[PATCH 00/12] hpsa updates

[PATCH 00/12] hpsa updates

[Don Brace]

These patches are based on Linus's tree

The changes are:

hpsa: update discovery polling
 - Correct corner case for RBODs with no initial volumes.

hpsa: add controller checkpoint
 - Provide method to generate a controller checkpoint.

hpsa: change timeout for internal cmds
 - There are times when the default time-out value has
   proven to be too small.

hpsa: correct smart path enabled
 - Correct some corner cases where the ioaccel path is not
   re-enabled after a multipath failover or a transformation.

hpsa: update queue depth for externals
 - A re-scan can change the queue-depth of external
   logical volumes.

hpsa: reduce warning messages on device removal
 - Repeated warnings are not necessary.

hpsa: correct logical volume removal
 - Also check for outstanding commands during logical volume removal.

hpsa: add enclosure logical identifier
 - Populate sysfs enclosure_identifier.

hpsa: destroy sas transport properties before scsi_host
 - This patch was proposed by Martin Wilck on 10/21/2016:
   https://www.spinics.net/lists/linux-scsi/msg102085.html
 - but had no effect until Hannes's patch:
   commit 9441284fbc39 ("scsi-fixup-kernel-warning-during-rmmod")
   was applied to the kernel.
 - Re-submitting with Martin Wilck's permission.

hpsa: cleanup sas_phy structures in sysfs when unloading
 - This patch was proposed by Martin Wilck on 10/21/2016:
   https://www.spinics.net/lists/linux-scsi/msg102083.html
 - cleans up dangling symlinks in sysfs.
 - Re-submitting with Martin Wilck's permission.

hpsa: clear tmpdevice in scan thread
 - Clean up some meta data issues for devices during a re-scan.

hpsa: bump driver version

---

Bader Ali Saleh (1):
      hpsa: update discovery polling

Don Brace (8):
      hpsa: add controller checkpoint
      hpsa: change timeout for internal cmds
      hpsa: correct smart path enabled
      hpsa: update queue depth for externals
      hpsa: reduce warning messages on device removal
      hpsa: correct logical volume removal
      hpsa: add enclosure logical identifier
      hpsa: bump driver version

Martin Wilck (2):
      hpsa: destroy sas transport properties before scsi_host
      hpsa: cleanup sas_phy structures in sysfs when unloading

Scott Teel (1):
      hpsa: clear tmpdevice in scan thread


 drivers/scsi/hpsa.c     |  356 +++++++++++++++++++++++++++++++++++++----------
 drivers/scsi/hpsa_cmd.h |    3 
 2 files changed, 284 insertions(+), 75 deletions(-)

--
Signature

[PATCH 01/12] hpsa: destroy sas transport properties before scsi_host

[Don Brace]

From: Martin Wilck <mwilck@suse.de>

This patch cleans up a lot of warnings when unloading the
driver.

A current example of the stack trace starts with:
    [  142.570715] sysfs group 'power' not found for kobject 'port-5:0'
There can be hundreds of these messages during a driver unload.

I am resubmitting this patch on behalf of Martin Wilck with his
permission.

His original patch can be found here:
https://www.spinics.net/lists/linux-scsi/msg102085.html

This patch did not help until Hannes's
commit 9441284fbc39 ("scsi-fixup-kernel-warning-during-rmmod")
was applied to the kernel.

---------------------------
Original patch description:
---------------------------

Unloading the hpsa driver causes warnings

[ 1063.793652] WARNING: CPU: 1 PID: 4850 at ../fs/sysfs/group.c:237 device_del+0x54/0x240()
[ 1063.793659] sysfs group ffffffff81cf21a0 not found for kobject 'port-2:0'

with two different stacks:
1)
[ 1063.793774]  [<ffffffff81448af4>] device_del+0x54/0x240
[ 1063.793780]  [<ffffffff8145178a>] transport_remove_classdev+0x4a/0x60
[ 1063.793784]  [<ffffffff81451216>] attribute_container_device_trigger+0xa6/0xb0
[ 1063.793802]  [<ffffffffa0105d46>] sas_port_delete+0x126/0x160 [scsi_transport_sas]
[ 1063.793819]  [<ffffffffa036ebcc>] hpsa_free_sas_port+0x3c/0x70 [hpsa]

2)
[ 1063.797103]  [<ffffffff81448af4>] device_del+0x54/0x240
[ 1063.797118]  [<ffffffffa0105d4e>] sas_port_delete+0x12e/0x160 [scsi_transport_sas]
[ 1063.797134]  [<ffffffffa036ebcc>] hpsa_free_sas_port+0x3c/0x70 [hpsa]

This is caused by the fact that host device hostX is deleted before the
SAS transport devices hostX/port-a:b.

This patch fixes this by reverting the order of device deletions.

Tested-by: Don Brace <don.brace@microsemi.com>
Reviewed-by: Don Brace <don.brace@microsemi.com>
Signed-off-by: Martin Wilck <mwilck@xxxxxxx>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 9abe810..aff4a4f 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -8684,6 +8684,8 @@ static void hpsa_remove_one(struct pci_dev *pdev)
 	destroy_workqueue(h->rescan_ctlr_wq);
 	destroy_workqueue(h->resubmit_wq);
 
+	hpsa_delete_sas_host(h);
+
 	/*
 	 * Call before disabling interrupts.
 	 * scsi_remove_host can trigger I/O operations especially
@@ -8718,8 +8720,6 @@ static void hpsa_remove_one(struct pci_dev *pdev)
 	h->lockup_detected = NULL;			/* init_one 2 */
 	/* (void) pci_disable_pcie_error_reporting(pdev); */	/* init_one 1 */
 
-	hpsa_delete_sas_host(h);
-
 	kfree(h);					/* init_one 1 */
 }
 

[PATCH 02/12] hpsa: cleanup sas_phy structures in sysfs when unloading

[Don Brace]

From: Martin Wilck <mwilck@suse.de>

I am resubmitting this patch on behalf of Martin Wilck
with his permission.

The original patch can be found here:
https://www.spinics.net/lists/linux-scsi/msg102083.html

This patch did not help until Hannes's
commit 9441284fbc39 ("scsi-fixup-kernel-warning-during-rmmod")
was applied to the kernel.

--------------------------------------
Original patch description from Martin:
--------------------------------------

When the hpsa module is unloaded using rmmod, dangling
symlinks remain under /sys/class/sas_phy. Fix this by
calling sas_phy_delete() rather than sas_phy_free (which,
according to comments, should not be called for PHYs that
have been set up successfully, anyway).

Tested-by: Don Brace <don.brace@microsemi.com>
Reviewed-by: Don Brace <don.brace@microsemi.com>
Signed-off-by: Martin Wilck <mwilck@xxxxxxx>
Signed-off-by: Don Brace <don.brace@microsemi.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 aff4a4f..76461c4 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -9207,9 +9207,9 @@ static void hpsa_free_sas_phy(struct hpsa_sas_phy *hpsa_sas_phy)
 	struct sas_phy *phy = hpsa_sas_phy->phy;
 
 	sas_port_delete_phy(hpsa_sas_phy->parent_port->port, phy);
-	sas_phy_free(phy);
 	if (hpsa_sas_phy->added_to_port)
 		list_del(&hpsa_sas_phy->phy_list_entry);
+	sas_phy_delete(phy);
 	kfree(hpsa_sas_phy);
 }
 

[PATCH 03/12] hpsa: clear tmpdevice in scan thread

[Don Brace]

From: Scott Teel <scott.teel@microsemi.com>

clean up stale information.

Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c |    2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 76461c4..0330f17 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -4258,6 +4258,8 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h)
 		int phys_dev_index = i - (raid_ctlr_position == 0);
 		bool skip_device = false;
 
+		memset(tmpdevice, 0, sizeof(*tmpdevice));
+
 		physical_device = i < nphysicals + (raid_ctlr_position == 0);
 
 		/* Figure out where the LUN ID info is coming from */

[PATCH 04/12] hpsa: add controller checkpoint

[Don Brace]

tell hpsa controller to generate a checkpoint for rare lockup
conditions.

Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c     |    4 ++++
 drivers/scsi/hpsa_cmd.h |    1 +
 2 files changed, 5 insertions(+)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 0330f17..9c491c5 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -8009,6 +8009,10 @@ static void controller_lockup_detected(struct ctlr_info *h)
 	spin_unlock_irqrestore(&h->lock, flags);
 	dev_warn(&h->pdev->dev, "Controller lockup detected: 0x%08x after %d\n",
 			lockup_detected, h->heartbeat_sample_interval / HZ);
+	if (lockup_detected == 0xffff0000) {
+		dev_warn(&h->pdev->dev, "Telling controller to do a CHKPT\n");
+		writel(DOORBELL_GENERATE_CHKPT, h->vaddr + SA5_DOORBELL);
+	}
 	pci_disable_device(h->pdev);
 	fail_all_outstanding_cmds(h);
 }
diff --git a/drivers/scsi/hpsa_cmd.h b/drivers/scsi/hpsa_cmd.h
index 078afe4..78c3b64 100644
--- a/drivers/scsi/hpsa_cmd.h
+++ b/drivers/scsi/hpsa_cmd.h
@@ -142,6 +142,7 @@
 #define DOORBELL_CTLR_RESET	0x00000004l
 #define DOORBELL_CTLR_RESET2	0x00000020l
 #define DOORBELL_CLEAR_EVENTS	0x00000040l
+#define DOORBELL_GENERATE_CHKPT	0x00000080l
 
 #define CFGTBL_Trans_Simple     0x00000002l
 #define CFGTBL_Trans_Performant 0x00000004l

[PATCH 05/12] hpsa: update discovery polling

[Don Brace]

From: Bader Ali Saleh <bader.alisaleh@microsemi.com>

Correct a corner case where newly created volumes are not
detected automatically on an external RAID controller that has
no configured volumes during initial device discovery.

