[PATCH 4.19] scsi: megaraid_sas: Do not initiate OCR if controller is not in ready state

[PATCH 4.19] scsi: megaraid_sas: Do not initiate OCR if controller is not in ready state

[Anand Lodnoor]

commit 6d7537270e3283b92f9b327da9d58a4de40fe8d0 upstream.

Driver initiates OCR if a DCMD command times out. But there is a
deadlock if the driver attempts to invoke another OCR before the
mutex lock (reset_mutex) is released from the previous session of OCR.

This patch takes care of the above scenario using new flag
MEGASAS_FUSION_OCR_NOT_POSSIBLE to indicate if OCR is possible.

Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/1579000882-20246-9-git-send-email-anand.lodnoor@broadcom.com
Signed-off-by: Shivasharan S <shivasharan.srikanteshwara@broadcom.com>
Signed-off-by: Anand Lodnoor <anand.lodnoor@broadcom.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/scsi/megaraid/megaraid_sas_base.c   | 3 ++-
 drivers/scsi/megaraid/megaraid_sas_fusion.c | 3 ++-
 drivers/scsi/megaraid/megaraid_sas_fusion.h | 1 +
 3 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c
index 2f31d26..a2da200 100644
--- a/drivers/scsi/megaraid/megaraid_sas_base.c
+++ b/drivers/scsi/megaraid/megaraid_sas_base.c
@@ -4177,7 +4177,8 @@ int megasas_alloc_cmds(struct megasas_instance *instance)
 	if (instance->adapter_type == MFI_SERIES)
 		return KILL_ADAPTER;
 	else if (instance->unload ||
-			test_bit(MEGASAS_FUSION_IN_RESET, &instance->reset_flags))
+			test_bit(MEGASAS_FUSION_OCR_NOT_POSSIBLE,
+				 &instance->reset_flags))
 		return IGNORE_TIMEOUT;
 	else
 		return INITIATE_OCR;
diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c b/drivers/scsi/megaraid/megaraid_sas_fusion.c
index f45c54f..b094a4e 100644
--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
@@ -4558,6 +4558,7 @@ int megasas_reset_fusion(struct Scsi_Host *shost, int reason)
 	if (instance->requestorId && !instance->skip_heartbeat_timer_del)
 		del_timer_sync(&instance->sriov_heartbeat_timer);
 	set_bit(MEGASAS_FUSION_IN_RESET, &instance->reset_flags);
+	set_bit(MEGASAS_FUSION_OCR_NOT_POSSIBLE, &instance->reset_flags);
 	atomic_set(&instance->adprecovery, MEGASAS_ADPRESET_SM_POLLING);
 	instance->instancet->disable_intr(instance);
 	megasas_sync_irqs((unsigned long)instance);
@@ -4747,7 +4748,7 @@ int megasas_reset_fusion(struct Scsi_Host *shost, int reason)
 		atomic_set(&instance->adprecovery, MEGASAS_HBA_OPERATIONAL);
 	}
 out:
-	clear_bit(MEGASAS_FUSION_IN_RESET, &instance->reset_flags);
+	clear_bit(MEGASAS_FUSION_OCR_NOT_POSSIBLE, &instance->reset_flags);
 	mutex_unlock(&instance->reset_mutex);
 	return retval;
 }
diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.h b/drivers/scsi/megaraid/megaraid_sas_fusion.h
index 8e5ebee..df7bbd0 100644
--- a/drivers/scsi/megaraid/megaraid_sas_fusion.h
+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.h
@@ -102,6 +102,7 @@ enum MR_RAID_FLAGS_IO_SUB_TYPE {
 
 #define MEGASAS_FP_CMD_LEN	16
 #define MEGASAS_FUSION_IN_RESET 0
+#define MEGASAS_FUSION_OCR_NOT_POSSIBLE 1
 #define THRESHOLD_REPLY_COUNT 50
 #define RAID_1_PEER_CMDS 2
 #define JBOD_MAPS_COUNT	2
-- 
1.8.3.1

Re: [PATCH 4.19] scsi: megaraid_sas: Do not initiate OCR if controller is not in ready state

[Greg Kroah-Hartman]

On Thu, Feb 13, 2020 at 04:12:58PM +0530, Anand Lodnoor wrote:
> commit 6d7537270e3283b92f9b327da9d58a4de40fe8d0 upstream.
> 
> Driver initiates OCR if a DCMD command times out. But there is a
> deadlock if the driver attempts to invoke another OCR before the
> mutex lock (reset_mutex) is released from the previous session of OCR.
> 
> This patch takes care of the above scenario using new flag
> MEGASAS_FUSION_OCR_NOT_POSSIBLE to indicate if OCR is possible.
> 
> Cc: stable@vger.kernel.org
> Link: https://lore.kernel.org/r/1579000882-20246-9-git-send-email-anand.lodnoor@broadcom.com
> Signed-off-by: Shivasharan S <shivasharan.srikanteshwara@broadcom.com>
> Signed-off-by: Anand Lodnoor <anand.lodnoor@broadcom.com>
> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

All 3 of these now applied, thanks!

greg k-h