[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20110216002106.032445595@clark.kroah.org>
Date: Tue, 15 Feb 2011 16:20:37 -0800
From: Greg KH <gregkh@...e.de>
To: linux-kernel@...r.kernel.org, stable@...nel.org
Cc: stable-review@...nel.org, torvalds@...ux-foundation.org,
akpm@...ux-foundation.org, alan@...rguk.ukuu.org.uk,
Kashyap Desai <kashyap.desai@....com>,
James Bottomley <James.Bottomley@...e.de>
Subject: [patch 069/176] [SCSI] mpt2sas: Fix the race between broadcast asyn event and scsi command completion
2.6.36-stable review patch. If anyone has any objections, please let us know.
------------------
From: Kashyap, Desai <kashyap.desai@....com>
commit ec07a053597bdab51cbd23619f9f9f392712508a upstream.
False timeout after hard resets, there were two issues which leads
to timeout.
(1) Panic because of invalid memory access in the broadcast asyn
event processing routine due to a race between accessing the scsi command
pointer from broadcast asyn event processing thread and completing
the same scsi command from the interrupt context.
(2) Broadcast asyn event notifcations are not handled due to events
ignored while the broadcast asyn event is activity being processed
from the event process kernel thread.
In addition, changed the ABRT_TASK_SET to ABORT_TASK in the
broadcast async event processing routine. This is less disruptive to other
request that generate Broadcast Asyn Primitives besides target
reset. e.g clear reservations, microcode download,and mode select.
Signed-off-by: Kashyap Desai <kashyap.desai@....com>
Signed-off-by: James Bottomley <James.Bottomley@...e.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@...e.de>
---
drivers/scsi/mpt2sas/mpt2sas_scsih.c | 54 +++++++++++++++++++++++++++++------
1 file changed, 45 insertions(+), 9 deletions(-)
--- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c
+++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c
@@ -819,7 +819,7 @@ _scsih_is_end_device(u32 device_info)
}
/**
- * mptscsih_get_scsi_lookup - returns scmd entry
+ * _scsih_scsi_lookup_get - returns scmd entry
* @ioc: per adapter object
* @smid: system request message index
*
@@ -832,6 +832,28 @@ _scsih_scsi_lookup_get(struct MPT2SAS_AD
}
/**
+ * _scsih_scsi_lookup_get_clear - returns scmd entry
+ * @ioc: per adapter object
+ * @smid: system request message index
+ *
+ * Returns the smid stored scmd pointer.
+ * Then will derefrence the stored scmd pointer.
+ */
+static inline struct scsi_cmnd *
+_scsih_scsi_lookup_get_clear(struct MPT2SAS_ADAPTER *ioc, u16 smid)
+{
+ unsigned long flags;
+ struct scsi_cmnd *scmd;
+
+ spin_lock_irqsave(&ioc->scsi_lookup_lock, flags);
+ scmd = ioc->scsi_lookup[smid - 1].scmd;
+ ioc->scsi_lookup[smid - 1].scmd = NULL;
+ spin_unlock_irqrestore(&ioc->scsi_lookup_lock, flags);
+
+ return scmd;
+}
+
+/**
* _scsih_scsi_lookup_find_by_scmd - scmd lookup
* @ioc: per adapter object
* @smid: system request message index
@@ -3183,7 +3205,7 @@ _scsih_flush_running_cmds(struct MPT2SAS
