Commit b4730fb6 authored by Sreekanth Reddy's avatar Sreekanth Reddy Committed by James Bottomley

[SCSI] mpt2sas: fix for driver fails EEH, recovery from injected pci bus error

This patch stops the driver to invoke kthread (which remove the dead ioc)
for some time while EEH recovery has started.

[thenzl: add a 'non_operational_loop' reset.]
Signed-off-by: default avatarSreekanth Reddy <Sreekanth.Reddy@lsi.com>
Signed-off-by: default avatarTomas Henzl <thenzl@redhat.com>
Signed-off-by: default avatarJames Bottomley <JBottomley@Parallels.com>
parent 25ef16d0
...@@ -155,7 +155,7 @@ _base_fault_reset_work(struct work_struct *work) ...@@ -155,7 +155,7 @@ _base_fault_reset_work(struct work_struct *work)
struct task_struct *p; struct task_struct *p;
spin_lock_irqsave(&ioc->ioc_reset_in_progress_lock, flags); spin_lock_irqsave(&ioc->ioc_reset_in_progress_lock, flags);
if (ioc->shost_recovery) if (ioc->shost_recovery || ioc->pci_error_recovery)
goto rearm_timer; goto rearm_timer;
spin_unlock_irqrestore(&ioc->ioc_reset_in_progress_lock, flags); spin_unlock_irqrestore(&ioc->ioc_reset_in_progress_lock, flags);
...@@ -164,6 +164,20 @@ _base_fault_reset_work(struct work_struct *work) ...@@ -164,6 +164,20 @@ _base_fault_reset_work(struct work_struct *work)
printk(MPT2SAS_INFO_FMT "%s : SAS host is non-operational !!!!\n", printk(MPT2SAS_INFO_FMT "%s : SAS host is non-operational !!!!\n",
ioc->name, __func__); ioc->name, __func__);
/* It may be possible that EEH recovery can resolve some of
* pci bus failure issues rather removing the dead ioc function
* by considering controller is in a non-operational state. So
* here priority is given to the EEH recovery. If it doesn't
* not resolve this issue, mpt2sas driver will consider this
* controller to non-operational state and remove the dead ioc
* function.
*/
if (ioc->non_operational_loop++ < 5) {
spin_lock_irqsave(&ioc->ioc_reset_in_progress_lock,
flags);
goto rearm_timer;
}
/* /*
* Call _scsih_flush_pending_cmds callback so that we flush all * Call _scsih_flush_pending_cmds callback so that we flush all
* pending commands back to OS. This call is required to aovid * pending commands back to OS. This call is required to aovid
...@@ -193,6 +207,8 @@ _base_fault_reset_work(struct work_struct *work) ...@@ -193,6 +207,8 @@ _base_fault_reset_work(struct work_struct *work)
return; /* don't rearm timer */ return; /* don't rearm timer */
} }
ioc->non_operational_loop = 0;
if ((doorbell & MPI2_IOC_STATE_MASK) == MPI2_IOC_STATE_FAULT) { if ((doorbell & MPI2_IOC_STATE_MASK) == MPI2_IOC_STATE_FAULT) {
rc = mpt2sas_base_hard_reset_handler(ioc, CAN_SLEEP, rc = mpt2sas_base_hard_reset_handler(ioc, CAN_SLEEP,
FORCE_BIG_HAMMER); FORCE_BIG_HAMMER);
...@@ -4386,6 +4402,7 @@ mpt2sas_base_attach(struct MPT2SAS_ADAPTER *ioc) ...@@ -4386,6 +4402,7 @@ mpt2sas_base_attach(struct MPT2SAS_ADAPTER *ioc)
if (missing_delay[0] != -1 && missing_delay[1] != -1) if (missing_delay[0] != -1 && missing_delay[1] != -1)
_base_update_missing_delay(ioc, missing_delay[0], _base_update_missing_delay(ioc, missing_delay[0],
missing_delay[1]); missing_delay[1]);
ioc->non_operational_loop = 0;
return 0; return 0;
......
...@@ -835,6 +835,7 @@ struct MPT2SAS_ADAPTER { ...@@ -835,6 +835,7 @@ struct MPT2SAS_ADAPTER {
u16 cpu_msix_table_sz; u16 cpu_msix_table_sz;
u32 ioc_reset_count; u32 ioc_reset_count;
MPT2SAS_FLUSH_RUNNING_CMDS schedule_dead_ioc_flush_running_cmds; MPT2SAS_FLUSH_RUNNING_CMDS schedule_dead_ioc_flush_running_cmds;
u32 non_operational_loop;
/* internal commands, callback index */ /* internal commands, callback index */
u8 scsi_io_cb_idx; u8 scsi_io_cb_idx;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment