[PATCH 00/15] qla2xxx bug fixes

[PATCH 00/15] qla2xxx bug fixes

[Nilesh Javali]

Martin,
Please apply the qla2xxx bug fixes to the scsi tree at your
earliest convenience.

Thanks,
Nilesh

Arun Easi (5):
  qla2xxx: Fix compilation issue in PPC systems
  qla2xxx: Fix crash during driver load on big endian machines
  qla2xxx: Fix FW initialization error on big endian machines
  qla2xxx: Fix flash update in 28XX adapters on big endian machines
  qla2xxx: Fix device loss on 4G and older HBAs.

Daniel Wagner (1):
  scsi: qla2xxx: Return EBUSY on fcport deletion

Nilesh Javali (1):
  qla2xxx: Update version to 10.02.00.104-k

Quinn Tran (3):
  qla2xxx: limit interrupt vectors to number of cpu
  qla2xxx: tear down session if FW say its down
  qla2xxx: fix N2N and NVME connect retry failure

Saurav Kashyap (5):
  qla2xxx: Change post del message from debug level to log level
  qla2xxx: Don't check for fw_started while posting nvme command
  qla2xxx: Handle aborts correctly for port undergoing deletion
  qla2xxx: Fix the call trace for flush workqueue
  qla2xxx: If fcport is undergoing deletion return IO with retry

 drivers/scsi/qla2xxx/qla_gs.c      |  8 ++--
 drivers/scsi/qla2xxx/qla_init.c    | 74 ++++++++++++++++++++++--------
 drivers/scsi/qla2xxx/qla_isr.c     | 34 ++++++++++++--
 drivers/scsi/qla2xxx/qla_mbx.c     |  9 ++--
 drivers/scsi/qla2xxx/qla_nvme.c    | 14 +++---
 drivers/scsi/qla2xxx/qla_nx.c      |  2 +-
 drivers/scsi/qla2xxx/qla_nx2.c     |  4 +-
 drivers/scsi/qla2xxx/qla_os.c      | 10 ++--
 drivers/scsi/qla2xxx/qla_sup.c     | 10 ++--
 drivers/scsi/qla2xxx/qla_tmpl.c    |  9 ++--
 drivers/scsi/qla2xxx/qla_tmpl.h    |  2 +-
 drivers/scsi/qla2xxx/qla_version.h |  4 +-
 12 files changed, 120 insertions(+), 60 deletions(-)


base-commit: cf4d4d8ebdb838ee996e09e3ee18deb9a7737dea
-- 
2.19.0.rc0

[PATCH 01/15] scsi: qla2xxx: Return EBUSY on fcport deletion

[Nilesh Javali]

From: Daniel Wagner <dwagner@suse.de>

When the fcport is about to be deleted we should return EBUSY instead of
ENODEV. Only for EBUSY will the request be requeued in a multipath setup.

Also return EBUSY when the firmware has not yet started to avoid dropping
the request.

Link: https://lore.kernel.org/r/20201014073048.36219-1-dwagner@suse.de
Reviewed-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
---
 drivers/scsi/qla2xxx/qla_nvme.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c
index 1f9005125313..b7a1dc24db38 100644
--- a/drivers/scsi/qla2xxx/qla_nvme.c
+++ b/drivers/scsi/qla2xxx/qla_nvme.c
@@ -554,10 +554,12 @@ static int qla_nvme_post_cmd(struct nvme_fc_local_port *lport,
 
 	fcport = qla_rport->fcport;
 
-	if (!qpair || !fcport || (qpair && !qpair->fw_started) ||
-	    (fcport && fcport->deleted))
+	if (!qpair || !fcport)
 		return -ENODEV;
 
+	if (!qpair->fw_started || fcport->deleted)
+		return -EBUSY;
+
 	vha = fcport->vha;
 
 	if (!(fcport->nvme_flag & NVME_FLAG_REGISTERED))
-- 
2.19.0.rc0

Re: [PATCH 01/15] scsi: qla2xxx: Return EBUSY on fcport deletion

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Daniel Wagner <dwagner@suse.de>
> 
> When the fcport is about to be deleted we should return EBUSY instead of
> ENODEV. Only for EBUSY will the request be requeued in a multipath setup.
> 
> Also return EBUSY when the firmware has not yet started to avoid dropping
> the request.
> 
> Link: https://lore.kernel.org/r/20201014073048.36219-1-dwagner@suse.de
> Reviewed-by: Arun Easi <aeasi@marvell.com>
> Signed-off-by: Daniel Wagner <dwagner@suse.de>
> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
> ---
> drivers/scsi/qla2xxx/qla_nvme.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c
> index 1f9005125313..b7a1dc24db38 100644
> --- a/drivers/scsi/qla2xxx/qla_nvme.c
> +++ b/drivers/scsi/qla2xxx/qla_nvme.c
> @@ -554,10 +554,12 @@ static int qla_nvme_post_cmd(struct nvme_fc_local_port *lport,
> 
> 	fcport = qla_rport->fcport;
> 
> -	if (!qpair || !fcport || (qpair && !qpair->fw_started) ||
> -	    (fcport && fcport->deleted))
> +	if (!qpair || !fcport)
> 		return -ENODEV;
> 
> +	if (!qpair->fw_started || fcport->deleted)
> +		return -EBUSY;
> +
> 	vha = fcport->vha;
> 
> 	if (!(fcport->nvme_flag & NVME_FLAG_REGISTERED))
> -- 
> 2.19.0.rc0
> 

Looks Good. 

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 02/15] qla2xxx: Change post del message from debug level to log level

[Nilesh Javali]

From: Saurav Kashyap <skashyap@marvell.com>

Change the message debug level.

Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_gs.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_gs.c b/drivers/scsi/qla2xxx/qla_gs.c
index e28c4b7ec55f..391ac75e3de3 100644
--- a/drivers/scsi/qla2xxx/qla_gs.c
+++ b/drivers/scsi/qla2xxx/qla_gs.c
@@ -3558,10 +3558,10 @@ void qla24xx_async_gnnft_done(scsi_qla_host_t *vha, srb_t *sp)
 					if (fcport->flags & FCF_FCP2_DEVICE)
 						fcport->logout_on_delete = 0;
 
-					ql_dbg(ql_dbg_disc, vha, 0x20f0,
-					    "%s %d %8phC post del sess\n",
-					    __func__, __LINE__,
-					    fcport->port_name);
+					ql_log(ql_log_warn, vha, 0x20f0,
+					       "%s %d %8phC post del sess\n",
+					       __func__, __LINE__,
+					       fcport->port_name);
 
 					qlt_schedule_sess_for_deletion(fcport);
 					continue;
-- 
2.19.0.rc0

Re: [PATCH 02/15] qla2xxx: Change post del message from debug level to log level

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Saurav Kashyap <skashyap@marvell.com>
> 
> Change the message debug level.
> 
> Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_gs.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_gs.c b/drivers/scsi/qla2xxx/qla_gs.c
> index e28c4b7ec55f..391ac75e3de3 100644
> --- a/drivers/scsi/qla2xxx/qla_gs.c
> +++ b/drivers/scsi/qla2xxx/qla_gs.c
> @@ -3558,10 +3558,10 @@ void qla24xx_async_gnnft_done(scsi_qla_host_t *vha, srb_t *sp)
> 					if (fcport->flags & FCF_FCP2_DEVICE)
> 						fcport->logout_on_delete = 0;
> 
> -					ql_dbg(ql_dbg_disc, vha, 0x20f0,
> -					    "%s %d %8phC post del sess\n",
> -					    __func__, __LINE__,
> -					    fcport->port_name);
> +					ql_log(ql_log_warn, vha, 0x20f0,
> +					       "%s %d %8phC post del sess\n",
> +					       __func__, __LINE__,
> +					       fcport->port_name);
> 
> 					qlt_schedule_sess_for_deletion(fcport);
> 					continue;
> -- 
> 2.19.0.rc0
> 

I am okay with the change just curious, Would it not flood message file for large number of sessions?

--
Himanshu Madhani	 Oracle Linux Engineering

RE: [PATCH 02/15] qla2xxx: Change post del message from debug level to log level

[Saurav Kashyap]

Hi Himasnhu,
Comments inline

> -----Original Message-----
> From: Himanshu Madhani <himanshu.madhani@oracle.com>
> Sent: Tuesday, December 1, 2020 9:15 PM
> To: Nilesh Javali <njavali@marvell.com>
> Cc: Martin K . Petersen <martin.petersen@oracle.com>; linux-
> scsi@vger.kernel.org; GR-QLogic-Storage-Upstream <GR-QLogic-Storage-
> Upstream@marvell.com>
> Subject: Re: [PATCH 02/15] qla2xxx: Change post del message from debug level
> to log level
> 
> 
> 
> > On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> >
> > From: Saurav Kashyap <skashyap@marvell.com>
> >
> > Change the message debug level.
> >
> > Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
> > Signed-off-by: Nilesh Javali <njavali@marvell.com>
> > ---
> > drivers/scsi/qla2xxx/qla_gs.c | 8 ++++----
> > 1 file changed, 4 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/scsi/qla2xxx/qla_gs.c b/drivers/scsi/qla2xxx/qla_gs.c
> > index e28c4b7ec55f..391ac75e3de3 100644
> > --- a/drivers/scsi/qla2xxx/qla_gs.c
> > +++ b/drivers/scsi/qla2xxx/qla_gs.c
> > @@ -3558,10 +3558,10 @@ void qla24xx_async_gnnft_done(scsi_qla_host_t
> *vha, srb_t *sp)
> > 					if (fcport->flags & FCF_FCP2_DEVICE)
> > 						fcport->logout_on_delete = 0;
> >
> > -					ql_dbg(ql_dbg_disc, vha, 0x20f0,
> > -					    "%s %d %8phC post del sess\n",
> > -					    __func__, __LINE__,
> > -					    fcport->port_name);
> > +					ql_log(ql_log_warn, vha, 0x20f0,
> > +					       "%s %d %8phC post del sess\n",
> > +					       __func__, __LINE__,
> > +					       fcport->port_name);
> >
> >
> 	qlt_schedule_sess_for_deletion(fcport);
> > 					continue;
> > --
> > 2.19.0.rc0
> >
> 
> I am okay with the change just curious, Would it not flood message file for
> large number of sessions?

This was added mainly for help in debugging, if debug is not enabled. Sometimes we get logs
where it's hard to tell what happened to particular session. Moreover session deletion is not
very common scenario.

