[PATCH 08/11] megaraid_sas: Do not initiate OCR if controller is not in ready state

[PATCH 08/11] megaraid_sas: Do not initiate OCR if controller is not in ready state

[Anand Lodnoor]

Driver can initiate OCR if DCMD command is timeout, but there is a
deadlock if the driver attempts to invoke another OCR  before the
mutex lock 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 or not.

Cc: stable@vger.kernel.org
Signed-off-by: Shivasharan S <shivasharan.srikanteshwara@broadcom.com>
Signed-off-by: Anand Lodnoor <anand.lodnoor@broadcom.com>
---
 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 8bc4076..da47c8b 100644
--- a/drivers/scsi/megaraid/megaraid_sas_base.c
+++ b/drivers/scsi/megaraid/megaraid_sas_base.c
@@ -4404,7 +4404,8 @@ 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 6860fd2..8b6cc1b 100644
--- a/drivers/scsi/megaraid/megaraid_sas_fusion.c
+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c
@@ -4851,6 +4851,7 @@ 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);
@@ -5059,7 +5060,7 @@ int megasas_reset_fusion(struct Scsi_Host *shost, int reason)
 	instance->skip_heartbeat_timer_del = 1;
 	retval = FAILED;
 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 8358b68..d57ecc7 100644
--- a/drivers/scsi/megaraid/megaraid_sas_fusion.h
+++ b/drivers/scsi/megaraid/megaraid_sas_fusion.h
@@ -89,6 +89,7 @@ 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 RAID_1_PEER_CMDS 2
 #define JBOD_MAPS_COUNT	2
 #define MEGASAS_REDUCE_QD_COUNT 64
-- 
1.8.3.1

Re: [PATCH 08/11] megaraid_sas: Do not initiate OCR if controller is not in ready state

[Sumit Saxena]

On Sat, Jan 4, 2020 at 11:01 AM Sasha Levin <sashal@kernel.org> wrote:
>
> Hi,
>
> [This is an automated email]
>
> This commit has been processed because it contains a -stable tag.
> The stable tag indicates that it's relevant for the following trees: all
>
> The bot has tested the following trees: v5.4.7, v5.3.18, v4.19.92, v4.14.161, v4.9.207, v4.4.207.
>
> v5.4.7: Build OK!
> v5.3.18: Build OK!
> v4.19.92: Failed to apply! Possible dependencies:
>     Unable to calculate
>
> v4.14.161: Failed to apply! Possible dependencies:
>     Unable to calculate
>
> v4.9.207: Failed to apply! Possible dependencies:
>     45f4f2eb3da3 ("scsi: megaraid_sas: Add new pci device Ids for SAS3.5 Generic Megaraid Controllers")
>     69c337c0f8d7 ("scsi: megaraid_sas: SAS3.5 Generic Megaraid Controllers Fast Path for RAID 1/10 Writes")
>     a73b0a4b5d17 ("scsi: megaraid_sas: Change RAID_1_10_RMW_CMDS to RAID_1_PEER_CMDS and set value to 2")
>     d0fc91d67c59 ("scsi: megaraid_sas: Send SYNCHRONIZE_CACHE for VD to firmware")
>     fdd84e2514b0 ("scsi: megaraid_sas: SAS3.5 Generic Megaraid Controllers Stream Detection and IO Coalescing")
>
> v4.4.207: Failed to apply! Possible dependencies:
>     179ac14291a0 ("megaraid_sas: Reply Descriptor Post Queue (RDPQ) support")
>     18365b138508 ("megaraid_sas: Task management support")
>     2c048351c8e3 ("megaraid_sas: Syncing request flags macro names with firmware")
>     308ec459bc19 ("megaraid_sas: Dual queue depth support")
>     69c337c0f8d7 ("scsi: megaraid_sas: SAS3.5 Generic Megaraid Controllers Fast Path for RAID 1/10 Writes")
>     6d40afbc7d13 ("megaraid_sas: MFI IO timeout handling")
>     8a01a41d8647 ("megaraid_sas: Make adprecovery variable atomic")
>     8f05024cd3db ("megaraid_sas: Fastpath region lock bypass")
>     a73b0a4b5d17 ("scsi: megaraid_sas: Change RAID_1_10_RMW_CMDS to RAID_1_PEER_CMDS and set value to 2")
>
>
> NOTE: The patch will not be queued to stable trees until it is upstream.
>
> How should we proceed with this patch?
Hi Sasha,

Please pick this patch for stable trees where it cleanly gets applied
after it is upstream.
We will backport this patch for stable trees where it does not get
applied cleanly.

Thanks,
Sumit
>
> --
> Thanks,
> Sasha