The fix is to set the discovery_polling flag when an external
RAID controller is detected. This causes a device rescan
every 20-30 seconds, so that newly created volumes will
be detected automatically.

Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c     |   39 ++++++++++++++++++++++++++-------------
 drivers/scsi/hpsa_cmd.h |    2 ++
 2 files changed, 28 insertions(+), 13 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 9c491c5..6056bbd 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -3472,6 +3472,30 @@ static void hpsa_get_sas_address(struct ctlr_info *h, unsigned char *scsi3addr,
 	dev->sas_address = sa;
 }
 
+static void hpsa_ext_ctrl_present(struct ctlr_info *h,
+	struct ReportExtendedLUNdata *physdev)
+{
+	u32 nphysicals;
+	int i;
+
+	if (h->discovery_polling)
+		return;
+
+	nphysicals = (get_unaligned_be32(physdev->LUNListLength) / 24) + 1;
+
+	for (i = 0; i < nphysicals; i++) {
+		if (physdev->LUN[i].device_type ==
+			BMIC_DEVICE_TYPE_CONTROLLER
+			&& !is_hba_lunid(physdev->LUN[i].lunid)) {
+			dev_info(&h->pdev->dev,
+				"External controller present, activate discovery polling and disable rld caching\n");
+			hpsa_disable_rld_caching(h);
+			h->discovery_polling = 1;
+			break;
+		}
+	}
+}
+
 /* Get a device id from inquiry page 0x83 */
 static bool hpsa_vpd_page_supported(struct ctlr_info *h,
 	unsigned char scsi3addr[], u8 page)
@@ -4228,6 +4252,8 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h)
 	 */
 	ndevs_to_allocate = nphysicals + nlogicals + MAX_EXT_TARGETS + 1;
 
+	hpsa_ext_ctrl_present(h, physdev_list);
+
 	/* Allocate the per device structures */
 	for (i = 0; i < ndevs_to_allocate; i++) {
 		if (i >= HPSA_MAX_DEVICES) {
@@ -4298,18 +4324,6 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h)
 		figure_bus_target_lun(h, lunaddrbytes, tmpdevice);
 		this_device = currentsd[ncurrent];
 
-		/* Turn on discovery_polling if there are ext target devices.
-		 * Event-based change notification is unreliable for those.
-		 */
-		if (!h->discovery_polling) {
-			if (tmpdevice->external) {
-				h->discovery_polling = 1;
-				dev_info(&h->pdev->dev,
-					"External target, activate discovery polling.\n");
-			}
-		}
-
-
 		*this_device = *tmpdevice;
 		this_device->physical_device = physical_device;
 
@@ -8247,7 +8261,6 @@ static void hpsa_rescan_ctlr_worker(struct work_struct *work)
 	if (h->drv_req_rescan || hpsa_offline_devices_ready(h)) {
 		hpsa_perform_rescan(h);
 	} else if (h->discovery_polling) {
-		hpsa_disable_rld_caching(h);
 		if (hpsa_luns_changed(h)) {
 			dev_info(&h->pdev->dev,
 				"driver discovery polling rescan.\n");
diff --git a/drivers/scsi/hpsa_cmd.h b/drivers/scsi/hpsa_cmd.h
index 78c3b64..21a726e 100644
--- a/drivers/scsi/hpsa_cmd.h
+++ b/drivers/scsi/hpsa_cmd.h
@@ -780,6 +780,8 @@ struct bmic_identify_physical_device {
 	u8     phys_bay_in_box;  /* phys drv bay this drive resides */
 	__le32 rpm;              /* Drive rotational speed in rpm */
 	u8     device_type;       /* type of drive */
+#define BMIC_DEVICE_TYPE_CONTROLLER	0x07
+
 	u8     sata_version;     /* only valid when drive_type is SATA */
 	__le64 big_total_block_count;
 	__le64 ris_starting_lba;

[PATCH 06/12] hpsa: change timeout for internal cmds

[Don Brace]

There are times when the DEFAULT_TIMEOUT (30 seconds)
is not enough.

Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c |   22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 6056bbd..4ab53df 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -2929,7 +2929,7 @@ static int hpsa_scsi_do_inquiry(struct ctlr_info *h, unsigned char *scsi3addr,
 		goto out;
 	}
 	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c,
-					PCI_DMA_FROMDEVICE, DEFAULT_TIMEOUT);
+					PCI_DMA_FROMDEVICE, NO_TIMEOUT);
 	if (rc)
 		goto out;
 	ei = c->err_info;
@@ -3213,7 +3213,7 @@ static int hpsa_get_raid_map(struct ctlr_info *h,
 		return -1;
 	}
 	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c,
-					PCI_DMA_FROMDEVICE, DEFAULT_TIMEOUT);
+					PCI_DMA_FROMDEVICE, NO_TIMEOUT);
 	if (rc)
 		goto out;
 	ei = c->err_info;
@@ -3256,7 +3256,7 @@ static int hpsa_bmic_sense_subsystem_information(struct ctlr_info *h,
 	c->Request.CDB[9] = (bmic_device_index >> 8) & 0xff;
 
 	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c,
-				PCI_DMA_FROMDEVICE, DEFAULT_TIMEOUT);
+				PCI_DMA_FROMDEVICE, NO_TIMEOUT);
 	if (rc)
 		goto out;
 	ei = c->err_info;
@@ -3284,7 +3284,7 @@ static int hpsa_bmic_id_controller(struct ctlr_info *h,
 		goto out;
 
 	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c,
-		PCI_DMA_FROMDEVICE, DEFAULT_TIMEOUT);
+		PCI_DMA_FROMDEVICE, NO_TIMEOUT);
 	if (rc)
 		goto out;
 	ei = c->err_info;
@@ -3315,7 +3315,7 @@ static int hpsa_bmic_id_physical_device(struct ctlr_info *h,
 	c->Request.CDB[9] = (bmic_device_index >> 8) & 0xff;
 
 	hpsa_scsi_do_simple_cmd_with_retry(h, c, PCI_DMA_FROMDEVICE,
-						DEFAULT_TIMEOUT);
+						NO_TIMEOUT);
 	ei = c->err_info;
 	if (ei->CommandStatus != 0 && ei->CommandStatus != CMD_DATA_UNDERRUN) {
 		hpsa_scsi_interpret_error(h, c);
@@ -3388,7 +3388,7 @@ static void hpsa_get_enclosure_info(struct ctlr_info *h,
 		c->Request.CDB[5] = 0;
 
 	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c, PCI_DMA_FROMDEVICE,
-						DEFAULT_TIMEOUT);
+						NO_TIMEOUT);
 	if (rc)
 		goto out;
 
@@ -3628,7 +3628,7 @@ static int hpsa_scsi_do_report_luns(struct ctlr_info *h, int logical,
 	if (extended_response)
 		c->Request.CDB[1] = extended_response;
 	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c,
-					PCI_DMA_FROMDEVICE, DEFAULT_TIMEOUT);
+					PCI_DMA_FROMDEVICE, NO_TIMEOUT);
 	if (rc)
 		goto out;
 	ei = c->err_info;
@@ -3763,7 +3763,7 @@ static unsigned char hpsa_volume_offline(struct ctlr_info *h,
 
 	(void) fill_cmd(c, TEST_UNIT_READY, h, NULL, 0, 0, scsi3addr, TYPE_CMD);
 	rc = hpsa_scsi_do_simple_cmd(h, c, DEFAULT_REPLY_QUEUE,
-					DEFAULT_TIMEOUT);
+					NO_TIMEOUT);
 	if (rc) {
 		cmd_free(h, c);
 		return HPSA_VPD_LV_STATUS_UNSUPPORTED;
@@ -8620,7 +8620,7 @@ static void hpsa_disable_rld_caching(struct ctlr_info *h)
 		goto errout;
 
 	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c,
-		PCI_DMA_FROMDEVICE, DEFAULT_TIMEOUT);
+		PCI_DMA_FROMDEVICE, NO_TIMEOUT);
 	if ((rc != 0) || (c->err_info->CommandStatus != 0))
 		goto errout;
 
@@ -8632,7 +8632,7 @@ static void hpsa_disable_rld_caching(struct ctlr_info *h)
 		goto errout;
 
 	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c,
-		PCI_DMA_TODEVICE, DEFAULT_TIMEOUT);
+		PCI_DMA_TODEVICE, NO_TIMEOUT);
 	if ((rc != 0)  || (c->err_info->CommandStatus != 0))
 		goto errout;
 
@@ -8642,7 +8642,7 @@ static void hpsa_disable_rld_caching(struct ctlr_info *h)
 		goto errout;
 
 	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c,
-		PCI_DMA_FROMDEVICE, DEFAULT_TIMEOUT);
+		PCI_DMA_FROMDEVICE, NO_TIMEOUT);
 	if ((rc != 0)  || (c->err_info->CommandStatus != 0))
 		goto errout;
 

[PATCH 07/12] hpsa: correct smart path enabled

[Don Brace]

- correct re-enabling ioaccel after:
  1) RAID transformations and
  2) multi-path fail-overs.

Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c |  189 +++++++++++++++++++++++++++++++++++++++++----------
 1 file changed, 151 insertions(+), 38 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 4ab53df..6a998b0 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -787,7 +787,12 @@ static ssize_t host_show_hp_ssd_smart_path_enabled(struct device *dev,
 	}
 	offload_enabled = hdev->offload_enabled;
 	spin_unlock_irqrestore(&h->lock, flags);
-	return snprintf(buf, 20, "%d\n", offload_enabled);
+
+	if (hdev->devtype == TYPE_DISK || hdev->devtype == TYPE_ZBC)
+		return snprintf(buf, 20, "%d\n", offload_enabled);
+	else
+		return snprintf(buf, 40, "%s\n",
+				"Not applicable for a controller");
 }
 
 #define MAX_PATHS 8
@@ -1270,7 +1275,7 @@ static void hpsa_show_dev_msg(const char *level, struct ctlr_info *h,
 			dev->model,
 			label,
 			dev->offload_config ? '+' : '-',
-			dev->offload_enabled ? '+' : '-',
+			dev->offload_to_be_enabled ? '+' : '-',
 			dev->expose_device);
 }
 
@@ -1345,36 +1350,42 @@ static int hpsa_scsi_add_entry(struct ctlr_info *h,
 	(*nadded)++;
 	hpsa_show_dev_msg(KERN_INFO, h, device,
 		device->expose_device ? "added" : "masked");
-	device->offload_to_be_enabled = device->offload_enabled;
-	device->offload_enabled = 0;
 	return 0;
 }
 
-/* Update an entry in h->dev[] array. */
+/*
+ * Called during a scan operation.
+ *
+ * Update an entry in h->dev[] array.
+ */
 static void hpsa_scsi_update_entry(struct ctlr_info *h,
 	int entry, struct hpsa_scsi_dev_t *new_entry)
 {
-	int offload_enabled;
 	/* assumes h->devlock is held */
 	BUG_ON(entry < 0 || entry >= HPSA_MAX_DEVICES);
 
 	/* Raid level changed. */
 	h->dev[entry]->raid_level = new_entry->raid_level;
 
+	/*
+	 * ioacccel_handle may have changed for a dual domain disk
+	 */
+	h->dev[entry]->ioaccel_handle = new_entry->ioaccel_handle;
+
 	/* Raid offload parameters changed.  Careful about the ordering. */
-	if (new_entry->offload_config && new_entry->offload_enabled) {
+	if (new_entry->offload_config && new_entry->offload_to_be_enabled) {
 		/*
 		 * if drive is newly offload_enabled, we want to copy the
 		 * raid map data first.  If previously offload_enabled and
 		 * offload_config were set, raid map data had better be
-		 * the same as it was before.  if raid map data is changed
+		 * the same as it was before. If raid map data has changed
 		 * then it had better be the case that
 		 * h->dev[entry]->offload_enabled is currently 0.
 		 */
 		h->dev[entry]->raid_map = new_entry->raid_map;
 		h->dev[entry]->ioaccel_handle = new_entry->ioaccel_handle;
 	}
-	if (new_entry->hba_ioaccel_enabled) {
+	if (new_entry->offload_to_be_enabled) {
 		h->dev[entry]->ioaccel_handle = new_entry->ioaccel_handle;
 		wmb(); /* set ioaccel_handle *before* hba_ioaccel_enabled */
 	}
@@ -1385,17 +1396,18 @@ static void hpsa_scsi_update_entry(struct ctlr_info *h,
 
 	/*
 	 * We can turn off ioaccel offload now, but need to delay turning
-	 * it on until we can update h->dev[entry]->phys_disk[], but we
+	 * ioaccel on until we can update h->dev[entry]->phys_disk[], but we
 	 * can't do that until all the devices are updated.
 	 */
-	h->dev[entry]->offload_to_be_enabled = new_entry->offload_enabled;
-	if (!new_entry->offload_enabled)
+	h->dev[entry]->offload_to_be_enabled = new_entry->offload_to_be_enabled;
+
+	/*
+	 * turn ioaccel off immediately if told to do so.
+	 */
+	if (!new_entry->offload_to_be_enabled)
 		h->dev[entry]->offload_enabled = 0;
 
-	offload_enabled = h->dev[entry]->offload_enabled;
-	h->dev[entry]->offload_enabled = h->dev[entry]->offload_to_be_enabled;
 	hpsa_show_dev_msg(KERN_INFO, h, h->dev[entry], "updated");
-	h->dev[entry]->offload_enabled = offload_enabled;
 }
 
 /* Replace an entry from h->dev[] array. */
@@ -1421,9 +1433,8 @@ static void hpsa_scsi_replace_entry(struct ctlr_info *h,
 	h->dev[entry] = new_entry;
 	added[*nadded] = new_entry;
 	(*nadded)++;
+
 	hpsa_show_dev_msg(KERN_INFO, h, new_entry, "replaced");
-	new_entry->offload_to_be_enabled = new_entry->offload_enabled;
-	new_entry->offload_enabled = 0;
 }
 
