfcp_erp_lun_reopen - initiate reopen of a LUN
|
|
|
|
* @sdev: SCSI device / LUN to be reopened
|
|
|
|
* @clear_mask: specifies flags in LUN status to be cleared
|
2010-12-02 21:16:16 +07:00
|
|
|
* @id: Id for debug trace event.
|
|
|
|
*
|
2005-04-17 05:20:36 +07:00
|
|
|
* Return: 0 on success, < 0 on error
|
|
|
|
*/
|
2010-12-02 21:16:16 +07:00
|
|
|
void zfcp_erp_lun_reopen(struct scsi_device *sdev, int clear, char *id)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
unsigned long flags;
|
2010-09-08 19:39:55 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
struct zfcp_port *port = zfcp_sdev->port;
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_adapter *adapter = port->adapter;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-11-24 22:53:58 +07:00
|
|
|
write_lock_irqsave(&adapter->erp_lock, flags);
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_lun_reopen(sdev, clear, id, 0);
|
2010-09-08 19:39:54 +07:00
|
|
|
write_unlock_irqrestore(&adapter->erp_lock, flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2010-09-08 19:39:55 +07:00
|
|
|
* zfcp_erp_lun_shutdown - Shutdown LUN
|
|
|
|
* @sdev: SCSI device / LUN to shut down.
|
2010-09-08 19:39:54 +07:00
|
|
|
* @clear: Status flags to clear.
|
|
|
|
* @id: Id for debug trace event.
|
|
|
|
*/
|
2010-12-02 21:16:16 +07:00
|
|
|
void zfcp_erp_lun_shutdown(struct scsi_device *sdev, int clear, char *id)
|
2010-09-08 19:39:54 +07:00
|
|
|
{
|
|
|
|
int flags = ZFCP_STATUS_COMMON_RUNNING | ZFCP_STATUS_COMMON_ERP_FAILED;
|
2010-12-02 21:16:16 +07:00
|
|
|
zfcp_erp_lun_reopen(sdev, clear | flags, id);
|
2010-09-08 19:39:54 +07:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2010-09-08 19:39:55 +07:00
|
|
|
* zfcp_erp_lun_shutdown_wait - Shutdown LUN and wait for erp completion
|
|
|
|
* @sdev: SCSI device / LUN to shut down.
|
2010-09-08 19:39:54 +07:00
|
|
|
* @id: Id for debug trace event.
|
|
|
|
*
|
2010-09-08 19:39:55 +07:00
|
|
|
* Do not acquire a reference for the LUN when creating the ERP
|
2010-09-08 19:39:54 +07:00
|
|
|
* action. It is safe, because this function waits for the ERP to
|
2010-09-08 19:39:55 +07:00
|
|
|
* complete first. This allows to shutdown the LUN, even when the SCSI
|
|
|
|
* device is in the state SDEV_DEL when scsi_device_get will fail.
|
2010-09-08 19:39:54 +07:00
|
|
|
*/
|
2010-09-08 19:39:55 +07:00
|
|
|
void zfcp_erp_lun_shutdown_wait(struct scsi_device *sdev, char *id)
|
2010-09-08 19:39:54 +07:00
|
|
|
{
|
|
|
|
unsigned long flags;
|
2010-09-08 19:39:55 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
struct zfcp_port *port = zfcp_sdev->port;
|
2010-09-08 19:39:54 +07:00
|
|
|
struct zfcp_adapter *adapter = port->adapter;
|
|
|
|
int clear = ZFCP_STATUS_COMMON_RUNNING | ZFCP_STATUS_COMMON_ERP_FAILED;
|
|
|
|
|
|
|
|
write_lock_irqsave(&adapter->erp_lock, flags);
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_lun_reopen(sdev, clear, id, ZFCP_STATUS_ERP_NO_REF);
|
2009-11-24 22:53:58 +07:00
|
|
|
write_unlock_irqrestore(&adapter->erp_lock, flags);
|
2010-09-08 19:39:54 +07:00
|
|
|
|
|
|
|
zfcp_erp_wait(adapter);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int status_change_set(unsigned long mask, atomic_t *status)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
return (atomic_read(status) ^ mask) & mask;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_adapter_unblock(struct zfcp_adapter *adapter)
|
2008-03-27 20:22:02 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
if (status_change_set(ZFCP_STATUS_COMMON_UNBLOCKED, &adapter->status))
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("eraubl1", &adapter->erp_action);
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(ZFCP_STATUS_COMMON_UNBLOCKED, &adapter->status);
|
2008-03-27 20:22:02 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_port_unblock(struct zfcp_port *port)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
if (status_change_set(ZFCP_STATUS_COMMON_UNBLOCKED, &port->status))
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("erpubl1", &port->erp_action);
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(ZFCP_STATUS_COMMON_UNBLOCKED, &port->status);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static void zfcp_erp_lun_unblock(struct scsi_device *sdev)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
|
|
|
|
if (status_change_set(ZFCP_STATUS_COMMON_UNBLOCKED, &zfcp_sdev->status))
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("erlubl1", &sdev_to_zfcp(sdev)->erp_action);
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(ZFCP_STATUS_COMMON_UNBLOCKED, &zfcp_sdev->status);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_action_to_running(struct zfcp_erp_action *erp_action)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
list_move(&erp_action->list, &erp_action->adapter->erp_running_head);
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("erator1", erp_action);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_strategy_check_fsfreq(struct zfcp_erp_action *act)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
2010-02-17 17:18:49 +07:00
|
|
|
struct zfcp_fsf_req *req;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-02-17 17:18:49 +07:00
|
|
|
if (!act->fsf_req_id)
|
2008-07-02 15:56:40 +07:00
|
|
|
return;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-02-17 17:18:50 +07:00
|
|
|
spin_lock(&adapter->req_list->lock);
|
|
|
|
req = _zfcp_reqlist_find(adapter->req_list, act->fsf_req_id);
|
2010-02-17 17:18:49 +07:00
|
|
|
if (req && req->erp_action == act) {
|
2008-07-02 15:56:40 +07:00
|
|
|
if (act->status & (ZFCP_STATUS_ERP_DISMISSED |
|
|
|
|
ZFCP_STATUS_ERP_TIMEDOUT)) {
|
2010-02-17 17:18:49 +07:00
|
|
|
req->status |= ZFCP_STATUS_FSFREQ_DISMISSED;
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("erscf_1", act);
|
2010-02-17 17:18:49 +07:00
|
|
|
req->erp_action = NULL;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2008-07-02 15:56:40 +07:00
|
|
|
if (act->status & ZFCP_STATUS_ERP_TIMEDOUT)
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("erscf_2", act);
|
2010-02-17 17:18:49 +07:00
|
|
|
if (req->status & ZFCP_STATUS_FSFREQ_DISMISSED)
|
|
|
|
act->fsf_req_id = 0;
|
2008-07-02 15:56:40 +07:00
|
|
|
} else
|
2010-02-17 17:18:49 +07:00
|
|
|
act->fsf_req_id = 0;
|
2010-02-17 17:18:50 +07:00
|
|
|
spin_unlock(&adapter->req_list->lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
|
|
|
* zfcp_erp_notify - Trigger ERP action.
|
|
|
|
* @erp_action: ERP action to continue.
|
|
|
|
* @set_mask: ERP action status flags to set.
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2008-07-02 15:56:40 +07:00
|
|
|
void zfcp_erp_notify(struct zfcp_erp_action *erp_action, unsigned long set_mask)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
struct zfcp_adapter *adapter = erp_action->adapter;
|
2008-07-02 15:56:40 +07:00
|
|
|
unsigned long flags;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
write_lock_irqsave(&adapter->erp_lock, flags);
|
2005-04-17 05:20:36 +07:00
|
|
|
if (zfcp_erp_action_exists(erp_action) == ZFCP_ERP_ACTION_RUNNING) {
|
|
|
|
erp_action->status |= set_mask;
|
|
|
|
zfcp_erp_action_ready(erp_action);
|
|
|
|
}
|
|
|
|
write_unlock_irqrestore(&adapter->erp_lock, flags);
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
|
|
|
* zfcp_erp_timeout_handler - Trigger ERP action from timed out ERP request
|
|
|
|
* @data: ERP action (from timer data)
|
2005-04-17 05:20:36 +07:00
|
|
|
*/
|
2008-07-02 15:56:40 +07:00
|
|
|
void zfcp_erp_timeout_handler(unsigned long data)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_erp_action *act = (struct zfcp_erp_action *) data;
|
|
|
|
zfcp_erp_notify(act, ZFCP_STATUS_ERP_TIMEDOUT);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_memwait_handler(unsigned long data)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
zfcp_erp_notify((struct zfcp_erp_action *)data, 0);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_strategy_memwait(struct zfcp_erp_action *erp_action)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
init_timer(&erp_action->timer);
|
|
|
|
erp_action->timer.function = zfcp_erp_memwait_handler;
|
|
|
|
erp_action->timer.data = (unsigned long) erp_action;
|
|
|
|
erp_action->timer.expires = jiffies + HZ;
|
|
|
|
add_timer(&erp_action->timer);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void _zfcp_erp_port_reopen_all(struct zfcp_adapter *adapter,
|
2010-12-02 21:16:16 +07:00
|
|
|
int clear, char *id)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_port *port;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-11-24 22:53:58 +07:00
|
|
|
read_lock(&adapter->port_list_lock);
|
|
|
|
list_for_each_entry(port, &adapter->port_list, list)
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_port_reopen(port, clear, id);
|
2009-11-24 22:53:58 +07:00
|
|
|
read_unlock(&adapter->port_list_lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static void _zfcp_erp_lun_reopen_all(struct zfcp_port *port, int clear,
|
2010-12-02 21:16:16 +07:00
|
|
|
char *id)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_lock(port->adapter->scsi_host->host_lock);
|
|
|
|
__shost_for_each_device(sdev, port->adapter->scsi_host)
|
2010-09-08 19:39:55 +07:00
|
|
|
if (sdev_to_zfcp(sdev)->port == port)
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_lun_reopen(sdev, clear, id, 0);
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_unlock(port->adapter->scsi_host->host_lock);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2009-07-13 20:06:09 +07:00
|
|
|
static void zfcp_erp_strategy_followup_failed(struct zfcp_erp_action *act)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
switch (act->action) {
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_ADAPTER:
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_adapter_reopen(act->adapter, 0, "ersff_1");
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED:
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_port_forced_reopen(act->port, 0, "ersff_2");
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT:
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_port_reopen(act->port, 0, "ersff_3");
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_LUN:
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_lun_reopen(act->sdev, 0, "ersff_4", 0);
|
2009-07-13 20:06:09 +07:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void zfcp_erp_strategy_followup_success(struct zfcp_erp_action *act)
|
|
|
|
{
|
|
|
|
switch (act->action) {
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_ADAPTER:
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_port_reopen_all(act->adapter, 0, "ersfs_1");
|
2009-07-13 20:06:09 +07:00
|
|
|
break;
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED:
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_port_reopen(act->port, 0, "ersfs_2");
|
2009-07-13 20:06:09 +07:00
|
|
|
break;
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT:
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_lun_reopen_all(act->port, 0, "ersfs_3");
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_wakeup(struct zfcp_adapter *adapter)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
unsigned long flags;
|
|
|
|
|
2009-11-24 22:53:58 +07:00
|
|
|
read_lock_irqsave(&adapter->erp_lock, flags);
|
2008-07-02 15:56:40 +07:00
|
|
|
if (list_empty(&adapter->erp_ready_head) &&
|
|
|
|
list_empty(&adapter->erp_running_head)) {
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_ADAPTER_ERP_PENDING,
|
2008-07-02 15:56:40 +07:00
|
|
|
&adapter->status);
|
|
|
|
wake_up(&adapter->erp_done_wqh);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2009-11-24 22:53:58 +07:00
|
|
|
read_unlock_irqrestore(&adapter->erp_lock, flags);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_enqueue_ptp_port(struct zfcp_adapter *adapter)
|
|
|
|
{
|
|
|
|
struct zfcp_port *port;
|
|
|
|
port = zfcp_port_enqueue(adapter, adapter->peer_wwpn, 0,
|
|
|
|
adapter->peer_d_id);
|
|
|
|
if (IS_ERR(port)) /* error or port already attached */
|
|
|
|
return;
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_port_reopen(port, 0, "ereptp1");
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_adapter_strat_fsf_xconf(struct zfcp_erp_action *erp_action)
|
|
|
|
{
|
|
|
|
int retries;
|
|
|
|
int sleep = 1;
|
|
|
|
struct zfcp_adapter *adapter = erp_action->adapter;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_ADAPTER_XCONFIG_OK, &adapter->status);
|
2008-07-02 15:56:40 +07:00
|
|
|
|
|
|
|
for (retries = 7; retries; retries--) {
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_ADAPTER_HOST_CON_INIT,
|
2008-07-02 15:56:40 +07:00
|
|
|
&adapter->status);
|
|
|
|
write_lock_irq(&adapter->erp_lock);
|
|
|
|
zfcp_erp_action_to_running(erp_action);
|
|
|
|
write_unlock_irq(&adapter->erp_lock);
|
|
|
|
if (zfcp_fsf_exchange_config_data(erp_action)) {
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_ADAPTER_HOST_CON_INIT,
|
2008-07-02 15:56:40 +07:00
|
|
|
&adapter->status);
|
|
|
|
return ZFCP_ERP_FAILED;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2009-08-18 20:43:25 +07:00
|
|
|
wait_event(adapter->erp_ready_wq,
|
|
|
|
!list_empty(&adapter->erp_ready_head));
|
2008-07-02 15:56:40 +07:00
|
|
|
if (erp_action->status & ZFCP_STATUS_ERP_TIMEDOUT)
|
|
|
|
break;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (!(atomic_read(&adapter->status) &
|
|
|
|
ZFCP_STATUS_ADAPTER_HOST_CON_INIT))
|
|
|
|
break;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
ssleep(sleep);
|
|
|
|
sleep *= 2;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_ADAPTER_HOST_CON_INIT,
|
2008-07-02 15:56:40 +07:00
|
|
|
&adapter->status);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (!(atomic_read(&adapter->status) & ZFCP_STATUS_ADAPTER_XCONFIG_OK))
|
|
|
|
return ZFCP_ERP_FAILED;
|
2007-09-07 14:15:31 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (fc_host_port_type(adapter->scsi_host) == FC_PORTTYPE_PTP)
|
|
|
|
zfcp_erp_enqueue_ptp_port(adapter);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_SUCCEEDED;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_adapter_strategy_open_fsf_xport(struct zfcp_erp_action *act)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int ret;
|
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
write_lock_irq(&adapter->erp_lock);
|
|
|
|
zfcp_erp_action_to_running(act);
|
|
|
|
write_unlock_irq(&adapter->erp_lock);
|
|
|
|
|
|
|
|
ret = zfcp_fsf_exchange_port_data(act);
|
|
|
|
if (ret == -EOPNOTSUPP)
|
|
|
|
return ZFCP_ERP_SUCCEEDED;
|
|
|
|
if (ret)
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("erasox1", act);
|
2009-08-18 20:43:25 +07:00
|
|
|
wait_event(adapter->erp_ready_wq,
|
|
|
|
!list_empty(&adapter->erp_ready_head));
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("erasox2", act);
|
2008-07-02 15:56:40 +07:00
|
|
|
if (act->status & ZFCP_STATUS_ERP_TIMEDOUT)
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
|
|
|
|
return ZFCP_ERP_SUCCEEDED;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_adapter_strategy_open_fsf(struct zfcp_erp_action *act)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
if (zfcp_erp_adapter_strat_fsf_xconf(act) == ZFCP_ERP_FAILED)
|
|
|
|
return ZFCP_ERP_FAILED;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (zfcp_erp_adapter_strategy_open_fsf_xport(act) == ZFCP_ERP_FAILED)
|
|
|
|
return ZFCP_ERP_FAILED;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2011-02-23 01:54:40 +07:00
|
|
|
if (mempool_resize(act->adapter->pool.sr_data,
|
2015-04-15 05:48:21 +07:00
|
|
|
act->adapter->stat_read_buf_num))
|
2010-06-21 15:11:33 +07:00
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
|
|
|
|
if (mempool_resize(act->adapter->pool.status_read_req,
|
2015-04-15 05:48:21 +07:00
|
|
|
act->adapter->stat_read_buf_num))
|
2010-06-21 15:11:33 +07:00
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
|
2010-04-30 23:09:36 +07:00
|
|
|
atomic_set(&act->adapter->stat_miss, act->adapter->stat_read_buf_num);
|
2008-07-02 15:56:40 +07:00
|
|
|
if (zfcp_status_read_refill(act->adapter))
|
|
|
|
return ZFCP_ERP_FAILED;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_SUCCEEDED;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-03-02 19:09:03 +07:00
|
|
|
static void zfcp_erp_adapter_strategy_close(struct zfcp_erp_action *act)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/* close queues to ensure that buffers are not accessed by adapter */
|
2009-08-18 20:43:19 +07:00
|
|
|
zfcp_qdio_close(adapter->qdio);
|
2008-07-02 15:56:40 +07:00
|
|
|
zfcp_fsf_req_dismiss_all(adapter);
|
|
|
|
adapter->fsf_req_seq_no = 0;
|
2009-08-18 20:43:12 +07:00
|
|
|
zfcp_fc_wka_ports_force_offline(adapter->gs);
|
2010-09-08 19:39:55 +07:00
|
|
|
/* all ports and LUNs are closed */
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_clear_adapter_status(adapter, ZFCP_STATUS_COMMON_OPEN);
|
2009-03-02 19:09:03 +07:00
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_ADAPTER_XCONFIG_OK |
|
2009-03-02 19:09:03 +07:00
|
|
|
ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, &adapter->status);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2009-03-02 19:09:03 +07:00
|
|
|
static int zfcp_erp_adapter_strategy_open(struct zfcp_erp_action *act)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2009-03-02 19:09:03 +07:00
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-12-02 21:16:17 +07:00
|
|
|
if (zfcp_qdio_open(adapter->qdio)) {
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_ADAPTER_XCONFIG_OK |
|
2009-03-02 19:09:03 +07:00
|
|
|
ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED,
|
|
|
|
&adapter->status);
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
}
|
2008-07-02 15:56:40 +07:00
|
|
|
|
2009-03-02 19:09:03 +07:00
|
|
|
if (zfcp_erp_adapter_strategy_open_fsf(act)) {
|
|
|
|
zfcp_erp_adapter_strategy_close(act);
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
}
|
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(ZFCP_STATUS_COMMON_OPEN, &adapter->status);
|
2009-03-02 19:09:03 +07:00
|
|
|
|
|
|
|
return ZFCP_ERP_SUCCEEDED;
|
|
|
|
}
|
2008-07-02 15:56:40 +07:00
|
|
|
|
2009-03-02 19:09:03 +07:00
|
|
|
static int zfcp_erp_adapter_strategy(struct zfcp_erp_action *act)
|
|
|
|
{
|
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
|
|
|
|
|
|
|
if (atomic_read(&adapter->status) & ZFCP_STATUS_COMMON_OPEN) {
|
|
|
|
zfcp_erp_adapter_strategy_close(act);
|
|
|
|
if (act->status & ZFCP_STATUS_ERP_CLOSE_ONLY)
|
|
|
|
return ZFCP_ERP_EXIT;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (zfcp_erp_adapter_strategy_open(act)) {
|
2008-07-02 15:56:40 +07:00
|
|
|
ssleep(8);
|
2009-03-02 19:09:03 +07:00
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-03-02 19:09:03 +07:00
|
|
|
return ZFCP_ERP_SUCCEEDED;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_port_forced_strategy_close(struct zfcp_erp_action *act)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int retval;
|
|
|
|
|
|
|
|
retval = zfcp_fsf_close_physical_port(act);
|
|
|
|
if (retval == -ENOMEM)
|
|
|
|
return ZFCP_ERP_NOMEM;
|
|
|
|
act->step = ZFCP_ERP_STEP_PHYS_PORT_CLOSING;
|
|
|
|
if (retval)
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
|
|
|
|
return ZFCP_ERP_CONTINUES;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_port_forced_strategy(struct zfcp_erp_action *erp_action)
|
|
|
|
{
|
|
|
|
struct zfcp_port *port = erp_action->port;
|
|
|
|
int status = atomic_read(&port->status);
|
|
|
|
|
|
|
|
switch (erp_action->step) {
|
|
|
|
case ZFCP_ERP_STEP_UNINITIALIZED:
|
|
|
|
if ((status & ZFCP_STATUS_PORT_PHYS_OPEN) &&
|
|
|
|
(status & ZFCP_STATUS_COMMON_OPEN))
|
|
|
|
return zfcp_erp_port_forced_strategy_close(erp_action);
|
|
|
|
else
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
|
|
|
|
case ZFCP_ERP_STEP_PHYS_PORT_CLOSING:
|
2009-07-13 20:06:08 +07:00
|
|
|
if (!(status & ZFCP_STATUS_PORT_PHYS_OPEN))
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_SUCCEEDED;
|
|
|
|
}
|
|
|
|
return ZFCP_ERP_FAILED;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_port_strategy_close(struct zfcp_erp_action *erp_action)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int retval;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
retval = zfcp_fsf_close_port(erp_action);
|
|
|
|
if (retval == -ENOMEM)
|
|
|
|
return ZFCP_ERP_NOMEM;
|
|
|
|
erp_action->step = ZFCP_ERP_STEP_PORT_CLOSING;
|
|
|
|
if (retval)
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
return ZFCP_ERP_CONTINUES;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_port_strategy_open_port(struct zfcp_erp_action *erp_action)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int retval;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
retval = zfcp_fsf_open_port(erp_action);
|
|
|
|
if (retval == -ENOMEM)
|
|
|
|
return ZFCP_ERP_NOMEM;
|
|
|
|
erp_action->step = ZFCP_ERP_STEP_PORT_OPENING;
|
|
|
|
if (retval)
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
return ZFCP_ERP_CONTINUES;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_open_ptp_port(struct zfcp_erp_action *act)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
|
|
|
struct zfcp_port *port = act->port;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (port->wwpn != adapter->peer_wwpn) {
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_set_port_status(port, ZFCP_STATUS_COMMON_ERP_FAILED);
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
}
|
|
|
|
port->d_id = adapter->peer_d_id;
|
|
|
|
return zfcp_erp_port_strategy_open_port(act);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int zfcp_erp_port_strategy_open_common(struct zfcp_erp_action *act)
|
|
|
|
{
|
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
|
|
|
struct zfcp_port *port = act->port;
|
|
|
|
int p_status = atomic_read(&port->status);
|
|
|
|
|
|
|
|
switch (act->step) {
|
|
|
|
case ZFCP_ERP_STEP_UNINITIALIZED:
|
|
|
|
case ZFCP_ERP_STEP_PHYS_PORT_CLOSING:
|
|
|
|
case ZFCP_ERP_STEP_PORT_CLOSING:
|
|
|
|
if (fc_host_port_type(adapter->scsi_host) == FC_PORTTYPE_PTP)
|
|
|
|
return zfcp_erp_open_ptp_port(act);
|
2008-12-19 22:56:59 +07:00
|
|
|
if (!port->d_id) {
|
2009-10-14 16:00:43 +07:00
|
|
|
zfcp_fc_trigger_did_lookup(port);
|
2009-08-18 20:43:20 +07:00
|
|
|
return ZFCP_ERP_EXIT;
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
return zfcp_erp_port_strategy_open_port(act);
|
|
|
|
|
|
|
|
case ZFCP_ERP_STEP_PORT_OPENING:
|
|
|
|
/* D_ID might have changed during open */
|
2008-10-01 17:42:17 +07:00
|
|
|
if (p_status & ZFCP_STATUS_COMMON_OPEN) {
|
2009-10-14 16:00:43 +07:00
|
|
|
if (!port->d_id) {
|
|
|
|
zfcp_fc_trigger_did_lookup(port);
|
|
|
|
return ZFCP_ERP_EXIT;
|
2008-10-01 17:42:17 +07:00
|
|
|
}
|
2009-10-14 16:00:43 +07:00
|
|
|
return ZFCP_ERP_SUCCEEDED;
|
2008-10-01 17:42:17 +07:00
|
|
|
}
|
2009-05-15 18:18:20 +07:00
|
|
|
if (port->d_id && !(p_status & ZFCP_STATUS_COMMON_NOESC)) {
|
|
|
|
port->d_id = 0;
|
2010-07-08 14:53:10 +07:00
|
|
|
return ZFCP_ERP_FAILED;
|
2009-05-15 18:18:20 +07:00
|
|
|
}
|
|
|
|
/* fall through otherwise */
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int zfcp_erp_port_strategy(struct zfcp_erp_action *erp_action)
|
|
|
|
{
|
|
|
|
struct zfcp_port *port = erp_action->port;
|
2009-10-14 16:00:43 +07:00
|
|
|
int p_status = atomic_read(&port->status);
|
2008-07-02 15:56:40 +07:00
|
|
|
|
2009-10-14 16:00:43 +07:00
|
|
|
if ((p_status & ZFCP_STATUS_COMMON_NOESC) &&
|
|
|
|
!(p_status & ZFCP_STATUS_COMMON_OPEN))
|
2008-10-01 17:42:17 +07:00
|
|
|
goto close_init_done;
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
switch (erp_action->step) {
|
|
|
|
case ZFCP_ERP_STEP_UNINITIALIZED:
|
2009-10-14 16:00:43 +07:00
|
|
|
if (p_status & ZFCP_STATUS_COMMON_OPEN)
|
2008-07-02 15:56:40 +07:00
|
|
|
return zfcp_erp_port_strategy_close(erp_action);
|
2005-04-17 05:20:36 +07:00
|
|
|
break;
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
case ZFCP_ERP_STEP_PORT_CLOSING:
|
2009-10-14 16:00:43 +07:00
|
|
|
if (p_status & ZFCP_STATUS_COMMON_OPEN)
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_FAILED;
|
2005-04-17 05:20:36 +07:00
|
|
|
break;
|
|
|
|
}
|
2008-10-01 17:42:17 +07:00
|
|
|
|
|
|
|
close_init_done:
|
2008-07-02 15:56:40 +07:00
|
|
|
if (erp_action->status & ZFCP_STATUS_ERP_CLOSE_ONLY)
|
|
|
|
return ZFCP_ERP_EXIT;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-10-01 17:42:17 +07:00
|
|
|
return zfcp_erp_port_strategy_open_common(erp_action);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static void zfcp_erp_lun_strategy_clearstati(struct scsi_device *sdev)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_COMMON_ACCESS_DENIED,
|
2010-09-08 19:39:55 +07:00
|
|
|
&zfcp_sdev->status);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static int zfcp_erp_lun_strategy_close(struct zfcp_erp_action *erp_action)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
int retval = zfcp_fsf_close_lun(erp_action);
|
2008-07-02 15:56:40 +07:00
|
|
|
if (retval == -ENOMEM)
|
|
|
|
return ZFCP_ERP_NOMEM;
|
2010-09-08 19:39:55 +07:00
|
|
|
erp_action->step = ZFCP_ERP_STEP_LUN_CLOSING;
|
2008-07-02 15:56:40 +07:00
|
|
|
if (retval)
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
return ZFCP_ERP_CONTINUES;
|
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static int zfcp_erp_lun_strategy_open(struct zfcp_erp_action *erp_action)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
int retval = zfcp_fsf_open_lun(erp_action);
|
2008-07-02 15:56:40 +07:00
|
|
|
if (retval == -ENOMEM)
|
|
|
|
return ZFCP_ERP_NOMEM;
|
2010-09-08 19:39:55 +07:00
|
|
|
erp_action->step = ZFCP_ERP_STEP_LUN_OPENING;
|
2008-07-02 15:56:40 +07:00
|
|
|
if (retval)
|
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
return ZFCP_ERP_CONTINUES;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static int zfcp_erp_lun_strategy(struct zfcp_erp_action *erp_action)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev = erp_action->sdev;
|
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
2008-07-02 15:56:40 +07:00
|
|
|
|
|
|
|
switch (erp_action->step) {
|
|
|
|
case ZFCP_ERP_STEP_UNINITIALIZED:
|
2010-09-08 19:39:55 +07:00
|
|
|
zfcp_erp_lun_strategy_clearstati(sdev);
|
|
|
|
if (atomic_read(&zfcp_sdev->status) & ZFCP_STATUS_COMMON_OPEN)
|
|
|
|
return zfcp_erp_lun_strategy_close(erp_action);
|
2008-07-02 15:56:40 +07:00
|
|
|
/* already closed, fall through */
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_STEP_LUN_CLOSING:
|
|
|
|
if (atomic_read(&zfcp_sdev->status) & ZFCP_STATUS_COMMON_OPEN)
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_FAILED;
|
|
|
|
if (erp_action->status & ZFCP_STATUS_ERP_CLOSE_ONLY)
|
|
|
|
return ZFCP_ERP_EXIT;
|
2010-09-08 19:39:55 +07:00
|
|
|
return zfcp_erp_lun_strategy_open(erp_action);
|
2008-07-02 15:56:40 +07:00
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_STEP_LUN_OPENING:
|
|
|
|
if (atomic_read(&zfcp_sdev->status) & ZFCP_STATUS_COMMON_OPEN)
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_SUCCEEDED;
|
|
|
|
}
|
|
|
|
return ZFCP_ERP_FAILED;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
static int zfcp_erp_strategy_check_lun(struct scsi_device *sdev, int result)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
switch (result) {
|
|
|
|
case ZFCP_ERP_SUCCEEDED :
|
2010-09-08 19:39:55 +07:00
|
|
|
atomic_set(&zfcp_sdev->erp_counter, 0);
|
|
|
|
zfcp_erp_lun_unblock(sdev);
|
2005-04-17 05:20:36 +07:00
|
|
|
break;
|
|
|
|
case ZFCP_ERP_FAILED :
|
2010-09-08 19:39:55 +07:00
|
|
|
atomic_inc(&zfcp_sdev->erp_counter);
|
|
|
|
if (atomic_read(&zfcp_sdev->erp_counter) > ZFCP_MAX_ERPS) {
|
|
|
|
dev_err(&zfcp_sdev->port->adapter->ccw_device->dev,
|
|
|
|
"ERP failed for LUN 0x%016Lx on "
|
2008-10-01 17:42:15 +07:00
|
|
|
"port 0x%016Lx\n",
|
2010-09-08 19:39:55 +07:00
|
|
|
(unsigned long long)zfcp_scsi_dev_lun(sdev),
|
|
|
|
(unsigned long long)zfcp_sdev->port->wwpn);
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_set_lun_status(sdev,
|
|
|
|
ZFCP_STATUS_COMMON_ERP_FAILED);
|
2008-10-01 17:42:15 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
if (atomic_read(&zfcp_sdev->status) & ZFCP_STATUS_COMMON_ERP_FAILED) {
|
|
|
|
zfcp_erp_lun_block(sdev, 0);
|
2005-04-17 05:20:36 +07:00
|
|
|
result = ZFCP_ERP_EXIT;
|
|
|
|
}
|
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_strategy_check_port(struct zfcp_port *port, int result)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
switch (result) {
|
|
|
|
case ZFCP_ERP_SUCCEEDED :
|
|
|
|
atomic_set(&port->erp_counter, 0);
|
|
|
|
zfcp_erp_port_unblock(port);
|
|
|
|
break;
|
2008-07-02 15:56:40 +07:00
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
case ZFCP_ERP_FAILED :
|
2008-07-02 15:56:40 +07:00
|
|
|
if (atomic_read(&port->status) & ZFCP_STATUS_COMMON_NOESC) {
|
2008-06-10 23:21:00 +07:00
|
|
|
zfcp_erp_port_block(port, 0);
|
|
|
|
result = ZFCP_ERP_EXIT;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
atomic_inc(&port->erp_counter);
|
2008-10-01 17:42:15 +07:00
|
|
|
if (atomic_read(&port->erp_counter) > ZFCP_MAX_ERPS) {
|
|
|
|
dev_err(&port->adapter->ccw_device->dev,
|
|
|
|
"ERP failed for remote port 0x%016Lx\n",
|
2008-10-01 17:42:18 +07:00
|
|
|
(unsigned long long)port->wwpn);
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_set_port_status(port,
|
|
|
|
ZFCP_STATUS_COMMON_ERP_FAILED);
|
2008-10-01 17:42:15 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (atomic_read(&port->status) & ZFCP_STATUS_COMMON_ERP_FAILED) {
|
|
|
|
zfcp_erp_port_block(port, 0);
|
2005-04-17 05:20:36 +07:00
|
|
|
result = ZFCP_ERP_EXIT;
|
|
|
|
}
|
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_strategy_check_adapter(struct zfcp_adapter *adapter,
|
|
|
|
int result)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
switch (result) {
|
|
|
|
case ZFCP_ERP_SUCCEEDED :
|
|
|
|
atomic_set(&adapter->erp_counter, 0);
|
|
|
|
zfcp_erp_adapter_unblock(adapter);
|
|
|
|
break;
|
2008-07-02 15:56:40 +07:00
|
|
|
|
2005-04-17 05:20:36 +07:00
|
|
|
case ZFCP_ERP_FAILED :
|
|
|
|
atomic_inc(&adapter->erp_counter);
|
2008-10-01 17:42:15 +07:00
|
|
|
if (atomic_read(&adapter->erp_counter) > ZFCP_MAX_ERPS) {
|
|
|
|
dev_err(&adapter->ccw_device->dev,
|
|
|
|
"ERP cannot recover an error "
|
|
|
|
"on the FCP device\n");
|
2010-09-08 19:40:01 +07:00
|
|
|
zfcp_erp_set_adapter_status(adapter,
|
|
|
|
ZFCP_STATUS_COMMON_ERP_FAILED);
|
2008-10-01 17:42:15 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (atomic_read(&adapter->status) & ZFCP_STATUS_COMMON_ERP_FAILED) {
|
|
|
|
zfcp_erp_adapter_block(adapter, 0);
|
2005-04-17 05:20:36 +07:00
|
|
|
result = ZFCP_ERP_EXIT;
|
|
|
|
}
|
|
|
|
return result;
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_strategy_check_target(struct zfcp_erp_action *erp_action,
|
|
|
|
int result)
|
2007-05-08 16:16:52 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_adapter *adapter = erp_action->adapter;
|
|
|
|
struct zfcp_port *port = erp_action->port;
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev = erp_action->sdev;
|
2008-07-02 15:56:40 +07:00
|
|
|
|
|
|
|
switch (erp_action->action) {
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_LUN:
|
|
|
|
result = zfcp_erp_strategy_check_lun(sdev, result);
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
|
|
|
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED:
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT:
|
|
|
|
result = zfcp_erp_strategy_check_port(port, result);
|
|
|
|
break;
|
|
|
|
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_ADAPTER:
|
|
|
|
result = zfcp_erp_strategy_check_adapter(adapter, result);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
return result;
|
2007-05-08 16:16:52 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_strat_change_det(atomic_t *target_status, u32 erp_status)
|
2007-05-08 16:16:52 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int status = atomic_read(target_status);
|
2007-05-08 16:16:52 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if ((status & ZFCP_STATUS_COMMON_RUNNING) &&
|
|
|
|
(erp_status & ZFCP_STATUS_ERP_CLOSE_ONLY))
|
|
|
|
return 1; /* take it online */
|
2007-05-08 16:16:52 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (!(status & ZFCP_STATUS_COMMON_RUNNING) &&
|
|
|
|
!(erp_status & ZFCP_STATUS_ERP_CLOSE_ONLY))
|
|
|
|
return 1; /* take it offline */
|
|
|
|
|
|
|
|
return 0;
|
2007-05-08 16:16:52 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_strategy_statechange(struct zfcp_erp_action *act, int ret)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
int action = act->action;
|
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
|
|
|
struct zfcp_port *port = act->port;
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev = act->sdev;
|
|
|
|
struct zfcp_scsi_dev *zfcp_sdev;
|
2008-07-02 15:56:40 +07:00
|
|
|
u32 erp_status = act->status;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
switch (action) {
|
2005-04-17 05:20:36 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_ADAPTER:
|
2008-07-02 15:56:40 +07:00
|
|
|
if (zfcp_erp_strat_change_det(&adapter->status, erp_status)) {
|
|
|
|
_zfcp_erp_adapter_reopen(adapter,
|
|
|
|
ZFCP_STATUS_COMMON_ERP_FAILED,
|
2010-12-02 21:16:16 +07:00
|
|
|
"ersscg1");
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_EXIT;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
break;
|
|
|
|
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED:
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT:
|
2008-07-02 15:56:40 +07:00
|
|
|
if (zfcp_erp_strat_change_det(&port->status, erp_status)) {
|
|
|
|
_zfcp_erp_port_reopen(port,
|
|
|
|
ZFCP_STATUS_COMMON_ERP_FAILED,
|
2010-12-02 21:16:16 +07:00
|
|
|
"ersscg2");
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_EXIT;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
break;
|
|
|
|
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_LUN:
|
|
|
|
zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
if (zfcp_erp_strat_change_det(&zfcp_sdev->status, erp_status)) {
|
|
|
|
_zfcp_erp_lun_reopen(sdev,
|
|
|
|
ZFCP_STATUS_COMMON_ERP_FAILED,
|
2010-12-02 21:16:16 +07:00
|
|
|
"ersscg3", 0);
|
2008-07-02 15:56:40 +07:00
|
|
|
return ZFCP_ERP_EXIT;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
break;
|
|
|
|
}
|
2008-07-02 15:56:40 +07:00
|
|
|
return ret;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_action_dequeue(struct zfcp_erp_action *erp_action)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_adapter *adapter = erp_action->adapter;
|
2010-09-08 19:39:55 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
adapter->erp_total_count--;
|
|
|
|
if (erp_action->status & ZFCP_STATUS_ERP_LOWMEM) {
|
|
|
|
adapter->erp_low_mem_count--;
|
|
|
|
erp_action->status &= ~ZFCP_STATUS_ERP_LOWMEM;
|
2008-03-27 20:22:05 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
list_del(&erp_action->list);
|
2010-12-02 21:16:12 +07:00
|
|
|
zfcp_dbf_rec_run("eractd1", erp_action);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
switch (erp_action->action) {
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_LUN:
|
|
|
|
zfcp_sdev = sdev_to_zfcp(erp_action->sdev);
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_COMMON_ERP_INUSE,
|
2010-09-08 19:39:55 +07:00
|
|
|
&zfcp_sdev->status);
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED:
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT:
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_COMMON_ERP_INUSE,
|
2008-07-02 15:56:40 +07:00
|
|
|
&erp_action->port->status);
|
|
|
|
break;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_ADAPTER:
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(ZFCP_STATUS_COMMON_ERP_INUSE,
|
2008-07-02 15:56:40 +07:00
|
|
|
&erp_action->adapter->status);
|
|
|
|
break;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static void zfcp_erp_action_cleanup(struct zfcp_erp_action *act, int result)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_adapter *adapter = act->adapter;
|
|
|
|
struct zfcp_port *port = act->port;
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev = act->sdev;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
switch (act->action) {
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_LUN:
|
2010-09-08 19:39:54 +07:00
|
|
|
if (!(act->status & ZFCP_STATUS_ERP_NO_REF))
|
2010-09-08 19:39:55 +07:00
|
|
|
scsi_device_put(sdev);
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT:
|
2009-03-02 19:09:08 +07:00
|
|
|
if (result == ZFCP_ERP_SUCCEEDED)
|
|
|
|
zfcp_scsi_schedule_rport_register(port);
|
2010-07-08 14:53:05 +07:00
|
|
|
/* fall through */
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED:
|
2010-02-17 17:18:56 +07:00
|
|
|
put_device(&port->dev);
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
|
|
|
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_ADAPTER:
|
2009-03-02 19:09:08 +07:00
|
|
|
if (result == ZFCP_ERP_SUCCEEDED) {
|
2008-12-25 19:38:50 +07:00
|
|
|
register_service_level(&adapter->service_level);
|
2012-09-04 20:23:35 +07:00
|
|
|
zfcp_fc_conditional_port_scan(adapter);
|
2011-02-23 01:54:48 +07:00
|
|
|
queue_work(adapter->work_queue, &adapter->ns_up_work);
|
2009-03-02 19:09:08 +07:00
|
|
|
} else
|
|
|
|
unregister_service_level(&adapter->service_level);
|
2011-02-23 01:54:48 +07:00
|
|
|
|
2009-11-24 22:53:59 +07:00
|
|
|
kref_put(&adapter->ref, zfcp_adapter_release);
|
2008-07-02 15:56:40 +07:00
|
|
|
break;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_strategy_do_action(struct zfcp_erp_action *erp_action)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
switch (erp_action->action) {
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_ADAPTER:
|
|
|
|
return zfcp_erp_adapter_strategy(erp_action);
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT_FORCED:
|
|
|
|
return zfcp_erp_port_forced_strategy(erp_action);
|
|
|
|
case ZFCP_ERP_ACTION_REOPEN_PORT:
|
|
|
|
return zfcp_erp_port_strategy(erp_action);
|
2010-09-08 19:39:55 +07:00
|
|
|
case ZFCP_ERP_ACTION_REOPEN_LUN:
|
|
|
|
return zfcp_erp_lun_strategy(erp_action);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
|
|
|
return ZFCP_ERP_FAILED;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_strategy(struct zfcp_erp_action *erp_action)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
int retval;
|
2008-07-02 15:56:40 +07:00
|
|
|
unsigned long flags;
|
2009-11-24 22:53:58 +07:00
|
|
|
struct zfcp_adapter *adapter = erp_action->adapter;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-11-24 22:53:59 +07:00
|
|
|
kref_get(&adapter->ref);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-11-24 22:53:59 +07:00
|
|
|
write_lock_irqsave(&adapter->erp_lock, flags);
|
2008-07-02 15:56:40 +07:00
|
|
|
zfcp_erp_strategy_check_fsfreq(erp_action);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (erp_action->status & ZFCP_STATUS_ERP_DISMISSED) {
|
|
|
|
zfcp_erp_action_dequeue(erp_action);
|
|
|
|
retval = ZFCP_ERP_DISMISSED;
|
|
|
|
goto unlock;
|
2005-06-13 18:15:15 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-07-08 14:53:09 +07:00
|
|
|
if (erp_action->status & ZFCP_STATUS_ERP_TIMEDOUT) {
|
|
|
|
retval = ZFCP_ERP_FAILED;
|
|
|
|
goto check_target;
|
|
|
|
}
|
|
|
|
|
2006-02-11 07:41:50 +07:00
|
|
|
zfcp_erp_action_to_running(erp_action);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/* no lock to allow for blocking operations */
|
2009-11-24 22:53:58 +07:00
|
|
|
write_unlock_irqrestore(&adapter->erp_lock, flags);
|
2008-07-02 15:56:40 +07:00
|
|
|
retval = zfcp_erp_strategy_do_action(erp_action);
|
2009-11-24 22:53:58 +07:00
|
|
|
write_lock_irqsave(&adapter->erp_lock, flags);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (erp_action->status & ZFCP_STATUS_ERP_DISMISSED)
|
|
|
|
retval = ZFCP_ERP_CONTINUES;
|
2008-06-10 23:21:00 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
switch (retval) {
|
|
|
|
case ZFCP_ERP_NOMEM:
|
|
|
|
if (!(erp_action->status & ZFCP_STATUS_ERP_LOWMEM)) {
|
|
|
|
++adapter->erp_low_mem_count;
|
|
|
|
erp_action->status |= ZFCP_STATUS_ERP_LOWMEM;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2008-07-02 15:56:40 +07:00
|
|
|
if (adapter->erp_total_count == adapter->erp_low_mem_count)
|
2010-12-02 21:16:16 +07:00
|
|
|
_zfcp_erp_adapter_reopen(adapter, 0, "erstgy1");
|
2008-07-02 15:56:40 +07:00
|
|
|
else {
|
|
|
|
zfcp_erp_strategy_memwait(erp_action);
|
|
|
|
retval = ZFCP_ERP_CONTINUES;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2008-07-02 15:56:40 +07:00
|
|
|
goto unlock;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
case ZFCP_ERP_CONTINUES:
|
|
|
|
if (erp_action->status & ZFCP_STATUS_ERP_LOWMEM) {
|
|
|
|
--adapter->erp_low_mem_count;
|
|
|
|
erp_action->status &= ~ZFCP_STATUS_ERP_LOWMEM;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2008-07-02 15:56:40 +07:00
|
|
|
goto unlock;
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2010-07-08 14:53:09 +07:00
|
|
|
check_target:
|
2008-07-02 15:56:40 +07:00
|
|
|
retval = zfcp_erp_strategy_check_target(erp_action, retval);
|
|
|
|
zfcp_erp_action_dequeue(erp_action);
|
|
|
|
retval = zfcp_erp_strategy_statechange(erp_action, retval);
|
|
|
|
if (retval == ZFCP_ERP_EXIT)
|
|
|
|
goto unlock;
|
2009-07-13 20:06:09 +07:00
|
|
|
if (retval == ZFCP_ERP_SUCCEEDED)
|
|
|
|
zfcp_erp_strategy_followup_success(erp_action);
|
|
|
|
if (retval == ZFCP_ERP_FAILED)
|
|
|
|
zfcp_erp_strategy_followup_failed(erp_action);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
unlock:
|
2009-11-24 22:53:58 +07:00
|
|
|
write_unlock_irqrestore(&adapter->erp_lock, flags);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (retval != ZFCP_ERP_CONTINUES)
|
|
|
|
zfcp_erp_action_cleanup(erp_action, retval);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-11-24 22:53:59 +07:00
|
|
|
kref_put(&adapter->ref, zfcp_adapter_release);
|
2005-04-17 05:20:36 +07:00
|
|
|
return retval;
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
static int zfcp_erp_thread(void *data)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2008-07-02 15:56:40 +07:00
|
|
|
struct zfcp_adapter *adapter = (struct zfcp_adapter *) data;
|
|
|
|
struct list_head *next;
|
|
|
|
struct zfcp_erp_action *act;
|
|
|
|
unsigned long flags;
|
2009-04-17 20:08:06 +07:00
|
|
|
|
2009-08-18 20:43:25 +07:00
|
|
|
for (;;) {
|
|
|
|
wait_event_interruptible(adapter->erp_ready_wq,
|
|
|
|
!list_empty(&adapter->erp_ready_head) ||
|
|
|
|
kthread_should_stop());
|
2009-04-17 20:08:06 +07:00
|
|
|
|
2009-08-18 20:43:25 +07:00
|
|
|
if (kthread_should_stop())
|
|
|
|
break;
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
write_lock_irqsave(&adapter->erp_lock, flags);
|
|
|
|
next = adapter->erp_ready_head.next;
|
|
|
|
write_unlock_irqrestore(&adapter->erp_lock, flags);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
if (next != &adapter->erp_ready_head) {
|
|
|
|
act = list_entry(next, struct zfcp_erp_action, list);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/* there is more to come after dismission, no notify */
|
|
|
|
if (zfcp_erp_strategy(act) != ZFCP_ERP_DISMISSED)
|
|
|
|
zfcp_erp_wakeup(adapter);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
return 0;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
|
|
|
* zfcp_erp_thread_setup - Start ERP thread for adapter
|
|
|
|
* @adapter: Adapter to start the ERP thread for
|
|
|
|
*
|
|
|
|
* Returns 0 on success or error code from kernel_thread()
|
|
|
|
*/
|
|
|
|
int zfcp_erp_thread_setup(struct zfcp_adapter *adapter)
|
|
|
|
{
|
2009-08-18 20:43:25 +07:00
|
|
|
struct task_struct *thread;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2009-08-18 20:43:25 +07:00
|
|
|
thread = kthread_run(zfcp_erp_thread, adapter, "zfcperp%s",
|
|
|
|
dev_name(&adapter->ccw_device->dev));
|
|
|
|
if (IS_ERR(thread)) {
|
2008-07-02 15:56:40 +07:00
|
|
|
dev_err(&adapter->ccw_device->dev,
|
2008-10-01 17:42:15 +07:00
|
|
|
"Creating an ERP thread for the FCP device failed.\n");
|
2009-08-18 20:43:25 +07:00
|
|
|
return PTR_ERR(thread);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
2009-08-18 20:43:25 +07:00
|
|
|
|
|
|
|
adapter->erp_thread = thread;
|
2008-07-02 15:56:40 +07:00
|
|
|
return 0;
|
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
|
|
|
* zfcp_erp_thread_kill - Stop ERP thread.
|
|
|
|
* @adapter: Adapter where the ERP thread should be stopped.
|
|
|
|
*
|
|
|
|
* The caller of this routine ensures that the specified adapter has
|
|
|
|
* been shut down and that this operation has been completed. Thus,
|
|
|
|
* there are no pending erp_actions which would need to be handled
|
|
|
|
* here.
|
|
|
|
*/
|
|
|
|
void zfcp_erp_thread_kill(struct zfcp_adapter *adapter)
|
|
|
|
{
|
2009-08-18 20:43:25 +07:00
|
|
|
kthread_stop(adapter->erp_thread);
|
|
|
|
adapter->erp_thread = NULL;
|
2009-08-18 20:43:27 +07:00
|
|
|
WARN_ON(!list_empty(&adapter->erp_ready_head));
|
|
|
|
WARN_ON(!list_empty(&adapter->erp_running_head));
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
2010-09-08 19:40:01 +07:00
|
|
|
* zfcp_erp_wait - wait for completion of error recovery on an adapter
|
|
|
|
* @adapter: adapter for which to wait for completion of its error recovery
|
2008-07-02 15:56:40 +07:00
|
|
|
*/
|
2010-09-08 19:40:01 +07:00
|
|
|
void zfcp_erp_wait(struct zfcp_adapter *adapter)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-09-08 19:40:01 +07:00
|
|
|
wait_event(adapter->erp_done_wqh,
|
|
|
|
!(atomic_read(&adapter->status) &
|
|
|
|
ZFCP_STATUS_ADAPTER_ERP_PENDING));
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
2010-09-08 19:40:01 +07:00
|
|
|
* zfcp_erp_set_adapter_status - set adapter status bits
|
|
|
|
* @adapter: adapter to change the status
|
|
|
|
* @mask: status bits to change
|
|
|
|
*
|
|
|
|
* Changes in common status bits are propagated to attached ports and LUNs.
|
2008-07-02 15:56:40 +07:00
|
|
|
*/
|
2010-09-08 19:40:01 +07:00
|
|
|
void zfcp_erp_set_adapter_status(struct zfcp_adapter *adapter, u32 mask)
|
2008-07-02 15:56:40 +07:00
|
|
|
{
|
2010-09-08 19:40:01 +07:00
|
|
|
struct zfcp_port *port;
|
|
|
|
struct scsi_device *sdev;
|
|
|
|
unsigned long flags;
|
|
|
|
u32 common_mask = mask & ZFCP_COMMON_FLAGS;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(mask, &adapter->status);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-09-08 19:40:01 +07:00
|
|
|
if (!common_mask)
|
|
|
|
return;
|
|
|
|
|
|
|
|
read_lock_irqsave(&adapter->port_list_lock, flags);
|
|
|
|
list_for_each_entry(port, &adapter->port_list, list)
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(common_mask, &port->status);
|
2010-09-08 19:40:01 +07:00
|
|
|
read_unlock_irqrestore(&adapter->port_list_lock, flags);
|
|
|
|
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_lock_irqsave(adapter->scsi_host->host_lock, flags);
|
|
|
|
__shost_for_each_device(sdev, adapter->scsi_host)
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(common_mask, &sdev_to_zfcp(sdev)->status);
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_unlock_irqrestore(adapter->scsi_host->host_lock, flags);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
2010-09-08 19:40:01 +07:00
|
|
|
* zfcp_erp_clear_adapter_status - clear adapter status bits
|
2008-07-02 15:56:40 +07:00
|
|
|
* @adapter: adapter to change the status
|
|
|
|
* @mask: status bits to change
|
|
|
|
*
|
2010-09-08 19:39:55 +07:00
|
|
|
* Changes in common status bits are propagated to attached ports and LUNs.
|
2008-07-02 15:56:40 +07:00
|
|
|
*/
|
2010-09-08 19:40:01 +07:00
|
|
|
void zfcp_erp_clear_adapter_status(struct zfcp_adapter *adapter, u32 mask)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
|
|
|
struct zfcp_port *port;
|
2010-09-08 19:40:01 +07:00
|
|
|
struct scsi_device *sdev;
|
2009-11-24 22:53:58 +07:00
|
|
|
unsigned long flags;
|
2008-07-02 15:56:40 +07:00
|
|
|
u32 common_mask = mask & ZFCP_COMMON_FLAGS;
|
2010-09-08 19:40:01 +07:00
|
|
|
u32 clear_counter = mask & ZFCP_STATUS_COMMON_ERP_FAILED;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(mask, &adapter->status);
|
2010-09-08 19:40:01 +07:00
|
|
|
|
|
|
|
if (!common_mask)
|
|
|
|
return;
|
|
|
|
|
|
|
|
if (clear_counter)
|
|
|
|
atomic_set(&adapter->erp_counter, 0);
|
|
|
|
|
|
|
|
read_lock_irqsave(&adapter->port_list_lock, flags);
|
|
|
|
list_for_each_entry(port, &adapter->port_list, list) {
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(common_mask, &port->status);
|
2010-09-08 19:40:01 +07:00
|
|
|
if (clear_counter)
|
|
|
|
atomic_set(&port->erp_counter, 0);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
2010-09-08 19:40:01 +07:00
|
|
|
read_unlock_irqrestore(&adapter->port_list_lock, flags);
|
2008-07-02 15:56:40 +07:00
|
|
|
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_lock_irqsave(adapter->scsi_host->host_lock, flags);
|
|
|
|
__shost_for_each_device(sdev, adapter->scsi_host) {
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(common_mask, &sdev_to_zfcp(sdev)->status);
|
2010-09-08 19:40:01 +07:00
|
|
|
if (clear_counter)
|
|
|
|
atomic_set(&sdev_to_zfcp(sdev)->erp_counter, 0);
|
2009-11-24 22:53:58 +07:00
|
|
|
}
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_unlock_irqrestore(adapter->scsi_host->host_lock, flags);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
2010-09-08 19:40:01 +07:00
|
|
|
* zfcp_erp_set_port_status - set port status bits
|
|
|
|
* @port: port to change the status
|
2008-07-02 15:56:40 +07:00
|
|
|
* @mask: status bits to change
|
|
|
|
*
|
2010-09-08 19:39:55 +07:00
|
|
|
* Changes in common status bits are propagated to attached LUNs.