u16 count = 0;
for (smid = 1; smid <= ioc->scsiio_depth; smid++) {
- scmd = _scsih_scsi_lookup_get(ioc, smid);
+ scmd = _scsih_scsi_lookup_get_clear(ioc, smid);
if (!scmd)
continue;
count++;
@@ -3775,7 +3797,7 @@ _scsih_io_done(struct MPT2SAS_ADAPTER *i
u32 response_code = 0;
mpi_reply = mpt2sas_base_get_reply_virt_addr(ioc, reply);
- scmd = _scsih_scsi_lookup_get(ioc, smid);
+ scmd = _scsih_scsi_lookup_get_clear(ioc, smid);
if (scmd == NULL)
return 1;
@@ -5037,6 +5059,7 @@ _scsih_sas_broadcast_primative_event(str
struct fw_event_work *fw_event)
{
struct scsi_cmnd *scmd;
+ struct scsi_device *sdev;
u16 smid, handle;
u32 lun;
struct MPT2SAS_DEVICE *sas_device_priv_data;
@@ -5047,12 +5070,17 @@ _scsih_sas_broadcast_primative_event(str
Mpi2EventDataSasBroadcastPrimitive_t *event_data = fw_event->event_data;
#endif
u16 ioc_status;
+ unsigned long flags;
+ int r;
+
dewtprintk(ioc, printk(MPT2SAS_INFO_FMT "broadcast primative: "
"phy number(%d), width(%d)\n", ioc->name, event_data->PhyNum,
event_data->PortWidth));
dtmprintk(ioc, printk(MPT2SAS_INFO_FMT "%s: enter\n", ioc->name,
__func__));
+ spin_lock_irqsave(&ioc->scsi_lookup_lock, flags);
+ ioc->broadcast_aen_busy = 0;
termination_count = 0;
query_count = 0;
mpi_reply = ioc->tm_cmds.reply;
@@ -5060,7 +5088,8 @@ _scsih_sas_broadcast_primative_event(str
scmd = _scsih_scsi_lookup_get(ioc, smid);
if (!scmd)
continue;
- sas_device_priv_data = scmd->device->hostdata;
+ sdev = scmd->device;
+ sas_device_priv_data = sdev->hostdata;
if (!sas_device_priv_data || !sas_device_priv_data->sas_target)
continue;
/* skip hidden raid components */
@@ -5076,6 +5105,7 @@ _scsih_sas_broadcast_primative_event(str
lun = sas_device_priv_data->lun;
query_count++;
+ spin_unlock_irqrestore(&ioc->scsi_lookup_lock, flags);
mpt2sas_scsih_issue_tm(ioc, handle, 0, 0, lun,
MPI2_SCSITASKMGMT_TASKTYPE_QUERY_TASK, smid, 30, NULL);
ioc->tm_cmds.status = MPT2_CMD_NOT_USED;
@@ -5085,14 +5115,20 @@ _scsih_sas_broadcast_primative_event(str
(mpi_reply->ResponseCode ==
MPI2_SCSITASKMGMT_RSP_TM_SUCCEEDED ||
mpi_reply->ResponseCode ==
- MPI2_SCSITASKMGMT_RSP_IO_QUEUED_ON_IOC))
+ MPI2_SCSITASKMGMT_RSP_IO_QUEUED_ON_IOC)) {
+ spin_lock_irqsave(&ioc->scsi_lookup_lock, flags);
continue;
-
- mpt2sas_scsih_issue_tm(ioc, handle, 0, 0, lun,
- MPI2_SCSITASKMGMT_TASKTYPE_ABRT_TASK_SET, 0, 30, NULL);
+ }
+ r = mpt2sas_scsih_issue_tm(ioc, handle, sdev->channel, sdev->id,
+ sdev->lun, MPI2_SCSITASKMGMT_TASKTYPE_ABORT_TASK, smid, 30,
+ scmd);
+ if (r == FAILED)
+ sdev_printk(KERN_WARNING, sdev, "task abort: FAILED "
+ "scmd(%p)\n", scmd);
termination_count += le32_to_cpu(mpi_reply->TerminationCount);
+ spin_lock_irqsave(&ioc->scsi_lookup_lock, flags);
}
- ioc->broadcast_aen_busy = 0;
+ spin_unlock_irqrestore(&ioc->scsi_lookup_lock, flags);
dtmprintk(ioc, printk(MPT2SAS_INFO_FMT
"%s - exit, query_count = %d termination_count = %d\n",
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists