[PATCH 0/3] hpsa compatility fixes

[PATCH 0/3] hpsa compatility fixes

[Hannes Reinecke]

Hi all,

this is a patchset to get hpsa working on older controllers.
Most notably older controller do not support the extended REPORT PHYS
command, so we need to fallback to the legacy version here.
I've also added a new sysfs HBA attribute 'ctlr_num', which displays
the current controller number. With that attribute one can construct
legacy '/dev/ccissX' symlinks to facilitate an update from older
installations using the cciss driver.

As usual, comments and reviews are welcome.

Hannes Reinecke (3):
  hpsa: use correct DID_NO_CONNECT hostbyte
  hpsa: fallback to use legacy REPORT PHYS command
  hpsa: add 'ctlr_num' sysfs attribute

 drivers/scsi/hpsa.c | 43 ++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 40 insertions(+), 3 deletions(-)

-- 
1.8.5.6

[PATCH 1/3] hpsa: use correct DID_NO_CONNECT hostbyte

[Hannes Reinecke]

NOT_READY is a sense key, not a legit scsi hostbyte value.
Use DID_NO_CONNECT instead.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/hpsa.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index ea64c01..488e17c 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -5493,7 +5493,7 @@ static int hpsa_scsi_queue_command(struct Scsi_Host *sh, struct scsi_cmnd *cmd)
 
 	dev = cmd->device->hostdata;
 	if (!dev) {
-		cmd->result = NOT_READY << 16; /* host byte */
+		cmd->result = DID_NO_CONNECT << 16;
 		cmd->scsi_done(cmd);
 		return 0;
 	}
-- 
1.8.5.6

[PATCH 2/3] hpsa: fallback to use legacy REPORT PHYS command

[Hannes Reinecke]

Older SmartArray controller do not support the extended REPORT PHYS
command, so fallback to use the legacy version here.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/hpsa.c | 28 ++++++++++++++++++++++++++--
 1 file changed, 26 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 488e17c..3d89f74 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -3628,8 +3628,32 @@ static int hpsa_scsi_do_report_luns(struct ctlr_info *h, int logical,
 static inline int hpsa_scsi_do_report_phys_luns(struct ctlr_info *h,
 		struct ReportExtendedLUNdata *buf, int bufsize)
 {
-	return hpsa_scsi_do_report_luns(h, 0, buf, bufsize,
-						HPSA_REPORT_PHYS_EXTENDED);
+	int rc;
+	struct ReportLUNdata *lbuf;
+
+	rc = hpsa_scsi_do_report_luns(h, 0, buf, bufsize,
+				      HPSA_REPORT_PHYS_EXTENDED);
+	if (!rc || !hpsa_allow_any)
+		return rc;
+
+	/* REPORT PHYS EXTENDED is not supported */
+	lbuf = kzalloc(sizeof(*lbuf), GFP_KERNEL);
+	if (!lbuf)
+		return -ENOMEM;
+
+	rc = hpsa_scsi_do_report_luns(h, 0, lbuf, sizeof(*lbuf), 0);
+	if (!rc) {
+		int i;
+		u32 nphys;
+
+		/* Copy ReportLUNdata header */
+		memcpy(buf, lbuf, 8);
+		nphys = be32_to_cpu(*((__be32 *)lbuf->LUNListLength)) / 24;
+		for (i = 0; i < nphys; i++)
+			memcpy(buf->LUN[i].lunid, lbuf->LUN[i], 8);
+	}
+	kfree(lbuf);
+	return rc;
 }
 
 static inline int hpsa_scsi_do_report_log_luns(struct ctlr_info *h,
-- 
1.8.5.6

[PATCH 3/3] hpsa: add 'ctlr_num' sysfs attribute

[Hannes Reinecke]

Add a sysfs attribute 'ctlr_num' holding the current HPSA controller
number. This is required to construct compability 'cciss' links.

Signed-off-by: Hannes Reinecke <hare@suse.com>
---
 drivers/scsi/hpsa.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 3d89f74..5834117 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -867,6 +867,16 @@ static ssize_t path_info_show(struct device *dev,
 	return output_len;
 }
 
+static ssize_t host_show_ctlr_num(struct device *dev,
+	struct device_attribute *attr, char *buf)
+{
+	struct ctlr_info *h;
+	struct Scsi_Host *shost = class_to_shost(dev);
+
+	h = shost_to_hba(shost);
+	return snprintf(buf, 20, "%d\n", h->ctlr);
+}
+
 static DEVICE_ATTR(raid_level, S_IRUGO, raid_level_show, NULL);
 static DEVICE_ATTR(lunid, S_IRUGO, lunid_show, NULL);
 static DEVICE_ATTR(unique_id, S_IRUGO, unique_id_show, NULL);
@@ -890,6 +900,8 @@ static DEVICE_ATTR(resettable, S_IRUGO,
 	host_show_resettable, NULL);
 static DEVICE_ATTR(lockup_detected, S_IRUGO,
 	host_show_lockup_detected, NULL);
+static DEVICE_ATTR(ctlr_num, S_IRUGO,
+	host_show_ctlr_num, NULL);
 
 static struct device_attribute *hpsa_sdev_attrs[] = {
 	&dev_attr_raid_level,
@@ -910,6 +922,7 @@ static DEVICE_ATTR(lockup_detected, S_IRUGO,
 	&dev_attr_hp_ssd_smart_path_status,
 	&dev_attr_raid_offload_debug,
 	&dev_attr_lockup_detected,
+	&dev_attr_ctlr_num,
 	NULL,
 };
 
-- 
1.8.5.6

Re: [PATCH 2/3] hpsa: fallback to use legacy REPORT PHYS command

[Martin Wilck]

On Fri, 2016-11-18 at 08:32 +0100, Hannes Reinecke wrote:
> Older SmartArray controller do not support the extended REPORT PHYS
> command, so fallback to use the legacy version here.
> 
> Signed-off-by: Hannes Reinecke <hare@suse.com>

Tested-by: Martin Wilck <mwilck@suse.com>

> ---
>  drivers/scsi/hpsa.c | 28 ++++++++++++++++++++++++++--
>  1 file changed, 26 insertions(+), 2 deletions(-)

-- 
Dr. Martin Wilck <mwilck@suse.com>, Tel. +49 (0)911 74053 2107
SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton
HRB 21284 (AG Nürnberg)

RE: [PATCH 1/3] hpsa: use correct DID_NO_CONNECT hostbyte

[Don Brace]

> -----Original Message-----
> From: Hannes Reinecke [mailto:hare@suse.de]
> Sent: Friday, November 18, 2016 1:33 AM
> To: Martin K. Petersen
> Cc: Christoph Hellwig; James Bottomley; Don Brace; Martin Wilck; linux-
> scsi@vger.kernel.org; Hannes Reinecke; Hannes Reinecke
> Subject: [PATCH 1/3] hpsa: use correct DID_NO_CONNECT hostbyte
> 
> EXTERNAL EMAIL
> 
> 
> NOT_READY is a sense key, not a legit scsi hostbyte value.
> Use DID_NO_CONNECT instead.
> 
> Signed-off-by: Hannes Reinecke <hare@suse.com>
> ---
>  drivers/scsi/hpsa.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
> index ea64c01..488e17c 100644
> --- a/drivers/scsi/hpsa.c
> +++ b/drivers/scsi/hpsa.c
> @@ -5493,7 +5493,7 @@ static int hpsa_scsi_queue_command(struct
> Scsi_Host *sh, struct scsi_cmnd *cmd)
> 
>         dev = cmd->device->hostdata;
>         if (!dev) {
> -               cmd->result = NOT_READY << 16; /* host byte */
> +               cmd->result = DID_NO_CONNECT << 16;
>                 cmd->scsi_done(cmd);
>                 return 0;
>         }
> --
> 1.8.5.6

Acked-by: Don Brace <don.brace@microsemi.com>

RE: [PATCH 3/3] hpsa: add 'ctlr_num' sysfs attribute

[Don Brace]

> -----Original Message-----
> From: Hannes Reinecke [mailto:hare@suse.de]
> Sent: Friday, November 18, 2016 1:33 AM
> To: Martin K. Petersen
> Cc: Christoph Hellwig; James Bottomley; Don Brace; Martin Wilck; linux-
> scsi@vger.kernel.org; Hannes Reinecke; Hannes Reinecke
> Subject: [PATCH 3/3] hpsa: add 'ctlr_num' sysfs attribute
> 
> EXTERNAL EMAIL
> 
> 
> Add a sysfs attribute 'ctlr_num' holding the current HPSA controller
> number. This is required to construct compability 'cciss' links.
> 
> Signed-off-by: Hannes Reinecke <hare@suse.com>
> ---
>  drivers/scsi/hpsa.c | 13 +++++++++++++
>  1 file changed, 13 insertions(+)
> 
> diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
> index 3d89f74..5834117 100644
> --- a/drivers/scsi/hpsa.c
> +++ b/drivers/scsi/hpsa.c
> @@ -867,6 +867,16 @@ static ssize_t path_info_show(struct device *dev,
>         return output_len;
>  }
> 
> +static ssize_t host_show_ctlr_num(struct device *dev,
> +       struct device_attribute *attr, char *buf)
> +{
> +       struct ctlr_info *h;
> +       struct Scsi_Host *shost = class_to_shost(dev);
> +
> +       h = shost_to_hba(shost);
> +       return snprintf(buf, 20, "%d\n", h->ctlr);
> +}
> +
>  static DEVICE_ATTR(raid_level, S_IRUGO, raid_level_show, NULL);
>  static DEVICE_ATTR(lunid, S_IRUGO, lunid_show, NULL);
>  static DEVICE_ATTR(unique_id, S_IRUGO, unique_id_show, NULL);
> @@ -890,6 +900,8 @@ static DEVICE_ATTR(resettable, S_IRUGO,
>         host_show_resettable, NULL);
>  static DEVICE_ATTR(lockup_detected, S_IRUGO,
>         host_show_lockup_detected, NULL);
> +static DEVICE_ATTR(ctlr_num, S_IRUGO,
> +       host_show_ctlr_num, NULL);
> 
>  static struct device_attribute *hpsa_sdev_attrs[] = {
>         &dev_attr_raid_level,
> @@ -910,6 +922,7 @@ static DEVICE_ATTR(lockup_detected, S_IRUGO,
>         &dev_attr_hp_ssd_smart_path_status,
>         &dev_attr_raid_offload_debug,
>         &dev_attr_lockup_detected,
> +       &dev_attr_ctlr_num,
>         NULL,
>  };
> 
> --
> 1.8.5.6

Acked-by: Don Brace <don.brace@microsemi.com>

Re: [PATCH 2/3] hpsa: fallback to use legacy REPORT PHYS command

[Martin K. Petersen]

>>>>> "Hannes" == Hannes Reinecke <hare@suse.de> writes:

Hannes> Older SmartArray controller do not support the extended REPORT
Hannes> PHYS command, so fallback to use the legacy version here.

Don: Please review.

Thanks!

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: [PATCH 0/3] hpsa compatility fixes

[Martin K. Petersen]

>>>>> "Hannes" == Hannes Reinecke <hare@suse.de> writes:

Hannes> Hi all, this is a patchset to get hpsa working on older
Hannes> controllers.  Most notably older controller do not support the
Hannes> extended REPORT PHYS command, so we need to fallback to the
Hannes> legacy version here.  I've also added a new sysfs HBA attribute
Hannes> 'ctlr_num', which displays the current controller number. With
Hannes> that attribute one can construct legacy '/dev/ccissX' symlinks
Hannes> to facilitate an update from older installations using the cciss
Hannes> driver.

Applied patches 1 and 3. Patch 2 is awaiting feedback from Microsemi.

-- 
Martin K. Petersen	Oracle Linux Engineering

RE: [PATCH 2/3] hpsa: fallback to use legacy REPORT PHYS command

[Don Brace]

> -----Original Message-----
> From: Hannes Reinecke [mailto:hare@suse.de]
> Sent: Friday, November 18, 2016 1:33 AM
> To: Martin K. Petersen
> Cc: Christoph Hellwig; James Bottomley; Don Brace; Martin Wilck; linux-
> scsi@vger.kernel.org; Hannes Reinecke; Hannes Reinecke
> Subject: [PATCH 2/3] hpsa: fallback to use legacy REPORT PHYS command
> 
> EXTERNAL EMAIL
> 
> 
> Older SmartArray controller do not support the extended REPORT PHYS
> command, so fallback to use the legacy version here.
> 
> Signed-off-by: Hannes Reinecke <hare@suse.com>
> ---
>  drivers/scsi/hpsa.c | 28 ++++++++++++++++++++++++++--
>  1 file changed, 26 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
> index 488e17c..3d89f74 100644
> --- a/drivers/scsi/hpsa.c
> +++ b/drivers/scsi/hpsa.c
> @@ -3628,8 +3628,32 @@ static int hpsa_scsi_do_report_luns(struct
> ctlr_info *h, int logical,
>  static inline int hpsa_scsi_do_report_phys_luns(struct ctlr_info *h,
>                 struct ReportExtendedLUNdata *buf, int bufsize)
>  {
> -       return hpsa_scsi_do_report_luns(h, 0, buf, bufsize,
> -                                               HPSA_REPORT_PHYS_EXTENDED);
> +       int rc;
> +       struct ReportLUNdata *lbuf;
> +
> +       rc = hpsa_scsi_do_report_luns(h, 0, buf, bufsize,
> +                                     HPSA_REPORT_PHYS_EXTENDED);
> +       if (!rc || !hpsa_allow_any)
> +               return rc;
> +
> +       /* REPORT PHYS EXTENDED is not supported */
> +       lbuf = kzalloc(sizeof(*lbuf), GFP_KERNEL);
> +       if (!lbuf)
> +               return -ENOMEM;
> +
> +       rc = hpsa_scsi_do_report_luns(h, 0, lbuf, sizeof(*lbuf), 0);
> +       if (!rc) {
> +               int i;
> +               u32 nphys;
> +
> +               /* Copy ReportLUNdata header */
> +               memcpy(buf, lbuf, 8);
> +               nphys = be32_to_cpu(*((__be32 *)lbuf->LUNListLength)) / 24;
> +               for (i = 0; i < nphys; i++)
> +                       memcpy(buf->LUN[i].lunid, lbuf->LUN[i], 8);
> +       }
> +       kfree(lbuf);
> +       return rc;
>  }
> 
>  static inline int hpsa_scsi_do_report_log_luns(struct ctlr_info *h,
> --
> 1.8.5.6

For a non-extended REPORT_PHYS, the buffer size is 8, so the divisor needs to be 8.
REPORT_PHYS_EXTENDED has been around a long time, which controller and fw are you testing this on so I can
Also run some tests.

Thanks,
Don Brace
ESC - Smart Storage
Microsemi Corporation

Re: [PATCH 2/3] hpsa: fallback to use legacy REPORT PHYS command

[Hannes Reinecke]

On 12/01/2016 10:50 PM, Don Brace wrote:
>
>> -----Original Message-----
>> From: Hannes Reinecke [mailto:hare@suse.de]
>> Sent: Friday, November 18, 2016 1:33 AM
>> To: Martin K. Petersen
>> Cc: Christoph Hellwig; James Bottomley; Don Brace; Martin Wilck; linux-
>> scsi@vger.kernel.org; Hannes Reinecke; Hannes Reinecke
>> Subject: [PATCH 2/3] hpsa: fallback to use legacy REPORT PHYS command
>>
>> EXTERNAL EMAIL
>>
>>
>> Older SmartArray controller do not support the extended REPORT PHYS
>> command, so fallback to use the legacy version here.
>>
>> Signed-off-by: Hannes Reinecke <hare@suse.com>
>> ---
>>  drivers/scsi/hpsa.c | 28 ++++++++++++++++++++++++++--
>>  1 file changed, 26 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
>> index 488e17c..3d89f74 100644
>> --- a/drivers/scsi/hpsa.c
>> +++ b/drivers/scsi/hpsa.c
>> @@ -3628,8 +3628,32 @@ static int hpsa_scsi_do_report_luns(struct
>> ctlr_info *h, int logical,
>>  static inline int hpsa_scsi_do_report_phys_luns(struct ctlr_info *h,
>>                 struct ReportExtendedLUNdata *buf, int bufsize)
>>  {
>> -       return hpsa_scsi_do_report_luns(h, 0, buf, bufsize,
>> -                                               HPSA_REPORT_PHYS_EXTENDED);
>> +       int rc;
>> +       struct ReportLUNdata *lbuf;
>> +
>> +       rc = hpsa_scsi_do_report_luns(h, 0, buf, bufsize,
>> +                                     HPSA_REPORT_PHYS_EXTENDED);
>> +       if (!rc || !hpsa_allow_any)
>> +               return rc;
>> +
>> +       /* REPORT PHYS EXTENDED is not supported */
>> +       lbuf = kzalloc(sizeof(*lbuf), GFP_KERNEL);
>> +       if (!lbuf)
>> +               return -ENOMEM;
>> +
>> +       rc = hpsa_scsi_do_report_luns(h, 0, lbuf, sizeof(*lbuf), 0);
>> +       if (!rc) {
>> +               int i;
>> +               u32 nphys;
>> +
>> +               /* Copy ReportLUNdata header */
>> +               memcpy(buf, lbuf, 8);
>> +               nphys = be32_to_cpu(*((__be32 *)lbuf->LUNListLength)) / 24;
>> +               for (i = 0; i < nphys; i++)
>> +                       memcpy(buf->LUN[i].lunid, lbuf->LUN[i], 8);
>> +       }
>> +       kfree(lbuf);
>> +       return rc;
>>  }
>>
>>  static inline int hpsa_scsi_do_report_log_luns(struct ctlr_info *h,
>> --
>> 1.8.5.6
>
> For a non-extended REPORT_PHYS, the buffer size is 8, so the divisor needs to be 8.
> REPORT_PHYS_EXTENDED has been around a long time, which controller and fw are you testing this on so I can
> Also run some tests.
>
That was seen on an Smart Array 64xx.
Yes, I know, horribly ancient.

But this was sort-of the idea, namely that we can use 'hpsa' as a 
complete replacement for cciss.

I'll be fixing up the patch and resubmit it.

Cheers,

Hannes
-- 
Dr. Hannes Reinecke		   Teamlead Storage & Networking
hare@suse.de			               +49 911 74053 688
SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: F. Imendörffer, J. Smithard, J. Guild, D. Upmanyu, G. Norton
HRB 21284 (AG Nürnberg)