|
2008-07-02 15:56:40 +07:00
|
|
|
*/
|
2010-09-08 19:40:01 +07:00
|
|
|
void zfcp_erp_set_port_status(struct zfcp_port *port, u32 mask)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-09-08 19:39:55 +07:00
|
|
|
struct scsi_device *sdev;
|
2008-07-02 15:56:40 +07:00
|
|
|
u32 common_mask = mask & ZFCP_COMMON_FLAGS;
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
unsigned long flags;
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(mask, &port->status);
|
2005-04-17 05:20:36 +07:00
|
|
|
|
2010-09-08 19:40:01 +07:00
|
|
|
if (!common_mask)
|
|
|
|
return;
|
|
|
|
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_lock_irqsave(port->adapter->scsi_host->host_lock, flags);
|
|
|
|
__shost_for_each_device(sdev, port->adapter->scsi_host)
|
2010-09-08 19:40:01 +07:00
|
|
|
if (sdev_to_zfcp(sdev)->port == port)
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(common_mask,
|
2010-09-08 19:40:01 +07:00
|
|
|
&sdev_to_zfcp(sdev)->status);
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_unlock_irqrestore(port->adapter->scsi_host->host_lock, flags);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
2010-09-08 19:40:01 +07:00
|
|
|
* zfcp_erp_clear_port_status - clear port status bits
|
|
|
|
* @port: adapter to change the status
|
2008-07-02 15:56:40 +07:00
|
|
|
* @mask: status bits to change
|
2010-09-08 19:40:01 +07:00
|
|
|
*
|
|
|
|
* Changes in common status bits are propagated to attached LUNs.
|
2008-07-02 15:56:40 +07:00
|
|
|
*/
|
2010-09-08 19:40:01 +07:00
|
|
|
void zfcp_erp_clear_port_status(struct zfcp_port *port, u32 mask)
|
2005-04-17 05:20:36 +07:00
|
|
|
{
|
2010-09-08 19:40:01 +07:00
|
|
|
struct scsi_device *sdev;
|
|
|
|
u32 common_mask = mask & ZFCP_COMMON_FLAGS;
|
|
|
|
u32 clear_counter = mask & ZFCP_STATUS_COMMON_ERP_FAILED;
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
unsigned long flags;
|
2010-09-08 19:40:01 +07:00
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(mask, &port->status);
|
2010-09-08 19:40:01 +07:00
|
|
|
|
|
|
|
if (!common_mask)
|
|
|
|
return;
|
2010-09-08 19:39:55 +07:00
|
|
|
|
2010-09-08 19:40:01 +07:00
|
|
|
if (clear_counter)
|
|
|
|
atomic_set(&port->erp_counter, 0);
|
|
|
|
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_lock_irqsave(port->adapter->scsi_host->host_lock, flags);
|
|
|
|
__shost_for_each_device(sdev, port->adapter->scsi_host)
|
2010-09-08 19:40:01 +07:00
|
|
|
if (sdev_to_zfcp(sdev)->port == port) {
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(common_mask,
|
2010-09-08 19:40:01 +07:00
|
|
|
&sdev_to_zfcp(sdev)->status);
|
|
|
|
if (clear_counter)
|
|
|
|
atomic_set(&sdev_to_zfcp(sdev)->erp_counter, 0);
|
2008-07-02 15:56:40 +07:00
|
|
|
}
|
[SCSI] zfcp: fix schedule-inside-lock in scsi_device list loops
BUG: sleeping function called from invalid context at kernel/workqueue.c:2752
in_atomic(): 1, irqs_disabled(): 1, pid: 360, name: zfcperp0.0.1700
CPU: 1 Not tainted 3.9.3+ #69
Process zfcperp0.0.1700 (pid: 360, task: 0000000075b7e080, ksp: 000000007476bc30)
<snip>
Call Trace:
([<00000000001165de>] show_trace+0x106/0x154)
[<00000000001166a0>] show_stack+0x74/0xf4
[<00000000006ff646>] dump_stack+0xc6/0xd4
[<000000000017f3a0>] __might_sleep+0x128/0x148
[<000000000015ece8>] flush_work+0x54/0x1f8
[<00000000001630de>] __cancel_work_timer+0xc6/0x128
[<00000000005067ac>] scsi_device_dev_release_usercontext+0x164/0x23c
[<0000000000161816>] execute_in_process_context+0x96/0xa8
[<00000000004d33d8>] device_release+0x60/0xc0
[<000000000048af48>] kobject_release+0xa8/0x1c4
[<00000000004f4bf2>] __scsi_iterate_devices+0xfa/0x130
[<000003ff801b307a>] zfcp_erp_strategy+0x4da/0x1014 [zfcp]
[<000003ff801b3caa>] zfcp_erp_thread+0xf6/0x2b0 [zfcp]
[<000000000016b75a>] kthread+0xf2/0xfc
[<000000000070c9de>] kernel_thread_starter+0x6/0xc
[<000000000070c9d8>] kernel_thread_starter+0x0/0xc
Apparently, the ref_count for some scsi_device drops down to zero,
triggering device removal through execute_in_process_context(), while
the lldd error recovery thread iterates through a scsi device list.
Unfortunately, execute_in_process_context() decides to immediately
execute that device removal function, instead of scheduling asynchronous
execution, since it detects process context and thinks it is safe to do
so. But almost all calls to shost_for_each_device() in our lldd are
inside spin_lock_irq, even in thread context. Obviously, schedule()
inside spin_lock_irq sections is a bad idea.
Change the lldd to use the proper iterator function,
__shost_for_each_device(), in combination with required locking.
Occurences that need to be changed include all calls in zfcp_erp.c,
since those might be executed in zfcp error recovery thread context
with a lock held.
Other occurences of shost_for_each_device() in zfcp_fsf.c do not
need to be changed (no process context, no surrounding locking).
The problem was introduced in Linux 2.6.37 by commit
b62a8d9b45b971a67a0f8413338c230e3117dff5
"[SCSI] zfcp: Use SCSI device data zfcp_scsi_dev instead of zfcp_unit".
Reported-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Peschke <mpeschke@linux.vnet.ibm.com>
Cc: stable@vger.kernel.org #2.6.37+
Signed-off-by: Steffen Maier <maier@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>
2013-08-22 22:45:37 +07:00
|
|
|
spin_unlock_irqrestore(port->adapter->scsi_host->host_lock, flags);
|
2005-04-17 05:20:36 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
2010-09-08 19:40:01 +07:00
|
|
|
* zfcp_erp_set_lun_status - set lun status bits
|
|
|
|
* @sdev: SCSI device / lun to set the status bits
|
|
|
|
* @mask: status bits to change
|
2008-07-02 15:56:40 +07:00
|
|
|
*/
|
2010-09-08 19:40:01 +07:00
|
|
|
void zfcp_erp_set_lun_status(struct scsi_device *sdev, u32 mask)
|
2005-06-13 18:23:57 +07:00
|
|
|
{
|
2010-09-08 19:40:01 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_or(mask, &zfcp_sdev->status);
|
2005-06-13 18:23:57 +07:00
|
|
|
}
|
|
|
|
|
2008-07-02 15:56:40 +07:00
|
|
|
/**
|
2010-09-08 19:40:01 +07:00
|
|
|
* zfcp_erp_clear_lun_status - clear lun status bits
|
|
|
|
* @sdev: SCSi device / lun to clear the status bits
|
|
|
|
* @mask: status bits to change
|
2008-07-02 15:56:40 +07:00
|
|
|
*/
|
2010-09-08 19:40:01 +07:00
|
|
|
void zfcp_erp_clear_lun_status(struct scsi_device *sdev, u32 mask)
|
2005-06-13 18:23:57 +07:00
|
|
|
{
|
2010-09-08 19:40:01 +07:00
|
|
|
struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev);
|
|
|
|
|
2015-04-24 06:12:32 +07:00
|
|
|
atomic_andnot(mask, &zfcp_sdev->status);
|
2010-09-08 19:40:01 +07:00
|
|
|
|
|
|
|
if (mask & ZFCP_STATUS_COMMON_ERP_FAILED)
|
|
|
|
atomic_set(&zfcp_sdev->erp_counter, 0);
|
2005-06-13 18:23:57 +07:00
|
|
|
}
|
2010-09-08 19:40:01 +07:00
|
|
|
|