Thanks,
~Saurav
> 
> --
> Himanshu Madhani	 Oracle Linux Engineering

Re: [PATCH 02/15] qla2xxx: Change post del message from debug level to log level

[Himanshu Madhani]

> On Dec 1, 2020, at 10:05 AM, Saurav Kashyap <skashyap@marvell.com> wrote:
> 
> Hi Himasnhu,
> Comments inline
> 
>> -----Original Message-----
>> From: Himanshu Madhani <himanshu.madhani@oracle.com>
>> Sent: Tuesday, December 1, 2020 9:15 PM
>> To: Nilesh Javali <njavali@marvell.com>
>> Cc: Martin K . Petersen <martin.petersen@oracle.com>; linux-
>> scsi@vger.kernel.org; GR-QLogic-Storage-Upstream <GR-QLogic-Storage-
>> Upstream@marvell.com>
>> Subject: Re: [PATCH 02/15] qla2xxx: Change post del message from debug level
>> to log level
>> 
>> 
>> 
>>> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
>>> 
>>> From: Saurav Kashyap <skashyap@marvell.com>
>>> 
>>> Change the message debug level.
>>> 
>>> Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
>>> Signed-off-by: Nilesh Javali <njavali@marvell.com>
>>> ---
>>> drivers/scsi/qla2xxx/qla_gs.c | 8 ++++----
>>> 1 file changed, 4 insertions(+), 4 deletions(-)
>>> 
>>> diff --git a/drivers/scsi/qla2xxx/qla_gs.c b/drivers/scsi/qla2xxx/qla_gs.c
>>> index e28c4b7ec55f..391ac75e3de3 100644
>>> --- a/drivers/scsi/qla2xxx/qla_gs.c
>>> +++ b/drivers/scsi/qla2xxx/qla_gs.c
>>> @@ -3558,10 +3558,10 @@ void qla24xx_async_gnnft_done(scsi_qla_host_t
>> *vha, srb_t *sp)
>>> 					if (fcport->flags & FCF_FCP2_DEVICE)
>>> 						fcport->logout_on_delete = 0;
>>> 
>>> -					ql_dbg(ql_dbg_disc, vha, 0x20f0,
>>> -					    "%s %d %8phC post del sess\n",
>>> -					    __func__, __LINE__,
>>> -					    fcport->port_name);
>>> +					ql_log(ql_log_warn, vha, 0x20f0,
>>> +					       "%s %d %8phC post del sess\n",
>>> +					       __func__, __LINE__,
>>> +					       fcport->port_name);
>>> 
>>> 
>> 	qlt_schedule_sess_for_deletion(fcport);
>>> 					continue;
>>> --
>>> 2.19.0.rc0
>>> 
>> 
>> I am okay with the change just curious, Would it not flood message file for
>> large number of sessions?
> 
> This was added mainly for help in debugging, if debug is not enabled. Sometimes we get logs
> where it's hard to tell what happened to particular session. Moreover session deletion is not
> very common scenario.
> 

In that case, I would also prefer to see message coming out from qlt_schedule_sess_for_deletion(), because that’s where you can track that session is scheduled for deletion. 

I would prefer this message also changed to ql_log_warn. 

        ql_dbg(ql_dbg_disc, sess->vha, 0xe001,
            "Scheduling sess %p for deletion %8phC\n",
            sess, sess->port_name);

> Thanks,
> ~Saurav
>> 
>> --
>> Himanshu Madhani	 Oracle Linux Engineering
> 

--
Himanshu Madhani	 Oracle Linux Engineering

RE: [EXT] Re: [PATCH 02/15] qla2xxx: Change post del message from debug level to log level

[Saurav Kashyap]

Hi Himanshu,

> -----Original Message-----
> From: Himanshu Madhani <himanshu.madhani@oracle.com>
> Sent: Tuesday, December 1, 2020 9:46 PM
> To: Saurav Kashyap <skashyap@marvell.com>
> Cc: Nilesh Javali <njavali@marvell.com>; Martin K . Petersen
> <martin.petersen@oracle.com>; linux-scsi@vger.kernel.org; GR-QLogic-
> Storage-Upstream <GR-QLogic-Storage-Upstream@marvell.com>
> Subject: [EXT] Re: [PATCH 02/15] qla2xxx: Change post del message from
> debug level to log level
> 
> External Email
> 
> ----------------------------------------------------------------------
> 
> 
> > On Dec 1, 2020, at 10:05 AM, Saurav Kashyap <skashyap@marvell.com>
> wrote:
> >
> > Hi Himasnhu,
> > Comments inline
> >
> >> -----Original Message-----
> >> From: Himanshu Madhani <himanshu.madhani@oracle.com>
> >> Sent: Tuesday, December 1, 2020 9:15 PM
> >> To: Nilesh Javali <njavali@marvell.com>
> >> Cc: Martin K . Petersen <martin.petersen@oracle.com>; linux-
> >> scsi@vger.kernel.org; GR-QLogic-Storage-Upstream <GR-QLogic-Storage-
> >> Upstream@marvell.com>
> >> Subject: Re: [PATCH 02/15] qla2xxx: Change post del message from debug
> level
> >> to log level
> >>
> >>
> >>
> >>> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> >>>
> >>> From: Saurav Kashyap <skashyap@marvell.com>
> >>>
> >>> Change the message debug level.
> >>>
> >>> Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
> >>> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> >>> ---
> >>> drivers/scsi/qla2xxx/qla_gs.c | 8 ++++----
> >>> 1 file changed, 4 insertions(+), 4 deletions(-)
> >>>
> >>> diff --git a/drivers/scsi/qla2xxx/qla_gs.c b/drivers/scsi/qla2xxx/qla_gs.c
> >>> index e28c4b7ec55f..391ac75e3de3 100644
> >>> --- a/drivers/scsi/qla2xxx/qla_gs.c
> >>> +++ b/drivers/scsi/qla2xxx/qla_gs.c
> >>> @@ -3558,10 +3558,10 @@ void
> qla24xx_async_gnnft_done(scsi_qla_host_t
> >> *vha, srb_t *sp)
> >>> 					if (fcport->flags & FCF_FCP2_DEVICE)
> >>> 						fcport->logout_on_delete = 0;
> >>>
> >>> -					ql_dbg(ql_dbg_disc, vha, 0x20f0,
> >>> -					    "%s %d %8phC post del sess\n",
> >>> -					    __func__, __LINE__,
> >>> -					    fcport->port_name);
> >>> +					ql_log(ql_log_warn, vha, 0x20f0,
> >>> +					       "%s %d %8phC post del sess\n",
> >>> +					       __func__, __LINE__,
> >>> +					       fcport->port_name);
> >>>
> >>>
> >> 	qlt_schedule_sess_for_deletion(fcport);
> >>> 					continue;
> >>> --
> >>> 2.19.0.rc0
> >>>
> >>
> >> I am okay with the change just curious, Would it not flood message file for
> >> large number of sessions?
> >
> > This was added mainly for help in debugging, if debug is not enabled.
> Sometimes we get logs
> > where it's hard to tell what happened to particular session. Moreover session
> deletion is not
> > very common scenario.
> >
> 
> In that case, I would also prefer to see message coming out from
> qlt_schedule_sess_for_deletion(), because that’s where you can track that
> session is scheduled for deletion.
> 
> I would prefer this message also changed to ql_log_warn.
> 
>         ql_dbg(ql_dbg_disc, sess->vha, 0xe001,
>             "Scheduling sess %p for deletion %8phC\n",
>             sess, sess->port_name);

Make sense, will update it in next version.

Thanks,
~Saurav
> 
> > Thanks,
> > ~Saurav
> >>
> >> --
> >> Himanshu Madhani	 Oracle Linux Engineering
> >
> 
> --
> Himanshu Madhani	 Oracle Linux Engineering

[PATCH 03/15] qla2xxx: limit interrupt vectors to number of cpu

[Nilesh Javali]

From: Quinn Tran <qutran@marvell.com>

Driver created too many QPairs(126) with 28xx adapter.
Limit the number of CPUs to lower wasted resources.

Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_isr.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
index a24b82de4aab..77dd7630c3f8 100644
--- a/drivers/scsi/qla2xxx/qla_isr.c
+++ b/drivers/scsi/qla2xxx/qla_isr.c
@@ -3952,10 +3952,12 @@ qla24xx_enable_msix(struct qla_hw_data *ha, struct rsp_que *rsp)
 	if (USER_CTRL_IRQ(ha) || !ha->mqiobase) {
 		/* user wants to control IRQ setting for target mode */
 		ret = pci_alloc_irq_vectors(ha->pdev, min_vecs,
-		    ha->msix_count, PCI_IRQ_MSIX);
+		    min((u16)ha->msix_count, (u16)num_online_cpus()),
+		    PCI_IRQ_MSIX);
 	} else
 		ret = pci_alloc_irq_vectors_affinity(ha->pdev, min_vecs,
-		    ha->msix_count, PCI_IRQ_MSIX | PCI_IRQ_AFFINITY,
+		    min((u16)ha->msix_count, (u16)num_online_cpus()),
+		    PCI_IRQ_MSIX | PCI_IRQ_AFFINITY,
 		    &desc);
 
 	if (ret < 0) {
-- 
2.19.0.rc0

Re: [PATCH 03/15] qla2xxx: limit interrupt vectors to number of cpu

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Quinn Tran <qutran@marvell.com>
> 
> Driver created too many QPairs(126) with 28xx adapter.
> Limit the number of CPUs to lower wasted resources.
> 
> Signed-off-by: Quinn Tran <qutran@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_isr.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
> index a24b82de4aab..77dd7630c3f8 100644
> --- a/drivers/scsi/qla2xxx/qla_isr.c
> +++ b/drivers/scsi/qla2xxx/qla_isr.c
> @@ -3952,10 +3952,12 @@ qla24xx_enable_msix(struct qla_hw_data *ha, struct rsp_que *rsp)
> 	if (USER_CTRL_IRQ(ha) || !ha->mqiobase) {
> 		/* user wants to control IRQ setting for target mode */
> 		ret = pci_alloc_irq_vectors(ha->pdev, min_vecs,
> -		    ha->msix_count, PCI_IRQ_MSIX);
> +		    min((u16)ha->msix_count, (u16)num_online_cpus()),
> +		    PCI_IRQ_MSIX);
> 	} else
> 		ret = pci_alloc_irq_vectors_affinity(ha->pdev, min_vecs,
> -		    ha->msix_count, PCI_IRQ_MSIX | PCI_IRQ_AFFINITY,
> +		    min((u16)ha->msix_count, (u16)num_online_cpus()),
> +		    PCI_IRQ_MSIX | PCI_IRQ_AFFINITY,
> 		    &desc);
> 
> 	if (ret < 0) {
> -- 
> 2.19.0.rc0
> 

Looks good.

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 04/15] qla2xxx: tear down session if FW say its down

[Nilesh Javali]

From: Quinn Tran <qutran@marvell.com>

The completion status 0x28 (ppc = be = 0x2800) below indicate session
is not there, trigger session deletion.

qla2xxx [000b:04:00.1]-8009:8: DEVICE RESET ISSUED nexus=8:1:51 cmd=c000001432d0f600.
qla2xxx [000b:04:00.1]-5039:8: Async-tmf error - hdl=67b completion status(2800).
qla2xxx [000b:04:00.1]-8030:8: TM IOCB failed (102).
qla2xxx [000b:04:00.1]-800c:8: do_reset failed for cmd=c000001432d0f600.
qla2xxx [000b:04:00.1]-800f:8: DEVICE RESET FAILED: Task management failed nexus=8:1:51 cmd=c000001432d0f600.
qla2xxx [000b:04:00.1]-8009:8: DEVICE RESET ISSUED nexus=8:1:52 cmd=c000001432d0c200.
qla2xxx [000b:04:00.1]-5039:8: Async-tmf error - hdl=67c completion status(2800).
qla2xxx [000b:04:00.1]-8030:8: TM IOCB failed (102).

Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_isr.c | 28 +++++++++++++++++++++++++++-
 1 file changed, 27 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
index 77dd7630c3f8..f9142dbec112 100644
--- a/drivers/scsi/qla2xxx/qla_isr.c
+++ b/drivers/scsi/qla2xxx/qla_isr.c
@@ -2226,11 +2226,13 @@ qla24xx_tm_iocb_entry(scsi_qla_host_t *vha, struct req_que *req, void *tsk)
 	srb_t *sp;
 	struct srb_iocb *iocb;
 	struct sts_entry_24xx *sts = (struct sts_entry_24xx *)tsk;
+	u16 comp_status;
 
 	sp = qla2x00_get_sp_from_handle(vha, func, req, tsk);
 	if (!sp)
 		return;
 
+	comp_status = le16_to_cpu(sts->comp_status);
 	iocb = &sp->u.iocb_cmd;
 	type = sp->name;
 	fcport = sp->fcport;
@@ -2244,7 +2246,7 @@ qla24xx_tm_iocb_entry(scsi_qla_host_t *vha, struct req_que *req, void *tsk)
 	} else if (sts->comp_status != cpu_to_le16(CS_COMPLETE)) {
 		ql_log(ql_log_warn, fcport->vha, 0x5039,
 		    "Async-%s error - hdl=%x completion status(%x).\n",
-		    type, sp->handle, sts->comp_status);
+		    type, sp->handle, comp_status);
 		iocb->u.tmf.data = QLA_FUNCTION_FAILED;
 	} else if ((le16_to_cpu(sts->scsi_status) &
 	    SS_RESPONSE_INFO_LEN_VALID)) {
@@ -2260,6 +2262,30 @@ qla24xx_tm_iocb_entry(scsi_qla_host_t *vha, struct req_que *req, void *tsk)
 		}
 	}
 
+	switch (comp_status) {
+	case CS_PORT_LOGGED_OUT:
+	case CS_PORT_CONFIG_CHG:
+	case CS_PORT_BUSY:
+	case CS_INCOMPLETE:
+	case CS_PORT_UNAVAILABLE:
+	case CS_TIMEOUT:
+	case CS_RESET:
+		if (atomic_read(&fcport->state) == FCS_ONLINE) {
+			ql_dbg(ql_dbg_disc, fcport->vha, 0x3021,
+			       "-Port to be marked lost on fcport=%02x%02x%02x, current port state= %s comp_status %x.\n",
+			       fcport->d_id.b.domain, fcport->d_id.b.area,
+			       fcport->d_id.b.al_pa,
+			       port_state_str[FCS_ONLINE],
+			       comp_status);
+
+			qlt_schedule_sess_for_deletion(fcport);
+		}
+		break;
+
+	default:
+		break;
+	}
+
 	if (iocb->u.tmf.data != QLA_SUCCESS)
 		ql_dump_buffer(ql_dbg_async + ql_dbg_buffer, sp->vha, 0x5055,
 		    sts, sizeof(*sts));
-- 
2.19.0.rc0

Re: [PATCH 04/15] qla2xxx: tear down session if FW say its down

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Quinn Tran <qutran@marvell.com>
> 
> The completion status 0x28 (ppc = be = 0x2800) below indicate session
> is not there, trigger session deletion.
> 
> qla2xxx [000b:04:00.1]-8009:8: DEVICE RESET ISSUED nexus=8:1:51 cmd=c000001432d0f600.
> qla2xxx [000b:04:00.1]-5039:8: Async-tmf error - hdl=67b completion status(2800).
> qla2xxx [000b:04:00.1]-8030:8: TM IOCB failed (102).
> qla2xxx [000b:04:00.1]-800c:8: do_reset failed for cmd=c000001432d0f600.
> qla2xxx [000b:04:00.1]-800f:8: DEVICE RESET FAILED: Task management failed nexus=8:1:51 cmd=c000001432d0f600.
> qla2xxx [000b:04:00.1]-8009:8: DEVICE RESET ISSUED nexus=8:1:52 cmd=c000001432d0c200.
> qla2xxx [000b:04:00.1]-5039:8: Async-tmf error - hdl=67c completion status(2800).
> qla2xxx [000b:04:00.1]-8030:8: TM IOCB failed (102).
> 
> Signed-off-by: Quinn Tran <qutran@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_isr.c | 28 +++++++++++++++++++++++++++-
> 1 file changed, 27 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
> index 77dd7630c3f8..f9142dbec112 100644
> --- a/drivers/scsi/qla2xxx/qla_isr.c
> +++ b/drivers/scsi/qla2xxx/qla_isr.c
> @@ -2226,11 +2226,13 @@ qla24xx_tm_iocb_entry(scsi_qla_host_t *vha, struct req_que *req, void *tsk)
> 	srb_t *sp;
> 	struct srb_iocb *iocb;
> 	struct sts_entry_24xx *sts = (struct sts_entry_24xx *)tsk;
> +	u16 comp_status;
> 
> 	sp = qla2x00_get_sp_from_handle(vha, func, req, tsk);
> 	if (!sp)
> 		return;
> 
> +	comp_status = le16_to_cpu(sts->comp_status);
> 	iocb = &sp->u.iocb_cmd;
> 	type = sp->name;
> 	fcport = sp->fcport;
> @@ -2244,7 +2246,7 @@ qla24xx_tm_iocb_entry(scsi_qla_host_t *vha, struct req_que *req, void *tsk)
> 	} else if (sts->comp_status != cpu_to_le16(CS_COMPLETE)) {
> 		ql_log(ql_log_warn, fcport->vha, 0x5039,
> 		    "Async-%s error - hdl=%x completion status(%x).\n",
> -		    type, sp->handle, sts->comp_status);
> +		    type, sp->handle, comp_status);
> 		iocb->u.tmf.data = QLA_FUNCTION_FAILED;
> 	} else if ((le16_to_cpu(sts->scsi_status) &
> 	    SS_RESPONSE_INFO_LEN_VALID)) {
> @@ -2260,6 +2262,30 @@ qla24xx_tm_iocb_entry(scsi_qla_host_t *vha, struct req_que *req, void *tsk)
> 		}
> 	}
> 
> +	switch (comp_status) {
> +	case CS_PORT_LOGGED_OUT:
> +	case CS_PORT_CONFIG_CHG:
> +	case CS_PORT_BUSY:
> +	case CS_INCOMPLETE:
> +	case CS_PORT_UNAVAILABLE:
> +	case CS_TIMEOUT:
> +	case CS_RESET:
> +		if (atomic_read(&fcport->state) == FCS_ONLINE) {
> +			ql_dbg(ql_dbg_disc, fcport->vha, 0x3021,
> +			       "-Port to be marked lost on fcport=%02x%02x%02x, current port state= %s comp_status %x.\n",
> +			       fcport->d_id.b.domain, fcport->d_id.b.area,
> +			       fcport->d_id.b.al_pa,
> +			       port_state_str[FCS_ONLINE],
> +			       comp_status);
> +
> +			qlt_schedule_sess_for_deletion(fcport);
> +		}
> +		break;
> +
> +	default:
> +		break;
> +	}
> +
> 	if (iocb->u.tmf.data != QLA_SUCCESS)
> 		ql_dump_buffer(ql_dbg_async + ql_dbg_buffer, sp->vha, 0x5055,
> 		    sts, sizeof(*sts));
> -- 
> 2.19.0.rc0
> 

Looks Good.

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 05/15] qla2xxx: Don't check for fw_started while posting nvme command

[Nilesh Javali]

From: Saurav Kashyap <skashyap@marvell.com>

NVMe commands can come only after successful addition of rport and nvme
connect, and rport is only registered after FW started bit is set. Remove the
redundant check.

Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_nvme.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c
index b7a1dc24db38..d4159d5a4ffd 100644
--- a/drivers/scsi/qla2xxx/qla_nvme.c
+++ b/drivers/scsi/qla2xxx/qla_nvme.c
@@ -554,19 +554,15 @@ static int qla_nvme_post_cmd(struct nvme_fc_local_port *lport,
 
 	fcport = qla_rport->fcport;
 
-	if (!qpair || !fcport)
-		return -ENODEV;
-
-	if (!qpair->fw_started || fcport->deleted)
+	if (unlikely(!qpair || !fcport || fcport->deleted))
 		return -EBUSY;
 
-	vha = fcport->vha;
-
 	if (!(fcport->nvme_flag & NVME_FLAG_REGISTERED))
 		return -ENODEV;
 
-	if (test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags) ||
-	    (qpair && !qpair->fw_started) || fcport->deleted)
+	vha = fcport->vha;
+
+	if (test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags))
 		return -EBUSY;
 
 	/*
-- 
2.19.0.rc0

Re: [PATCH 05/15] qla2xxx: Don't check for fw_started while posting nvme command

[Daniel Wagner]

On Tue, Dec 01, 2020 at 12:27:20AM -0800, Nilesh Javali wrote:
> From: Saurav Kashyap <skashyap@marvell.com>
> 
> NVMe commands can come only after successful addition of rport and nvme
> connect, and rport is only registered after FW started bit is set. Remove the
> redundant check.
> 
> Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
>  drivers/scsi/qla2xxx/qla_nvme.c | 12 ++++--------
>  1 file changed, 4 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c
> index b7a1dc24db38..d4159d5a4ffd 100644
> --- a/drivers/scsi/qla2xxx/qla_nvme.c
> +++ b/drivers/scsi/qla2xxx/qla_nvme.c
> @@ -554,19 +554,15 @@ static int qla_nvme_post_cmd(struct nvme_fc_local_port *lport,
>  
>  	fcport = qla_rport->fcport;
>  
> -	if (!qpair || !fcport)
> -		return -ENODEV;
> -
> -	if (!qpair->fw_started || fcport->deleted)
> +	if (unlikely(!qpair || !fcport || fcport->deleted))
>  		return -EBUSY;

This reverts the fix from patch #1 in this series. What's the reasoning
that needs to return EBUSY when !qpair || !fcport is true?

RE: [PATCH 05/15] qla2xxx: Don't check for fw_started while posting nvme command

[Saurav Kashyap]

Hi Daniel,
Comments inline..

> -----Original Message-----
> From: Daniel Wagner <dwagner@suse.de>
> Sent: Tuesday, December 1, 2020 2:32 PM
> To: Nilesh Javali <njavali@marvell.com>
> Cc: martin.petersen@oracle.com; linux-scsi@vger.kernel.org; GR-QLogic-
> Storage-Upstream <GR-QLogic-Storage-Upstream@marvell.com>
> Subject: Re: [PATCH 05/15] qla2xxx: Don't check for fw_started while posting
> nvme command
> 
> On Tue, Dec 01, 2020 at 12:27:20AM -0800, Nilesh Javali wrote:
> > From: Saurav Kashyap <skashyap@marvell.com>
> >
> > NVMe commands can come only after successful addition of rport and nvme
> > connect, and rport is only registered after FW started bit is set. Remove the
> > redundant check.
> >
> > Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
> > Signed-off-by: Nilesh Javali <njavali@marvell.com>
> > ---
> >  drivers/scsi/qla2xxx/qla_nvme.c | 12 ++++--------
> >  1 file changed, 4 insertions(+), 8 deletions(-)
> >
> > diff --git a/drivers/scsi/qla2xxx/qla_nvme.c
> b/drivers/scsi/qla2xxx/qla_nvme.c
> > index b7a1dc24db38..d4159d5a4ffd 100644
> > --- a/drivers/scsi/qla2xxx/qla_nvme.c
> > +++ b/drivers/scsi/qla2xxx/qla_nvme.c
> > @@ -554,19 +554,15 @@ static int qla_nvme_post_cmd(struct
> nvme_fc_local_port *lport,
> >
> >  	fcport = qla_rport->fcport;
> >
> > -	if (!qpair || !fcport)
> > -		return -ENODEV;
> > -
> > -	if (!qpair->fw_started || fcport->deleted)
> > +	if (unlikely(!qpair || !fcport || fcport->deleted))
> >  		return -EBUSY;
> 
> This reverts the fix from patch #1 in this series. What's the reasoning
> that needs to return EBUSY when !qpair || !fcport is true?
<SK> Ideally driver should not hit (!qpair || !fcport) case.  The patch was to remove fw_started flag and consolidate other checks.
We want IO to retry until remote port is deleted and below condition is hit.
----------<condition>--------
        if (!(fcport->nvme_flag & NVME_FLAG_REGISTERED))
                return rval;
----------<condition>---------

Thanks,
~Saurav

Re: [PATCH 05/15] qla2xxx: Don't check for fw_started while posting nvme command

[Daniel Wagner]

Hi Saurav

On Tue, Dec 01, 2020 at 09:39:05AM +0000, Saurav Kashyap wrote:
> > > --- a/drivers/scsi/qla2xxx/qla_nvme.c
> > > +++ b/drivers/scsi/qla2xxx/qla_nvme.c
> > > @@ -554,19 +554,15 @@ static int qla_nvme_post_cmd(struct
> > nvme_fc_local_port *lport,
> > >
> > >  	fcport = qla_rport->fcport;
> > >
> > > -	if (!qpair || !fcport)
> > > -		return -ENODEV;
> > > -
> > > -	if (!qpair->fw_started || fcport->deleted)
> > > +	if (unlikely(!qpair || !fcport || fcport->deleted))
> > >  		return -EBUSY;
> > 
> > This reverts the fix from patch #1 in this series. What's the reasoning
> > that needs to return EBUSY when !qpair || !fcport is true?
>
> Ideally driver should not hit (!qpair || !fcport) case.  The patch was
> to remove fw_started flag and consolidate other checks.

Looking again on the patch I think I got confused.

> We want IO to retry until remote port is deleted and below condition is hit.

The result of this patch is that in EBUSY will be returned in all the
cases, not just for the case of fcport->deleted. So all is good from my
point of view. Thanks for explaining.

Sorry for the noise.

Thanks,
Daniel

RE: [EXT] Re: [PATCH 05/15] qla2xxx: Don't check for fw_started while posting nvme command

[Saurav Kashyap]

Hi Daniel,

> -----Original Message-----
> From: Daniel Wagner <dwagner@suse.de>
> Sent: Tuesday, December 1, 2020 3:24 PM
> To: Saurav Kashyap <skashyap@marvell.com>
> Cc: Nilesh Javali <njavali@marvell.com>; martin.petersen@oracle.com; linux-
> scsi@vger.kernel.org; GR-QLogic-Storage-Upstream <GR-QLogic-Storage-
> Upstream@marvell.com>
> Subject: [EXT] Re: [PATCH 05/15] qla2xxx: Don't check for fw_started while
> posting nvme command
> 
> External Email
> 
> ----------------------------------------------------------------------
> Hi Saurav
> 
> On Tue, Dec 01, 2020 at 09:39:05AM +0000, Saurav Kashyap wrote:
> > > > --- a/drivers/scsi/qla2xxx/qla_nvme.c
> > > > +++ b/drivers/scsi/qla2xxx/qla_nvme.c
> > > > @@ -554,19 +554,15 @@ static int qla_nvme_post_cmd(struct
> > > nvme_fc_local_port *lport,
> > > >
> > > >  	fcport = qla_rport->fcport;
> > > >
> > > > -	if (!qpair || !fcport)
> > > > -		return -ENODEV;
> > > > -
> > > > -	if (!qpair->fw_started || fcport->deleted)
> > > > +	if (unlikely(!qpair || !fcport || fcport->deleted))
> > > >  		return -EBUSY;
> > >
> > > This reverts the fix from patch #1 in this series. What's the reasoning
> > > that needs to return EBUSY when !qpair || !fcport is true?
> >
> > Ideally driver should not hit (!qpair || !fcport) case.  The patch was
> > to remove fw_started flag and consolidate other checks.
> 
> Looking again on the patch I think I got confused.
> 
> > We want IO to retry until remote port is deleted and below condition is hit.
> 
> The result of this patch is that in EBUSY will be returned in all the
> cases, not just for the case of fcport->deleted. So all is good from my
> point of view. Thanks for explaining.
> 
> Sorry for the noise.

No problem, most welcome.

Thanks,
~Saurav
> 
> Thanks,
> Daniel

Re: [PATCH 05/15] qla2xxx: Don't check for fw_started while posting nvme command

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Saurav Kashyap <skashyap@marvell.com>
> 
> NVMe commands can come only after successful addition of rport and nvme
> connect, and rport is only registered after FW started bit is set. Remove the
> redundant check.
> 
> Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_nvme.c | 12 ++++--------
> 1 file changed, 4 insertions(+), 8 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c
> index b7a1dc24db38..d4159d5a4ffd 100644
> --- a/drivers/scsi/qla2xxx/qla_nvme.c
> +++ b/drivers/scsi/qla2xxx/qla_nvme.c
> @@ -554,19 +554,15 @@ static int qla_nvme_post_cmd(struct nvme_fc_local_port *lport,
> 
> 	fcport = qla_rport->fcport;
> 
> -	if (!qpair || !fcport)
> -		return -ENODEV;
> -
> -	if (!qpair->fw_started || fcport->deleted)
> +	if (unlikely(!qpair || !fcport || fcport->deleted))
> 		return -EBUSY;
> 
> -	vha = fcport->vha;
> -
> 	if (!(fcport->nvme_flag & NVME_FLAG_REGISTERED))
> 		return -ENODEV;
> 
> -	if (test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags) ||
> -	    (qpair && !qpair->fw_started) || fcport->deleted)
> +	vha = fcport->vha;
> +
> +	if (test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags))
> 		return -EBUSY;
> 
> 	/*
> -- 
> 2.19.0.rc0
> 

Looks Good

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 06/15] qla2xxx: Fix compilation issue in PPC systems

[Nilesh Javali]

From: Arun Easi <aeasi@marvell.com>

Fix compile time errors reported on PPC systems.

Signed-off-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_nx.c  | 2 +-
 drivers/scsi/qla2xxx/qla_nx2.c | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c
index b3ba0de5d4fb..fd994e36200a 100644
--- a/drivers/scsi/qla2xxx/qla_nx.c
+++ b/drivers/scsi/qla2xxx/qla_nx.c
@@ -965,7 +965,7 @@ qla82xx_read_status_reg(struct qla_hw_data *ha, uint32_t *val)
 static int
 qla82xx_flash_wait_write_finish(struct qla_hw_data *ha)
 {
-	uint32_t val;
+	uint32_t val = 0;
 	int i, ret;
 	scsi_qla_host_t *vha = pci_get_drvdata(ha->pdev);
 
diff --git a/drivers/scsi/qla2xxx/qla_nx2.c b/drivers/scsi/qla2xxx/qla_nx2.c
index 01ccd4526707..68a16c95dcb7 100644
--- a/drivers/scsi/qla2xxx/qla_nx2.c
+++ b/drivers/scsi/qla2xxx/qla_nx2.c
@@ -139,7 +139,7 @@ qla8044_poll_wait_for_ready(struct scsi_qla_host *vha, uint32_t addr1,
 	uint32_t mask)
 {
 	unsigned long timeout;
-	uint32_t temp;
+	uint32_t temp = 0;
 
 	/* jiffies after 100ms */
 	timeout = jiffies + msecs_to_jiffies(TIMEOUT_100_MS);
@@ -2594,7 +2594,7 @@ qla8044_minidump_process_rdmux(struct scsi_qla_host *vha,
 	struct qla8044_minidump_entry_hdr *entry_hdr,
 	uint32_t **d_ptr)
 {
-	uint32_t r_addr, s_stride, s_addr, s_value, loop_cnt, i, r_value;
+	uint32_t r_addr, s_stride, s_addr, s_value, loop_cnt, i, r_value = 0;
 	struct qla8044_minidump_entry_mux *mux_hdr;
 	uint32_t *data_ptr = *d_ptr;
 
-- 
2.19.0.rc0

Re: [PATCH 06/15] qla2xxx: Fix compilation issue in PPC systems

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Arun Easi <aeasi@marvell.com>
> 
> Fix compile time errors reported on PPC systems.
> 

What is the error? Can you please add details in commit message for reference 

> Signed-off-by: Arun Easi <aeasi@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_nx.c  | 2 +-
> drivers/scsi/qla2xxx/qla_nx2.c | 4 ++--
> 2 files changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c
> index b3ba0de5d4fb..fd994e36200a 100644
> --- a/drivers/scsi/qla2xxx/qla_nx.c
> +++ b/drivers/scsi/qla2xxx/qla_nx.c
> @@ -965,7 +965,7 @@ qla82xx_read_status_reg(struct qla_hw_data *ha, uint32_t *val)
> static int
> qla82xx_flash_wait_write_finish(struct qla_hw_data *ha)
> {
> -	uint32_t val;
> +	uint32_t val = 0;
> 	int i, ret;
> 	scsi_qla_host_t *vha = pci_get_drvdata(ha->pdev);
> 
> diff --git a/drivers/scsi/qla2xxx/qla_nx2.c b/drivers/scsi/qla2xxx/qla_nx2.c
> index 01ccd4526707..68a16c95dcb7 100644
> --- a/drivers/scsi/qla2xxx/qla_nx2.c
> +++ b/drivers/scsi/qla2xxx/qla_nx2.c
> @@ -139,7 +139,7 @@ qla8044_poll_wait_for_ready(struct scsi_qla_host *vha, uint32_t addr1,
> 	uint32_t mask)
> {
> 	unsigned long timeout;
> -	uint32_t temp;
> +	uint32_t temp = 0;
> 
> 	/* jiffies after 100ms */
> 	timeout = jiffies + msecs_to_jiffies(TIMEOUT_100_MS);
> @@ -2594,7 +2594,7 @@ qla8044_minidump_process_rdmux(struct scsi_qla_host *vha,
> 	struct qla8044_minidump_entry_hdr *entry_hdr,
> 	uint32_t **d_ptr)
> {
> -	uint32_t r_addr, s_stride, s_addr, s_value, loop_cnt, i, r_value;
> +	uint32_t r_addr, s_stride, s_addr, s_value, loop_cnt, i, r_value = 0;
> 	struct qla8044_minidump_entry_mux *mux_hdr;
> 	uint32_t *data_ptr = *d_ptr;
> 
> -- 
> 2.19.0.rc0
> 

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 07/15] qla2xxx: Fix crash during driver load on big endian machines

[Nilesh Javali]

From: Arun Easi <aeasi@marvell.com>

Crash stack:
	[576544.715489] Unable to handle kernel paging request for data at address 0xd00000000f970000
	[576544.715497] Faulting instruction address: 0xd00000000f880f64
	[576544.715503] Oops: Kernel access of bad area, sig: 11 [#1]
	[576544.715506] SMP NR_CPUS=2048 NUMA pSeries
	:
	[576544.715703] NIP [d00000000f880f64] .qla27xx_fwdt_template_valid+0x94/0x100 [qla2xxx]
	[576544.715722] LR [d00000000f7952dc] .qla24xx_load_risc_flash+0x2fc/0x590 [qla2xxx]
	[576544.715726] Call Trace:
	[576544.715731] [c0000004d0ffb000] [c0000006fe02c350] 0xc0000006fe02c350 (unreliable)
	[576544.715750] [c0000004d0ffb080] [d00000000f7952dc] .qla24xx_load_risc_flash+0x2fc/0x590 [qla2xxx]
	[576544.715770] [c0000004d0ffb170] [d00000000f7aa034] .qla81xx_load_risc+0x84/0x1a0 [qla2xxx]
	[576544.715789] [c0000004d0ffb210] [d00000000f79f7c8] .qla2x00_setup_chip+0xc8/0x910 [qla2xxx]
	[576544.715808] [c0000004d0ffb300] [d00000000f7a631c] .qla2x00_initialize_adapter+0x4dc/0xb00 [qla2xxx]
	[576544.715826] [c0000004d0ffb3e0] [d00000000f78ce28] .qla2x00_probe_one+0xf08/0x2200 [qla2xxx]

Fixes: f73cb695d3ec ("[SCSI] qla2xxx: Add support for ISP2071.")
Signed-off-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_tmpl.c | 9 +++++----
 drivers/scsi/qla2xxx/qla_tmpl.h | 2 +-
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_tmpl.c b/drivers/scsi/qla2xxx/qla_tmpl.c
index 84f4416d366f..a6bb1c0e2245 100644
--- a/drivers/scsi/qla2xxx/qla_tmpl.c
+++ b/drivers/scsi/qla2xxx/qla_tmpl.c
@@ -928,7 +928,8 @@ qla27xx_template_checksum(void *p, ulong size)
 static inline int
 qla27xx_verify_template_checksum(struct qla27xx_fwdt_template *tmp)
 {
-	return qla27xx_template_checksum(tmp, tmp->template_size) == 0;
+	return qla27xx_template_checksum(tmp,
+		le32_to_cpu(tmp->template_size)) == 0;
 }
 
 static inline int
@@ -944,7 +945,7 @@ qla27xx_execute_fwdt_template(struct scsi_qla_host *vha,
 	ulong len = 0;
 
 	if (qla27xx_fwdt_template_valid(tmp)) {
-		len = tmp->template_size;
+		len = le32_to_cpu(tmp->template_size);
 		tmp = memcpy(buf, tmp, len);
 		ql27xx_edit_template(vha, tmp);
 		qla27xx_walk_template(vha, tmp, buf, &len);
@@ -960,7 +961,7 @@ qla27xx_fwdt_calculate_dump_size(struct scsi_qla_host *vha, void *p)
 	ulong len = 0;
 
 	if (qla27xx_fwdt_template_valid(tmp)) {
-		len = tmp->template_size;
+		len = le32_to_cpu(tmp->template_size);
 		qla27xx_walk_template(vha, tmp, NULL, &len);
 	}
 
@@ -972,7 +973,7 @@ qla27xx_fwdt_template_size(void *p)
 {
 	struct qla27xx_fwdt_template *tmp = p;
 
-	return tmp->template_size;
+	return le32_to_cpu(tmp->template_size);
 }
 
 int
diff --git a/drivers/scsi/qla2xxx/qla_tmpl.h b/drivers/scsi/qla2xxx/qla_tmpl.h
index c47184db5081..6e0987edfceb 100644
--- a/drivers/scsi/qla2xxx/qla_tmpl.h
+++ b/drivers/scsi/qla2xxx/qla_tmpl.h
@@ -12,7 +12,7 @@
 struct __packed qla27xx_fwdt_template {
 	__le32 template_type;
 	__le32 entry_offset;
-	uint32_t template_size;
+	__le32 template_size;
 	uint32_t count;		/* borrow field for running/residual count */
 
 	__le32 entry_count;
-- 
2.19.0.rc0

Re: [PATCH 07/15] qla2xxx: Fix crash during driver load on big endian machines

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Arun Easi <aeasi@marvell.com>
> 
> Crash stack:
> 	[576544.715489] Unable to handle kernel paging request for data at address 0xd00000000f970000
> 	[576544.715497] Faulting instruction address: 0xd00000000f880f64
> 	[576544.715503] Oops: Kernel access of bad area, sig: 11 [#1]
> 	[576544.715506] SMP NR_CPUS=2048 NUMA pSeries
> 	:
> 	[576544.715703] NIP [d00000000f880f64] .qla27xx_fwdt_template_valid+0x94/0x100 [qla2xxx]
> 	[576544.715722] LR [d00000000f7952dc] .qla24xx_load_risc_flash+0x2fc/0x590 [qla2xxx]
> 	[576544.715726] Call Trace:
> 	[576544.715731] [c0000004d0ffb000] [c0000006fe02c350] 0xc0000006fe02c350 (unreliable)
> 	[576544.715750] [c0000004d0ffb080] [d00000000f7952dc] .qla24xx_load_risc_flash+0x2fc/0x590 [qla2xxx]
> 	[576544.715770] [c0000004d0ffb170] [d00000000f7aa034] .qla81xx_load_risc+0x84/0x1a0 [qla2xxx]
> 	[576544.715789] [c0000004d0ffb210] [d00000000f79f7c8] .qla2x00_setup_chip+0xc8/0x910 [qla2xxx]
> 	[576544.715808] [c0000004d0ffb300] [d00000000f7a631c] .qla2x00_initialize_adapter+0x4dc/0xb00 [qla2xxx]
> 	[576544.715826] [c0000004d0ffb3e0] [d00000000f78ce28] .qla2x00_probe_one+0xf08/0x2200 [qla2xxx]
> 
> Fixes: f73cb695d3ec ("[SCSI] qla2xxx: Add support for ISP2071.")
> Signed-off-by: Arun Easi <aeasi@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>

I think this needs stable tag as well. 

> ---
> drivers/scsi/qla2xxx/qla_tmpl.c | 9 +++++----
> drivers/scsi/qla2xxx/qla_tmpl.h | 2 +-
> 2 files changed, 6 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_tmpl.c b/drivers/scsi/qla2xxx/qla_tmpl.c
> index 84f4416d366f..a6bb1c0e2245 100644
> --- a/drivers/scsi/qla2xxx/qla_tmpl.c
> +++ b/drivers/scsi/qla2xxx/qla_tmpl.c
> @@ -928,7 +928,8 @@ qla27xx_template_checksum(void *p, ulong size)
> static inline int
> qla27xx_verify_template_checksum(struct qla27xx_fwdt_template *tmp)
> {
> -	return qla27xx_template_checksum(tmp, tmp->template_size) == 0;
> +	return qla27xx_template_checksum(tmp,
> +		le32_to_cpu(tmp->template_size)) == 0;
> }
> 
> static inline int
> @@ -944,7 +945,7 @@ qla27xx_execute_fwdt_template(struct scsi_qla_host *vha,
> 	ulong len = 0;
> 
> 	if (qla27xx_fwdt_template_valid(tmp)) {
> -		len = tmp->template_size;
> +		len = le32_to_cpu(tmp->template_size);
> 		tmp = memcpy(buf, tmp, len);
> 		ql27xx_edit_template(vha, tmp);
> 		qla27xx_walk_template(vha, tmp, buf, &len);
> @@ -960,7 +961,7 @@ qla27xx_fwdt_calculate_dump_size(struct scsi_qla_host *vha, void *p)
> 	ulong len = 0;
> 
> 	if (qla27xx_fwdt_template_valid(tmp)) {
> -		len = tmp->template_size;
> +		len = le32_to_cpu(tmp->template_size);
> 		qla27xx_walk_template(vha, tmp, NULL, &len);
> 	}
> 
> @@ -972,7 +973,7 @@ qla27xx_fwdt_template_size(void *p)
> {
> 	struct qla27xx_fwdt_template *tmp = p;
> 
> -	return tmp->template_size;
> +	return le32_to_cpu(tmp->template_size);
> }
> 
> int
> diff --git a/drivers/scsi/qla2xxx/qla_tmpl.h b/drivers/scsi/qla2xxx/qla_tmpl.h
> index c47184db5081..6e0987edfceb 100644
> --- a/drivers/scsi/qla2xxx/qla_tmpl.h
> +++ b/drivers/scsi/qla2xxx/qla_tmpl.h
> @@ -12,7 +12,7 @@
> struct __packed qla27xx_fwdt_template {
> 	__le32 template_type;
> 	__le32 entry_offset;
> -	uint32_t template_size;
> +	__le32 template_size;
> 	uint32_t count;		/* borrow field for running/residual count */
> 
> 	__le32 entry_count;
> -- 
> 2.19.0.rc0
> 

Otherwise.. Looks good

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 08/15] qla2xxx: Fix FW initialization error on big endian machines

[Nilesh Javali]

From: Arun Easi <aeasi@marvell.com>

Some fields are not correctly byte swapped causing failure
during initialization. As probe() returns failure, HBAs
will not be claimed when this happens.

Signed-off-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_mbx.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
index 40af7f1524ce..1b4261c3c476 100644
--- a/drivers/scsi/qla2xxx/qla_mbx.c
+++ b/drivers/scsi/qla2xxx/qla_mbx.c
@@ -1129,7 +1129,7 @@ qla2x00_get_fw_version(scsi_qla_host_t *vha)
 		if (ha->flags.scm_supported_a &&
 		    (ha->fw_attributes_ext[0] & FW_ATTR_EXT0_SCM_SUPPORTED)) {
 			ha->flags.scm_supported_f = 1;
-			ha->sf_init_cb->flags |= BIT_13;
+			ha->sf_init_cb->flags |= cpu_to_le16(BIT_13);
 		}
 		ql_log(ql_log_info, vha, 0x11a3, "SCM in FW: %s\n",
 		       (ha->flags.scm_supported_f) ? "Supported" :
@@ -1137,9 +1137,9 @@ qla2x00_get_fw_version(scsi_qla_host_t *vha)
 
 		if (vha->flags.nvme2_enabled) {
 			/* set BIT_15 of special feature control block for SLER */
-			ha->sf_init_cb->flags |= BIT_15;
+			ha->sf_init_cb->flags |= cpu_to_le16(BIT_15);
 			/* set BIT_14 of special feature control block for PI CTRL*/
-			ha->sf_init_cb->flags |= BIT_14;
+			ha->sf_init_cb->flags |= cpu_to_le16(BIT_14);
 		}
 	}
 
-- 
2.19.0.rc0

Re: [PATCH 08/15] qla2xxx: Fix FW initialization error on big endian machines

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Arun Easi <aeasi@marvell.com>
> 
> Some fields are not correctly byte swapped causing failure
> during initialization. As probe() returns failure, HBAs
> will not be claimed when this happens.
> 
> Signed-off-by: Arun Easi <aeasi@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>

I think this patch needs following fixes tag

Fixes: 9f2475fe7406b ("scsi: qla2xxx: SAN congestion management implementation")
Fixes: cf3c54fb49a4e ("scsi: qla2xxx: Add SLER and PI control support”)

> ---
> drivers/scsi/qla2xxx/qla_mbx.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
> index 40af7f1524ce..1b4261c3c476 100644
> --- a/drivers/scsi/qla2xxx/qla_mbx.c
> +++ b/drivers/scsi/qla2xxx/qla_mbx.c
> @@ -1129,7 +1129,7 @@ qla2x00_get_fw_version(scsi_qla_host_t *vha)
> 		if (ha->flags.scm_supported_a &&
> 		    (ha->fw_attributes_ext[0] & FW_ATTR_EXT0_SCM_SUPPORTED)) {
> 			ha->flags.scm_supported_f = 1;
> -			ha->sf_init_cb->flags |= BIT_13;
> +			ha->sf_init_cb->flags |= cpu_to_le16(BIT_13);
> 		}
> 		ql_log(ql_log_info, vha, 0x11a3, "SCM in FW: %s\n",
> 		       (ha->flags.scm_supported_f) ? "Supported" :
> @@ -1137,9 +1137,9 @@ qla2x00_get_fw_version(scsi_qla_host_t *vha)
> 
> 		if (vha->flags.nvme2_enabled) {
> 			/* set BIT_15 of special feature control block for SLER */
> -			ha->sf_init_cb->flags |= BIT_15;
> +			ha->sf_init_cb->flags |= cpu_to_le16(BIT_15);
> 			/* set BIT_14 of special feature control block for PI CTRL*/
> -			ha->sf_init_cb->flags |= BIT_14;
> +			ha->sf_init_cb->flags |= cpu_to_le16(BIT_14);
> 		}
> 	}
> 
> -- 
> 2.19.0.rc0
> 

Otherwise, Looks good.

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 09/15] qla2xxx: fix N2N and NVME connect retry failure

[Nilesh Javali]

From: Quinn Tran <qutran@marvell.com>

FC-NVMe target discovery failed when initiator wwpn < target wwpn in an
N2N (Direct Attach) config, where the driver was stuck on FCP PRLI
mode and failed to retry with NVME PRLI.

Signed-off-by: Quinn Tran <qutran@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_init.c | 71 ++++++++++++++++++++++++---------
 drivers/scsi/qla2xxx/qla_mbx.c  |  3 --
 2 files changed, 52 insertions(+), 22 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index 5626e9b6949f..12e3b05baf41 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -1268,9 +1268,10 @@ qla24xx_async_prli(struct scsi_qla_host *vha, fc_port_t *fcport)
 		lio->u.logio.flags |= SRB_LOGIN_NVME_PRLI;
 
 	ql_dbg(ql_dbg_disc, vha, 0x211b,
-	    "Async-prli - %8phC hdl=%x, loopid=%x portid=%06x retries=%d %s.\n",
+	    "Async-prli - %8phC hdl=%x, loopid=%x portid=%06x retries=%d fc4type %x priority %x %s.\n",
 	    fcport->port_name, sp->handle, fcport->loop_id, fcport->d_id.b24,
-	    fcport->login_retry, NVME_TARGET(vha->hw, fcport) ? "nvme" : "fc");
+	    fcport->login_retry, fcport->fc4_type, vha->hw->fc4_type_priority,
+	    NVME_TARGET(vha->hw, fcport) ? "nvme" : "fcp");
 
 	rval = qla2x00_start_sp(sp);
 	if (rval != QLA_SUCCESS) {
@@ -1932,26 +1933,58 @@ qla24xx_handle_prli_done_event(struct scsi_qla_host *vha, struct event_arg *ea)
 			break;
 		}
 
-		/*
-		 * Retry PRLI with other FC-4 type if failure occurred on dual
-		 * FCP/NVMe port
-		 */
-		if (NVME_FCP_TARGET(ea->fcport)) {
-			ql_dbg(ql_dbg_disc, vha, 0x2118,
-				"%s %d %8phC post %s prli\n",
-				__func__, __LINE__, ea->fcport->port_name,
-				(ea->fcport->fc4_type & FS_FC4TYPE_NVME) ?
-				"NVMe" : "FCP");
-			if (vha->hw->fc4_type_priority == FC4_PRIORITY_NVME)
+		ql_dbg(ql_dbg_disc, vha, 0x2118,
+		       "%s %d %8phC priority %s, fc4type %x\n",
+		       __func__, __LINE__, ea->fcport->port_name,
+		       vha->hw->fc4_type_priority == FC4_PRIORITY_FCP ?
+		       "FCP" : "NVMe", ea->fcport->fc4_type);
+
+		if (N2N_TOPO(vha->hw)) {
+			if (vha->hw->fc4_type_priority == FC4_PRIORITY_NVME) {
 				ea->fcport->fc4_type &= ~FS_FC4TYPE_NVME;
-			else
+				ea->fcport->fc4_type |= FS_FC4TYPE_FCP;
+			} else {
 				ea->fcport->fc4_type &= ~FS_FC4TYPE_FCP;
-		}
+				ea->fcport->fc4_type |= FS_FC4TYPE_NVME;
+			}
 
-		ea->fcport->flags &= ~FCF_ASYNC_SENT;
-		ea->fcport->keep_nport_handle = 0;
-		ea->fcport->logout_on_delete = 1;
-		qlt_schedule_sess_for_deletion(ea->fcport);
+			if (ea->fcport->n2n_link_reset_cnt < 3) {
+				ea->fcport->n2n_link_reset_cnt++;
+				vha->relogin_jif = jiffies + 2 * HZ;
+				/*
+				 * PRLI failed. Reset link to kick start
+				 * state machine
+				 */
+				set_bit(N2N_LINK_RESET, &vha->dpc_flags);
+			} else {
+				ql_log(ql_log_warn, vha, 0x2119,
+				       "%s %d %8phC Unable to reconnect\n",
+				       __func__, __LINE__,
+				       ea->fcport->port_name);
+			}
+		} else {
+			/*
+			 * switch connect. login failed. Take connection down
+			 * and allow relogin to retrigger
+			 */
+			if (NVME_FCP_TARGET(ea->fcport)) {
+				ql_dbg(ql_dbg_disc, vha, 0x2118,
+				       "%s %d %8phC post %s prli\n",
+				       __func__, __LINE__,
+				       ea->fcport->port_name,
+				       (ea->fcport->fc4_type & FS_FC4TYPE_NVME)
+				       ? "NVMe" : "FCP");
+				if (vha->hw->fc4_type_priority == FC4_PRIORITY_NVME)
+					ea->fcport->fc4_type &= ~FS_FC4TYPE_NVME;
+				else
+					ea->fcport->fc4_type &= ~FS_FC4TYPE_FCP;
+			}
+
+			ea->fcport->flags &= ~FCF_ASYNC_SENT;
+			ea->fcport->keep_nport_handle = 0;
+			ea->fcport->logout_on_delete = 1;
+			qlt_schedule_sess_for_deletion(ea->fcport);
+		}
 		break;
 	}
 }
diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
index 1b4261c3c476..d7d4ab65009c 100644
--- a/drivers/scsi/qla2xxx/qla_mbx.c
+++ b/drivers/scsi/qla2xxx/qla_mbx.c
@@ -3998,9 +3998,6 @@ qla24xx_report_id_acquisition(scsi_qla_host_t *vha,
 				fcport->scan_state = QLA_FCPORT_FOUND;
 				fcport->n2n_flag = 1;
 				fcport->keep_nport_handle = 1;
-				fcport->fc4_type = FS_FC4TYPE_FCP;
-				if (vha->flags.nvme_enabled)
-					fcport->fc4_type |= FS_FC4TYPE_NVME;
 
 				if (wwn_to_u64(vha->port_name) >
 				    wwn_to_u64(fcport->port_name)) {
-- 
2.19.0.rc0

Re: [PATCH 09/15] qla2xxx: fix N2N and NVME connect retry failure

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Quinn Tran <qutran@marvell.com>
> 
> FC-NVMe target discovery failed when initiator wwpn < target wwpn in an
> N2N (Direct Attach) config, where the driver was stuck on FCP PRLI
> mode and failed to retry with NVME PRLI.
> 
> Signed-off-by: Quinn Tran <qutran@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>

Missing Fixes tag 

Fixes: 84ed362ac40ca ("scsi: qla2xxx: Dual FCP-NVMe target port support”)
Fixes: 983f127603fac ("scsi: qla2xxx: Retry PLOGI on FC-NVMe PRLI failure”)

> ---
> drivers/scsi/qla2xxx/qla_init.c | 71 ++++++++++++++++++++++++---------
> drivers/scsi/qla2xxx/qla_mbx.c  |  3 --
> 2 files changed, 52 insertions(+), 22 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
> index 5626e9b6949f..12e3b05baf41 100644
> --- a/drivers/scsi/qla2xxx/qla_init.c
> +++ b/drivers/scsi/qla2xxx/qla_init.c
> @@ -1268,9 +1268,10 @@ qla24xx_async_prli(struct scsi_qla_host *vha, fc_port_t *fcport)
> 		lio->u.logio.flags |= SRB_LOGIN_NVME_PRLI;
> 
> 	ql_dbg(ql_dbg_disc, vha, 0x211b,
> -	    "Async-prli - %8phC hdl=%x, loopid=%x portid=%06x retries=%d %s.\n",
> +	    "Async-prli - %8phC hdl=%x, loopid=%x portid=%06x retries=%d fc4type %x priority %x %s.\n",
> 	    fcport->port_name, sp->handle, fcport->loop_id, fcport->d_id.b24,
> -	    fcport->login_retry, NVME_TARGET(vha->hw, fcport) ? "nvme" : "fc");
> +	    fcport->login_retry, fcport->fc4_type, vha->hw->fc4_type_priority,
> +	    NVME_TARGET(vha->hw, fcport) ? "nvme" : "fcp");
> 
> 	rval = qla2x00_start_sp(sp);
> 	if (rval != QLA_SUCCESS) {
> @@ -1932,26 +1933,58 @@ qla24xx_handle_prli_done_event(struct scsi_qla_host *vha, struct event_arg *ea)
> 			break;
> 		}
> 
> -		/*
> -		 * Retry PRLI with other FC-4 type if failure occurred on dual
> -		 * FCP/NVMe port
> -		 */
> -		if (NVME_FCP_TARGET(ea->fcport)) {
> -			ql_dbg(ql_dbg_disc, vha, 0x2118,
> -				"%s %d %8phC post %s prli\n",
> -				__func__, __LINE__, ea->fcport->port_name,
> -				(ea->fcport->fc4_type & FS_FC4TYPE_NVME) ?
> -				"NVMe" : "FCP");
> -			if (vha->hw->fc4_type_priority == FC4_PRIORITY_NVME)
> +		ql_dbg(ql_dbg_disc, vha, 0x2118,
> +		       "%s %d %8phC priority %s, fc4type %x\n",
> +		       __func__, __LINE__, ea->fcport->port_name,
> +		       vha->hw->fc4_type_priority == FC4_PRIORITY_FCP ?
> +		       "FCP" : "NVMe", ea->fcport->fc4_type);
> +
> +		if (N2N_TOPO(vha->hw)) {
> +			if (vha->hw->fc4_type_priority == FC4_PRIORITY_NVME) {
> 				ea->fcport->fc4_type &= ~FS_FC4TYPE_NVME;
> -			else
> +				ea->fcport->fc4_type |= FS_FC4TYPE_FCP;
> +			} else {
> 				ea->fcport->fc4_type &= ~FS_FC4TYPE_FCP;
> -		}
> +				ea->fcport->fc4_type |= FS_FC4TYPE_NVME;
> +			}
> 
> -		ea->fcport->flags &= ~FCF_ASYNC_SENT;
> -		ea->fcport->keep_nport_handle = 0;
> -		ea->fcport->logout_on_delete = 1;
> -		qlt_schedule_sess_for_deletion(ea->fcport);
> +			if (ea->fcport->n2n_link_reset_cnt < 3) {
> +				ea->fcport->n2n_link_reset_cnt++;
> +				vha->relogin_jif = jiffies + 2 * HZ;
> +				/*
> +				 * PRLI failed. Reset link to kick start
> +				 * state machine
> +				 */
> +				set_bit(N2N_LINK_RESET, &vha->dpc_flags);
> +			} else {
> +				ql_log(ql_log_warn, vha, 0x2119,
> +				       "%s %d %8phC Unable to reconnect\n",
> +				       __func__, __LINE__,
> +				       ea->fcport->port_name);
> +			}
> +		} else {
> +			/*
> +			 * switch connect. login failed. Take connection down
> +			 * and allow relogin to retrigger
> +			 */
> +			if (NVME_FCP_TARGET(ea->fcport)) {
> +				ql_dbg(ql_dbg_disc, vha, 0x2118,
> +				       "%s %d %8phC post %s prli\n",
> +				       __func__, __LINE__,
> +				       ea->fcport->port_name,
> +				       (ea->fcport->fc4_type & FS_FC4TYPE_NVME)
> +				       ? "NVMe" : "FCP");
> +				if (vha->hw->fc4_type_priority == FC4_PRIORITY_NVME)
> +					ea->fcport->fc4_type &= ~FS_FC4TYPE_NVME;
> +				else
> +					ea->fcport->fc4_type &= ~FS_FC4TYPE_FCP;
> +			}
> +
> +			ea->fcport->flags &= ~FCF_ASYNC_SENT;
> +			ea->fcport->keep_nport_handle = 0;
> +			ea->fcport->logout_on_delete = 1;
> +			qlt_schedule_sess_for_deletion(ea->fcport);
> +		}
> 		break;
> 	}
> }
> diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
> index 1b4261c3c476..d7d4ab65009c 100644
> --- a/drivers/scsi/qla2xxx/qla_mbx.c
> +++ b/drivers/scsi/qla2xxx/qla_mbx.c
> @@ -3998,9 +3998,6 @@ qla24xx_report_id_acquisition(scsi_qla_host_t *vha,
> 				fcport->scan_state = QLA_FCPORT_FOUND;
> 				fcport->n2n_flag = 1;
> 				fcport->keep_nport_handle = 1;
> -				fcport->fc4_type = FS_FC4TYPE_FCP;
> -				if (vha->flags.nvme_enabled)
> -					fcport->fc4_type |= FS_FC4TYPE_NVME;
> 
> 				if (wwn_to_u64(vha->port_name) >
> 				    wwn_to_u64(fcport->port_name)) {
> -- 
> 2.19.0.rc0
> 

