From 4dc2757a2e9a9d1f2faee4fc6119276fc0061c16 Mon Sep 17 00:00:00 2001 From: "Kashyap, Desai" Date: Tue, 4 Jan 2011 11:32:13 +0530 Subject: [SCSI] mpt2sas: Fix device removal handshake for zoned devices When zoning end devices, the driver is not sending device removal handshake alogrithm to firmware. This results in controller firmware not sending sas topology add events the next time the device is added. The fix is the driver should be doing the device removal handshake even though the PHYSTATUS_VACANT bit is set in the PhyStatus of the event data. The current design is avoiding the handshake when the VACANT bit is set in the phy status. Signed-off-by: Kashyap Desai Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/scsi/mpt2sas/mpt2sas_scsih.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'drivers/scsi/mpt2sas/mpt2sas_scsih.c') diff --git a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c index eda347c57979..95d82743d7b1 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c +++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c @@ -2981,9 +2981,6 @@ _scsih_check_topo_delete_events(struct MPT2SAS_ADAPTER *ioc, u16 handle; for (i = 0 ; i < event_data->NumEntries; i++) { - if (event_data->PHY[i].PhyStatus & - MPI2_EVENT_SAS_TOPO_PHYSTATUS_VACANT) - continue; handle = le16_to_cpu(event_data->PHY[i].AttachedDevHandle); if (!handle) continue; -- cgit v1.2.3 From efe82a16bc0f9f9e1fc8fa706eb0309fcd57770a Mon Sep 17 00:00:00 2001 From: "Kashyap, Desai" Date: Tue, 4 Jan 2011 11:34:17 +0530 Subject: [SCSI] mpt2sas: fix internal device reset for older firmware prior to MPI Rev K The "internal device reset complete" event is not supported for older firmware prior to MPI Rev K We added a check in the driver so the "internal device reset" event is ignored for older firmware. When ignored, the tm_busy flag doesn't get set nor cleared. Without this fix, IO queues would be froozen indefinetly after the "internal device reset" event, as the "complete" event never sent to clear the flag. Signed-off-by: Kashyap Desai Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/scsi/mpt2sas/mpt2sas_scsih.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'drivers/scsi/mpt2sas/mpt2sas_scsih.c') diff --git a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c index 95d82743d7b1..a16f2a05736f 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c +++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c @@ -5002,6 +5002,12 @@ _scsih_sas_device_status_change_event(struct MPT2SAS_ADAPTER *ioc, event_data); #endif + /* In MPI Revision K (0xC), the internal device reset complete was + * implemented, so avoid setting tm_busy flag for older firmware. + */ + if ((ioc->facts.HeaderVersion >> 8) < 0xC) + return; + if (event_data->ReasonCode != MPI2_EVENT_SAS_DEV_STAT_RC_INTERNAL_DEVICE_RESET && event_data->ReasonCode != -- cgit v1.2.3 From ec07a053597bdab51cbd23619f9f9f392712508a Mon Sep 17 00:00:00 2001 From: "Kashyap, Desai" Date: Wed, 5 Jan 2011 17:54:32 +0530 Subject: [SCSI] mpt2sas: Fix the race between broadcast asyn event and scsi command completion 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 Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/scsi/mpt2sas/mpt2sas_scsih.c | 54 ++++++++++++++++++++++++++++++------ 1 file changed, 45 insertions(+), 9 deletions(-) (limited to 'drivers/scsi/mpt2sas/mpt2sas_scsih.c') diff --git a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c index a16f2a05736f..db287d79bdcf 100644 --- 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 * @@ -831,6 +831,28 @@ _scsih_scsi_lookup_get(struct MPT2SAS_ADAPTER *ioc, u16 smid) return ioc->scsi_lookup[smid - 1].scmd; } +/** + * _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 @@ -3207,7 +3229,7 @@ _scsih_flush_running_cmds(struct MPT2SAS_ADAPTER *ioc) 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++; @@ -3801,7 +3823,7 @@ _scsih_io_done(struct MPT2SAS_ADAPTER *ioc, u16 smid, u8 msix_index, u32 reply) 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; @@ -5102,6 +5124,7 @@ _scsih_sas_broadcast_primative_event(struct MPT2SAS_ADAPTER *ioc, 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; @@ -5112,12 +5135,17 @@ _scsih_sas_broadcast_primative_event(struct MPT2SAS_ADAPTER *ioc, 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; @@ -5125,7 +5153,8 @@ _scsih_sas_broadcast_primative_event(struct MPT2SAS_ADAPTER *ioc, 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 */ @@ -5141,6 +5170,7 @@ _scsih_sas_broadcast_primative_event(struct MPT2SAS_ADAPTER *ioc, 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; @@ -5150,14 +5180,20 @@ _scsih_sas_broadcast_primative_event(struct MPT2SAS_ADAPTER *ioc, (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", -- cgit v1.2.3 From 3a9c913a3e57b170887d39456e04c18f2305ec67 Mon Sep 17 00:00:00 2001 From: "Kashyap, Desai" Date: Tue, 4 Jan 2011 11:40:23 +0530 Subject: [SCSI] mpt2sas: fix Integrated Raid unsynced on shutdown problem Issue: IR shutdown(sending) and IR shutdown(complete) messages not listed in /var/log/messages when driver is removed. The driver needs to issue a MPI2_RAID_ACTION_SYSTEM_SHUTDOWN_INITIATED request when the driver is unloaded so the IR metadata journal is updated. If this request is not sent, then the volume would need a "check consistency" issued on the next bootup if the volume was roamed from one initiator to another. The current driver supports this feature only when the system is rebooted, however this also need to be supported if the driver is unloaded Fix: To fix this issue, the driver is going to need to call the _scsih_ir_shutdown prior to reporting the volumes missing from the OS, hence the device handles are still present. Signed-off-by: Kashyap Desai Cc: stable@kernel.org Signed-off-by: James Bottomley --- drivers/scsi/mpt2sas/mpt2sas_scsih.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/scsi/mpt2sas/mpt2sas_scsih.c') diff --git a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c index db287d79bdcf..5ded3db6e316 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c +++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c @@ -6665,6 +6665,7 @@ _scsih_remove(struct pci_dev *pdev) destroy_workqueue(wq); /* release all the volumes */ + _scsih_ir_shutdown(ioc); list_for_each_entry_safe(raid_device, next, &ioc->raid_device_list, list) { if (raid_device->starget) { -- cgit v1.2.3