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 10/15] qla2xxx: Handle aborts correctly for port undergoing deletion

[Nilesh Javali]

From: Saurav Kashyap <skashyap@marvell.com>

Call trace observed while shutting down the adapter ports (LINK DOWN).
Handle aborts correctly.

Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_nvme.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c
index d4159d5a4ffd..eab559b3b257 100644
--- a/drivers/scsi/qla2xxx/qla_nvme.c
+++ b/drivers/scsi/qla2xxx/qla_nvme.c
@@ -227,7 +227,7 @@ static void qla_nvme_abort_work(struct work_struct *work)
 	       "%s called for sp=%p, hndl=%x on fcport=%p deleted=%d\n",
 	       __func__, sp, sp->handle, fcport, fcport->deleted);
 
-	if (!ha->flags.fw_started && fcport->deleted)
+	if (!ha->flags.fw_started || fcport->deleted)
 		goto out;
 
 	if (ha->flags.host_shutting_down) {
-- 
2.19.0.rc0

Re: [PATCH 10/15] qla2xxx: Handle aborts correctly for port undergoing deletion

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Saurav Kashyap <skashyap@marvell.com>
> 
> Call trace observed while shutting down the adapter ports (LINK DOWN).
> Handle aborts correctly.
> 

Could you post call trace in commit message. 

> Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_nvme.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c
> index d4159d5a4ffd..eab559b3b257 100644
> --- a/drivers/scsi/qla2xxx/qla_nvme.c
> +++ b/drivers/scsi/qla2xxx/qla_nvme.c
> @@ -227,7 +227,7 @@ static void qla_nvme_abort_work(struct work_struct *work)
> 	       "%s called for sp=%p, hndl=%x on fcport=%p deleted=%d\n",
> 	       __func__, sp, sp->handle, fcport, fcport->deleted);
> 
> -	if (!ha->flags.fw_started && fcport->deleted)
> +	if (!ha->flags.fw_started || fcport->deleted)
> 		goto out;
> 
> 	if (ha->flags.host_shutting_down) {
> -- 
> 2.19.0.rc0
> 

Change itself is fine. 

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 11/15] qla2xxx: Fix flash update in 28XX adapters on big endian machines

[Nilesh Javali]

From: Arun Easi <aeasi@marvell.com>

Flash update failed due to missing endian conversion in FLT region access
as well as in checksum computation.

Signed-off-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_sup.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_sup.c b/drivers/scsi/qla2xxx/qla_sup.c
index 0f92e9a044dc..f771fabcba59 100644
--- a/drivers/scsi/qla2xxx/qla_sup.c
+++ b/drivers/scsi/qla2xxx/qla_sup.c
@@ -2634,14 +2634,14 @@ qla28xx_extract_sfub_and_verify(struct scsi_qla_host *vha, uint32_t *buf,
 	    sizeof(struct secure_flash_update_block));
 
 	for (i = 0; i < (sizeof(struct secure_flash_update_block) >> 2); i++)
-		check_sum += p[i];
+		check_sum += le32_to_cpu(p[i]);
 
 	check_sum = (~check_sum) + 1;
 
-	if (check_sum != p[i]) {
+	if (check_sum != le32_to_cpu(p[i])) {
 		ql_log(ql_log_warn, vha, 0x7097,
 		    "SFUB checksum failed, 0x%x, 0x%x\n",
-		    check_sum, p[i]);
+		    check_sum, le32_to_cpu(p[i]));
 		return QLA_COMMAND_ERROR;
 	}
 
@@ -2721,7 +2721,7 @@ qla28xx_write_flash_data(scsi_qla_host_t *vha, uint32_t *dwptr, uint32_t faddr,
 	if (ha->flags.secure_adapter && region.attribute) {
 
 		ql_log(ql_log_warn + ql_dbg_verbose, vha, 0xffff,
-		    "Region %x is secure\n", region.code);
+		    "Region %x is secure\n", le16_to_cpu(region.code));
 
 		switch (le16_to_cpu(region.code)) {
 		case FLT_REG_FW:
@@ -2775,7 +2775,7 @@ qla28xx_write_flash_data(scsi_qla_host_t *vha, uint32_t *dwptr, uint32_t faddr,
 		default:
 			ql_log(ql_log_warn + ql_dbg_verbose, vha,
 			    0xffff, "Secure region %x not supported\n",
-			    region.code);
+			    le16_to_cpu(region.code));
 			rval = QLA_COMMAND_ERROR;
 			goto done;
 		}
-- 
2.19.0.rc0

Re: [PATCH 11/15] qla2xxx: Fix flash update in 28XX adapters on big endian machines

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Arun Easi <aeasi@marvell.com>
> 
> Flash update failed due to missing endian conversion in FLT region access
> as well as in checksum computation.
> 
> Signed-off-by: Arun Easi <aeasi@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_sup.c | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_sup.c b/drivers/scsi/qla2xxx/qla_sup.c
> index 0f92e9a044dc..f771fabcba59 100644
> --- a/drivers/scsi/qla2xxx/qla_sup.c
> +++ b/drivers/scsi/qla2xxx/qla_sup.c
> @@ -2634,14 +2634,14 @@ qla28xx_extract_sfub_and_verify(struct scsi_qla_host *vha, uint32_t *buf,
> 	    sizeof(struct secure_flash_update_block));
> 
> 	for (i = 0; i < (sizeof(struct secure_flash_update_block) >> 2); i++)
> -		check_sum += p[i];
> +		check_sum += le32_to_cpu(p[i]);
> 
> 	check_sum = (~check_sum) + 1;
> 
> -	if (check_sum != p[i]) {
> +	if (check_sum != le32_to_cpu(p[i])) {
> 		ql_log(ql_log_warn, vha, 0x7097,
> 		    "SFUB checksum failed, 0x%x, 0x%x\n",
> -		    check_sum, p[i]);
> +		    check_sum, le32_to_cpu(p[i]));
> 		return QLA_COMMAND_ERROR;
> 	}
> 
> @@ -2721,7 +2721,7 @@ qla28xx_write_flash_data(scsi_qla_host_t *vha, uint32_t *dwptr, uint32_t faddr,
> 	if (ha->flags.secure_adapter && region.attribute) {
> 
> 		ql_log(ql_log_warn + ql_dbg_verbose, vha, 0xffff,
> -		    "Region %x is secure\n", region.code);
> +		    "Region %x is secure\n", le16_to_cpu(region.code));
> 
> 		switch (le16_to_cpu(region.code)) {
> 		case FLT_REG_FW:
> @@ -2775,7 +2775,7 @@ qla28xx_write_flash_data(scsi_qla_host_t *vha, uint32_t *dwptr, uint32_t faddr,
> 		default:
> 			ql_log(ql_log_warn + ql_dbg_verbose, vha,
> 			    0xffff, "Secure region %x not supported\n",
> -			    region.code);
> +			    le16_to_cpu(region.code));
> 			rval = QLA_COMMAND_ERROR;
> 			goto done;
> 		}
> -- 
> 2.19.0.rc0
> 

Looks Good.

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 12/15] qla2xxx: Fix the call trace for flush workqueue

[Nilesh Javali]

From: Saurav Kashyap <skashyap@marvell.com>

The call trace was because workqueue was allocated without any
flags, added WQ_MEM_RECLAIM as flag while allocation.

kernel: workqueue: WQ_MEM_RECLAIM
kblockd:blk_mq_run_work_fn is flushing !WQ_MEM_RECLAIM qla2xxx_wq:0x0
kernel: WARNING: CPU: 0 PID: 2475 at
kernel/workqueue.c:2593 check_flush_dependency+0x110/0x130
kernel: CPU: 0 PID: 2475 Comm: kworker/0:1H Kdump:
loaded Tainted: G           OE    --------- -  - 4.18.0-193.el8.x86_64 #1
kernel: Hardware name: HPE ProLiant XL170r Gen10/ProLiant XL170r Gen10, BIOS U38 05/21/2019
kernel: Workqueue: kblockd blk_mq_run_work_fn
kernel: RIP: 0010:check_flush_dependency+0x110/0x130
kernel: Code: ff ff 48 8b 50 18 48 8d 8b b0 00 00 00 49 89 e8 48 81 c6 b0 00 00 00 48 c7 c7 00 1e e9
	95 c6 05 dc 9a 2f 01 01 e8 1a 42 fe ff <0f> 0b e9 0a ff ff ff 80 3d ca 9a 2f 01 0 0 75 95 e9 41 ff ff ff 90
kernel: RSP: 0018:ffffa40f48b2baf8 EFLAGS: 00010282
kernel: RAX: 0000000000000000 RBX: ffff946795282600 RCX: 0000000000000000
kernel: RDX: 000000000000005f RSI: ffffffff96a1af7f RDI: 0000000000000246
kernel: RBP: 0000000000000000 R08: ffffffff96a1af20 R09: 0000000000029480
kernel: R10: 00080c89bb3e7462 R11: 00000000000009ab R12: ffff946773628000
kernel: R13: 0000000000000282 R14: 0000000000000246 R15: ffffa40f48b2bb40
kernel: FS: 	0000000000000000(0000) 	GS:ffff94679fa00000(0000) knlGS:0000000000000000
kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
kernel: CR2: 00005570c4b60110 CR3: 000000029140a005 CR4: 00000000007606f0
kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
kernel: PKRU: 55555554
kernel: Call Trace:
kernel: flush_workqueue+0x13a/0x440
kernel: qla2x00_wait_for_sess_deletion+0x1d6/0x200 [qla2xxx]
kernel: ? finish_wait+0x80/0x80
kernel: qla2xxx_disable_port+0x2b/0x30 [qla2xxx]
kernel: qla2x00_process_vendor_specific+0x1dc9/0x2d20 [qla2xxx]
kernel: ? blk_rq_map_sg+0x195/0x570
kernel: qla24xx_bsg_request+0x1a3/0xf90 [qla2xxx]

Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_os.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index f9c8ae9d669e..a75edba2b334 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -3265,7 +3265,7 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
 	    "req->req_q_in=%p req->req_q_out=%p rsp->rsp_q_in=%p rsp->rsp_q_out=%p.\n",
 	    req->req_q_in, req->req_q_out, rsp->rsp_q_in, rsp->rsp_q_out);
 
-	ha->wq = alloc_workqueue("qla2xxx_wq", 0, 0);
+	ha->wq = alloc_workqueue("qla2xxx_wq", WQ_MEM_RECLAIM, 0);
 	if (unlikely(!ha->wq)) {
 		ret = -ENOMEM;
 		goto probe_failed;
-- 
2.19.0.rc0

Re: [PATCH 12/15] qla2xxx: Fix the call trace for flush workqueue

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Saurav Kashyap <skashyap@marvell.com>
> 
> The call trace was because workqueue was allocated without any
> flags, added WQ_MEM_RECLAIM as flag while allocation.
> 
> kernel: workqueue: WQ_MEM_RECLAIM
> kblockd:blk_mq_run_work_fn is flushing !WQ_MEM_RECLAIM qla2xxx_wq:0x0
> kernel: WARNING: CPU: 0 PID: 2475 at
> kernel/workqueue.c:2593 check_flush_dependency+0x110/0x130
> kernel: CPU: 0 PID: 2475 Comm: kworker/0:1H Kdump:
> loaded Tainted: G           OE    --------- -  - 4.18.0-193.el8.x86_64 #1
> kernel: Hardware name: HPE ProLiant XL170r Gen10/ProLiant XL170r Gen10, BIOS U38 05/21/2019
> kernel: Workqueue: kblockd blk_mq_run_work_fn
> kernel: RIP: 0010:check_flush_dependency+0x110/0x130
> kernel: Code: ff ff 48 8b 50 18 48 8d 8b b0 00 00 00 49 89 e8 48 81 c6 b0 00 00 00 48 c7 c7 00 1e e9
> 	95 c6 05 dc 9a 2f 01 01 e8 1a 42 fe ff <0f> 0b e9 0a ff ff ff 80 3d ca 9a 2f 01 0 0 75 95 e9 41 ff ff ff 90
> kernel: RSP: 0018:ffffa40f48b2baf8 EFLAGS: 00010282
> kernel: RAX: 0000000000000000 RBX: ffff946795282600 RCX: 0000000000000000
> kernel: RDX: 000000000000005f RSI: ffffffff96a1af7f RDI: 0000000000000246
> kernel: RBP: 0000000000000000 R08: ffffffff96a1af20 R09: 0000000000029480
> kernel: R10: 00080c89bb3e7462 R11: 00000000000009ab R12: ffff946773628000
> kernel: R13: 0000000000000282 R14: 0000000000000246 R15: ffffa40f48b2bb40
> kernel: FS: 	0000000000000000(0000) 	GS:ffff94679fa00000(0000) knlGS:0000000000000000
> kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> kernel: CR2: 00005570c4b60110 CR3: 000000029140a005 CR4: 00000000007606f0
> kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> kernel: PKRU: 55555554
> kernel: Call Trace:
> kernel: flush_workqueue+0x13a/0x440
> kernel: qla2x00_wait_for_sess_deletion+0x1d6/0x200 [qla2xxx]
> kernel: ? finish_wait+0x80/0x80
> kernel: qla2xxx_disable_port+0x2b/0x30 [qla2xxx]
> kernel: qla2x00_process_vendor_specific+0x1dc9/0x2d20 [qla2xxx]
> kernel: ? blk_rq_map_sg+0x195/0x570
> kernel: qla24xx_bsg_request+0x1a3/0xf90 [qla2xxx]
> 
> Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_os.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
> index f9c8ae9d669e..a75edba2b334 100644
> --- a/drivers/scsi/qla2xxx/qla_os.c
> +++ b/drivers/scsi/qla2xxx/qla_os.c
> @@ -3265,7 +3265,7 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id)
> 	    "req->req_q_in=%p req->req_q_out=%p rsp->rsp_q_in=%p rsp->rsp_q_out=%p.\n",
> 	    req->req_q_in, req->req_q_out, rsp->rsp_q_in, rsp->rsp_q_out);
> 
> -	ha->wq = alloc_workqueue("qla2xxx_wq", 0, 0);
> +	ha->wq = alloc_workqueue("qla2xxx_wq", WQ_MEM_RECLAIM, 0);
> 	if (unlikely(!ha->wq)) {
> 		ret = -ENOMEM;
> 		goto probe_failed;
> -- 
> 2.19.0.rc0
> 

Looks Good. 

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 13/15] qla2xxx: If fcport is undergoing deletion return IO with retry

[Nilesh Javali]

From: Saurav Kashyap <skashyap@marvell.com>

Driver unload with IOs causes server to crash.
Return IO with retry if fcport undergoing deletion.

Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_os.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index a75edba2b334..be9d10092dd3 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -884,8 +884,8 @@ qla2xxx_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd)
 			goto qc24_fail_command;
 	}
 
-	if (!fcport) {
-		cmd->result = DID_NO_CONNECT << 16;
+	if (!fcport || fcport->deleted) {
+		cmd->result = DID_IMM_RETRY << 16;
 		goto qc24_fail_command;
 	}
 
@@ -966,8 +966,8 @@ qla2xxx_mqueuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd,
 		goto qc24_fail_command;
 	}
 
-	if (!fcport) {
-		cmd->result = DID_NO_CONNECT << 16;
+	if (!fcport || fcport->deleted) {
+		cmd->result = DID_IMM_RETRY << 16;
 		goto qc24_fail_command;
 	}
 
-- 
2.19.0.rc0

Re: [PATCH 13/15] qla2xxx: If fcport is undergoing deletion return IO with retry

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Saurav Kashyap <skashyap@marvell.com>
> 
> Driver unload with IOs causes server to crash.
> Return IO with retry if fcport undergoing deletion.
> 

Any call stack or panic signature to share in commit message?

> Signed-off-by: Saurav Kashyap <skashyap@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_os.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
> index a75edba2b334..be9d10092dd3 100644
> --- a/drivers/scsi/qla2xxx/qla_os.c
> +++ b/drivers/scsi/qla2xxx/qla_os.c
> @@ -884,8 +884,8 @@ qla2xxx_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd)
> 			goto qc24_fail_command;
> 	}
> 
> -	if (!fcport) {
> -		cmd->result = DID_NO_CONNECT << 16;
> +	if (!fcport || fcport->deleted) {
> +		cmd->result = DID_IMM_RETRY << 16;
> 		goto qc24_fail_command;
> 	}
> 
> @@ -966,8 +966,8 @@ qla2xxx_mqueuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd,
> 		goto qc24_fail_command;
> 	}
> 
> -	if (!fcport) {
> -		cmd->result = DID_NO_CONNECT << 16;
> +	if (!fcport || fcport->deleted) {
> +		cmd->result = DID_IMM_RETRY << 16;
> 		goto qc24_fail_command;
> 	}
> 
> -- 
> 2.19.0.rc0
> 

Patch itself looks good.

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 14/15] qla2xxx: Fix device loss on 4G and older HBAs.

[Nilesh Javali]

From: Arun Easi <aeasi@marvell.com>

Due to a bug in the older scan logic, when a once lost device
re-appeared, it was not discovered. Fix this by resetting login_retry
counter upon device discovery.

This is applicable only for 4G and older HBAs.

Signed-off-by: Arun Easi <aeasi@marvell.com>
Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_init.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index 12e3b05baf41..dcc0f0d823db 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -5982,6 +5982,9 @@ qla2x00_find_all_fabric_devs(scsi_qla_host_t *vha)
 				break;
 			}
 
+			if (fcport->login_retry == 0)
+				fcport->login_retry =
+					vha->hw->login_retry_count;
 			/*
 			 * If device was not a fabric device before.
 			 */
-- 
2.19.0.rc0

Re: [PATCH 14/15] qla2xxx: Fix device loss on 4G and older HBAs.

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> From: Arun Easi <aeasi@marvell.com>
> 
> Due to a bug in the older scan logic, when a once lost device
> re-appeared, it was not discovered. Fix this by resetting login_retry
> counter upon device discovery.
> 
> This is applicable only for 4G and older HBAs.
> 
> Signed-off-by: Arun Easi <aeasi@marvell.com>
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_init.c | 3 +++
> 1 file changed, 3 insertions(+)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
> index 12e3b05baf41..dcc0f0d823db 100644
> --- a/drivers/scsi/qla2xxx/qla_init.c
> +++ b/drivers/scsi/qla2xxx/qla_init.c
> @@ -5982,6 +5982,9 @@ qla2x00_find_all_fabric_devs(scsi_qla_host_t *vha)
> 				break;
> 			}
> 
> +			if (fcport->login_retry == 0)
> +				fcport->login_retry =
> +					vha->hw->login_retry_count;
> 			/*
> 			 * If device was not a fabric device before.
> 			 */
> -- 
> 2.19.0.rc0
> 


Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering

[PATCH 15/15] qla2xxx: Update version to 10.02.00.104-k

[Nilesh Javali]

Signed-off-by: Nilesh Javali <njavali@marvell.com>
---
 drivers/scsi/qla2xxx/qla_version.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_version.h b/drivers/scsi/qla2xxx/qla_version.h
index c2d4da52f4a9..ccec858875dd 100644
--- a/drivers/scsi/qla2xxx/qla_version.h
+++ b/drivers/scsi/qla2xxx/qla_version.h
@@ -6,9 +6,9 @@
 /*
  * Driver version
  */
-#define QLA2XXX_VERSION      "10.02.00.103-k"
+#define QLA2XXX_VERSION      "10.02.00.104-k"
 
 #define QLA_DRIVER_MAJOR_VER	10
 #define QLA_DRIVER_MINOR_VER	2
 #define QLA_DRIVER_PATCH_VER	0
-#define QLA_DRIVER_BETA_VER	103
+#define QLA_DRIVER_BETA_VER	104
-- 
2.19.0.rc0

Re: [PATCH 15/15] qla2xxx: Update version to 10.02.00.104-k

[Himanshu Madhani]

> On Dec 1, 2020, at 2:27 AM, Nilesh Javali <njavali@marvell.com> wrote:
> 
> Signed-off-by: Nilesh Javali <njavali@marvell.com>
> ---
> drivers/scsi/qla2xxx/qla_version.h | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/scsi/qla2xxx/qla_version.h b/drivers/scsi/qla2xxx/qla_version.h
> index c2d4da52f4a9..ccec858875dd 100644
> --- a/drivers/scsi/qla2xxx/qla_version.h
> +++ b/drivers/scsi/qla2xxx/qla_version.h
> @@ -6,9 +6,9 @@
> /*
>  * Driver version
>  */
> -#define QLA2XXX_VERSION      "10.02.00.103-k"
> +#define QLA2XXX_VERSION      "10.02.00.104-k"
> 
> #define QLA_DRIVER_MAJOR_VER	10
> #define QLA_DRIVER_MINOR_VER	2
> #define QLA_DRIVER_PATCH_VER	0
> -#define QLA_DRIVER_BETA_VER	103
> +#define QLA_DRIVER_BETA_VER	104
> -- 
> 2.19.0.rc0
> 

Reviewed-by: Himanshu Madhani <himanshu.madhani@oracle.com>

--
Himanshu Madhani	 Oracle Linux Engineering