 /* Remove an entry from h->dev[] array. */
@@ -1513,11 +1524,22 @@ static inline int device_updated(struct hpsa_scsi_dev_t *dev1,
 		return 1;
 	if (dev1->offload_config != dev2->offload_config)
 		return 1;
-	if (dev1->offload_enabled != dev2->offload_enabled)
+	if (dev1->offload_to_be_enabled != dev2->offload_to_be_enabled)
 		return 1;
 	if (!is_logical_dev_addr_mode(dev1->scsi3addr))
 		if (dev1->queue_depth != dev2->queue_depth)
 			return 1;
+	/*
+	 * This can happen for dual domain devices. An active
+	 * path change causes the ioaccel handle to change
+	 *
+	 * for example note the handle differences between p0 and p1
+	 * Device                    WWN               ,WWN hash,Handle
+	 * D016 p0|0x3 [02]P2E:01:01,0x5000C5005FC4DACA,0x9B5616,0x01030003
+	 *	p1                   0x5000C5005FC4DAC9,0x6798C0,0x00040004
+	 */
+	if (dev1->ioaccel_handle != dev2->ioaccel_handle)
+		return 1;
 	return 0;
 }
 
@@ -1727,6 +1749,11 @@ static void hpsa_figure_phys_disk_ptrs(struct ctlr_info *h,
 		 * be 0, but we'll turn it off here just in case
 		 */
 		if (!logical_drive->phys_disk[i]) {
+			dev_warn(&h->pdev->dev,
+				"%s: [%d:%d:%d:%d] A phys disk component of LV is missing, turning off offload_enabled for LV.\n",
+				__func__,
+				h->scsi_host->host_no, logical_drive->bus,
+				logical_drive->target, logical_drive->lun);
 			logical_drive->offload_enabled = 0;
 			logical_drive->offload_to_be_enabled = 0;
 			logical_drive->queue_depth = 8;
@@ -1759,13 +1786,24 @@ static void hpsa_update_log_drive_phys_drive_ptrs(struct ctlr_info *h,
 		/*
 		 * If offload is currently enabled, the RAID map and
 		 * phys_disk[] assignment *better* not be changing
-		 * and since it isn't changing, we do not need to
-		 * update it.
+		 * because we would be changing ioaccel phsy_disk[] pointers
+		 * on a ioaccel volume processing I/O requests.
+		 *
+		 * If an ioaccel volume status changed, initially because it was
+		 * re-configured and thus underwent a transformation, or
+		 * a drive failed, we would have received a state change
+		 * request and ioaccel should have been turned off. When the
+		 * transformation completes, we get another state change
+		 * request to turn ioaccel back on. In this case, we need
+		 * to update the ioaccel information.
+		 *
+		 * Thus: If it is not currently enabled, but will be after
+		 * the scan completes, make sure the ioaccel pointers
+		 * are up to date.
 		 */
-		if (dev[i]->offload_enabled)
-			continue;
 
-		hpsa_figure_phys_disk_ptrs(h, dev, ndevices, dev[i]);
+		if (!dev[i]->offload_enabled && dev[i]->offload_to_be_enabled)
+			hpsa_figure_phys_disk_ptrs(h, dev, ndevices, dev[i]);
 	}
 }
 
@@ -1965,8 +2003,13 @@ static void adjust_hpsa_scsi_table(struct ctlr_info *h,
 	}
 	hpsa_update_log_drive_phys_drive_ptrs(h, h->dev, h->ndevices);
 
-	/* Now that h->dev[]->phys_disk[] is coherent, we can enable
+	/*
+	 * Now that h->dev[]->phys_disk[] is coherent, we can enable
 	 * any logical drives that need it enabled.
+	 *
+	 * The raid map should be current by now.
+	 *
+	 * We are updating the device list used for I/O requests.
 	 */
 	for (i = 0; i < h->ndevices; i++) {
 		if (h->dev[i] == NULL)
@@ -2441,7 +2484,7 @@ static void process_ioaccel2_completion(struct ctlr_info *h,
 
 	/*
 	 * Any RAID offload error results in retry which will use
-	 * the normal I/O path so the controller can handle whatever's
+	 * the normal I/O path so the controller can handle whatever is
 	 * wrong.
 	 */
 	if (is_logical_device(dev) &&
@@ -3540,6 +3583,13 @@ static bool hpsa_vpd_page_supported(struct ctlr_info *h,
 	return true;
 }
 
+/*
+ * Called during a scan operation.
+ * Sets ioaccel status on the new device list, not the existing device list
+ *
+ * The device list used during I/O will be updated later in
+ * adjust_hpsa_scsi_table.
+ */
 static void hpsa_get_ioaccel_status(struct ctlr_info *h,
 	unsigned char *scsi3addr, struct hpsa_scsi_dev_t *this_device)
 {
@@ -3568,12 +3618,12 @@ static void hpsa_get_ioaccel_status(struct ctlr_info *h,
 	this_device->offload_config =
 		!!(ioaccel_status & OFFLOAD_CONFIGURED_BIT);
 	if (this_device->offload_config) {
-		this_device->offload_enabled =
+		this_device->offload_to_be_enabled =
 			!!(ioaccel_status & OFFLOAD_ENABLED_BIT);
 		if (hpsa_get_raid_map(h, scsi3addr, this_device))
-			this_device->offload_enabled = 0;
+			this_device->offload_to_be_enabled = 0;
 	}
-	this_device->offload_to_be_enabled = this_device->offload_enabled;
+
 out:
 	kfree(buf);
 	return;
@@ -4307,7 +4357,7 @@ static void hpsa_update_scsi_devices(struct ctlr_info *h)
 				continue;
 		}
 
-		/* Get device type, vendor, model, device id */
+		/* Get device type, vendor, model, device id, raid_map */
 		rc = hpsa_update_device_info(h, lunaddrbytes, tmpdevice,
 							&is_OBDR);
 		if (rc == -ENOMEM) {
@@ -8067,9 +8117,79 @@ static int detect_controller_lockup(struct ctlr_info *h)
 	return false;
 }
 
-static void hpsa_ack_ctlr_events(struct ctlr_info *h)
+/*
+ * Set ioaccel status for all ioaccel volumes.
+ *
+ * Called from monitor controller worker (hpsa_event_monitor_worker)
+ *
+ * A Volume (or Volumes that comprise an Array set may be undergoing a
+ * transformation, so we will be turning off ioaccel for all volumes that
+ * make up the Array.
+ */
+static void hpsa_set_ioaccel_status(struct ctlr_info *h)
 {
+	int rc;
 	int i;
+	u8 ioaccel_status;
+	unsigned char *buf;
+	struct hpsa_scsi_dev_t *device;
+
+	if (!h)
+		return;
+
+	buf = kmalloc(64, GFP_KERNEL);
+	if (!buf)
+		return;
+
+	/*
+	 * Run through current device list used during I/O requests.
+	 */
+	for (i = 0; i < h->ndevices; i++) {
+		device = h->dev[i];
+
+		if (!device)
+			continue;
+		if (!device->scsi3addr)
+			continue;
+		if (!hpsa_vpd_page_supported(h, device->scsi3addr,
+						HPSA_VPD_LV_IOACCEL_STATUS))
+			continue;
+
+		memset(buf, 0, 64);
+
+		rc = hpsa_scsi_do_inquiry(h, device->scsi3addr,
+					VPD_PAGE | HPSA_VPD_LV_IOACCEL_STATUS,
+					buf, 64);
+		if (rc != 0)
+			continue;
+
+		ioaccel_status = buf[IOACCEL_STATUS_BYTE];
+		device->offload_config =
+				!!(ioaccel_status & OFFLOAD_CONFIGURED_BIT);
+		if (device->offload_config)
+			device->offload_to_be_enabled =
+				!!(ioaccel_status & OFFLOAD_ENABLED_BIT);
+
+		/*
+		 * Immediately turn off ioaccel for any volume the
+		 * controller tells us to. Some of the reasons could be:
+		 *    transformation - change to the LVs of an Array.
+		 *    degraded volume - component failure
+		 *
+		 * If ioaccel is to be re-enabled, re-enable later during the
+		 * scan operation so the driver can get a fresh raidmap
+		 * before turning ioaccel back on.
+		 *
+		 */
+		if (!device->offload_to_be_enabled)
+			device->offload_enabled = 0;
+	}
+
+	kfree(buf);
+}
+
+static void hpsa_ack_ctlr_events(struct ctlr_info *h)
+{
 	char *event_type;
 
 	if (!(h->fw_support & MISC_FW_EVENT_NOTIFY))
@@ -8087,10 +8207,7 @@ static void hpsa_ack_ctlr_events(struct ctlr_info *h)
 			event_type = "configuration change";
 		/* Stop sending new RAID offload reqs via the IO accelerator */
 		scsi_block_requests(h->scsi_host);
-		for (i = 0; i < h->ndevices; i++) {
-			h->dev[i]->offload_enabled = 0;
-			h->dev[i]->offload_to_be_enabled = 0;
-		}
+		hpsa_set_ioaccel_status(h);
 		hpsa_drain_accel_commands(h);
 		/* Set 'accelerator path config change' bit */
 		dev_warn(&h->pdev->dev,
@@ -8107,10 +8224,6 @@ static void hpsa_ack_ctlr_events(struct ctlr_info *h)
 		writel(h->events, &(h->cfgtable->clear_event_notify));
 		writel(DOORBELL_CLEAR_EVENTS, h->vaddr + SA5_DOORBELL);
 		hpsa_wait_for_clear_event_notify_ack(h);
-#if 0
-		writel(CFGTBL_ChangeReq, h->vaddr + SA5_DOORBELL);
-		hpsa_wait_for_mode_change_ack(h);
-#endif
 	}
 	return;
 }

[PATCH 08/12] hpsa: update queue depth for externals

[Don Brace]

- preserve external device queue depth during a
  scan operation.

Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 6a998b0..27a1aca 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -1765,8 +1765,12 @@ static void hpsa_figure_phys_disk_ptrs(struct ctlr_info *h,
 		 * way too high for partial stripe writes
 		 */
 		logical_drive->queue_depth = qdepth;
-	else
-		logical_drive->queue_depth = h->nr_cmds;
+	else {
+		if (logical_drive->external)
+			logical_drive->queue_depth = EXTERNAL_QD;
+		else
+			logical_drive->queue_depth = h->nr_cmds;
+	}
 }
 
 static void hpsa_update_log_drive_phys_drive_ptrs(struct ctlr_info *h,

[PATCH 09/12] hpsa: reduce warning messages on device removal

[Don Brace]

Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 27a1aca..228d744 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -1865,11 +1865,13 @@ static void hpsa_wait_for_outstanding_commands_for_dev(struct ctlr_info *h,
 			break;
 		if (++waits > 20)
 			break;
+		msleep(1000);
+	}
+
+	if (waits > 20)
 		dev_warn(&h->pdev->dev,
 			"%s: removing device with %d outstanding commands!\n",
 			__func__, cmds);
-		msleep(1000);
-	}
 }
 
 static void hpsa_remove_device(struct ctlr_info *h,

[PATCH 10/12] hpsa: correct logical volume removal

[Don Brace]

Suggested-by: Martin Wilck <mwilck@suse.com>
Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c |    9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index 228d744..a536ed6 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -1882,6 +1882,12 @@ static void hpsa_remove_device(struct ctlr_info *h,
 	if (!h->scsi_host)
 		return;
 
+	/*
+	 * Allow for commands to drain
+	 */
+	device->removed = 1;
+	hpsa_wait_for_outstanding_commands_for_dev(h, device);
+
 	if (is_logical_device(device)) { /* RAID */
 		sdev = scsi_device_lookup(h->scsi_host, device->bus,
 						device->target, device->lun);
@@ -1899,9 +1905,6 @@ static void hpsa_remove_device(struct ctlr_info *h,
 		}
 	} else { /* HBA */
 
-		device->removed = 1;
-		hpsa_wait_for_outstanding_commands_for_dev(h, device);
-
 		hpsa_remove_sas_device(device);
 	}
 }

[PATCH 11/12] hpsa: add enclosure logical identifier

[Don Brace]

add support for enclosure logical identifier

Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Reviewed-by: Scott Teel <scott.teel@microsemi.com>
Reviewed-by: Kevin Barnett <kevin.barnett@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.com>
---
 drivers/scsi/hpsa.c |   69 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 67 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c
index a536ed6..5428a4c 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -2965,6 +2965,57 @@ static void hpsa_scsi_interpret_error(struct ctlr_info *h,
 	}
 }
 
+static int hpsa_do_receive_diagnostic(struct ctlr_info *h, u8 *scsi3addr,
+					u8 page, u8 *buf, size_t bufsize)
+{
+	int rc = IO_OK;
+	struct CommandList *c;
+	struct ErrorInfo *ei;
+
+	c = cmd_alloc(h);
+	if (fill_cmd(c, RECEIVE_DIAGNOSTIC, h, buf, bufsize,
+			page, scsi3addr, TYPE_CMD)) {
+		rc = -1;
+		goto out;
+	}
+	rc = hpsa_scsi_do_simple_cmd_with_retry(h, c,
+		PCI_DMA_FROMDEVICE, NO_TIMEOUT);
+	if (rc)
+		goto out;
+	ei = c->err_info;
+	if (ei->CommandStatus != 0 && ei->CommandStatus != CMD_DATA_UNDERRUN) {
+		hpsa_scsi_interpret_error(h, c);
+		rc = -1;
+	}
+out:
+	cmd_free(h, c);
+	return rc;
+}
+
+static u64 hpsa_get_enclosure_logical_identifier(struct ctlr_info *h,
+						u8 *scsi3addr)
+{
+	u8 *buf;
+	u64 sa = 0;
+	int rc = 0;
+
+	buf = kzalloc(1024, GFP_KERNEL);
+	if (!buf)
+		return 0;
+
+	rc = hpsa_do_receive_diagnostic(h, scsi3addr, RECEIVE_DIAGNOSTIC,
+					buf, 1024);
+
+	if (rc)
+		goto out;
+
+	sa = get_unaligned_be64(buf+12);
+
+out:
+	kfree(buf);
+	return sa;
+}
+
 static int hpsa_scsi_do_inquiry(struct ctlr_info *h, unsigned char *scsi3addr,
 			u16 page, unsigned char *buf,
 			unsigned char bufsize)
@@ -3400,6 +3451,9 @@ static void hpsa_get_enclosure_info(struct ctlr_info *h,
 
 	bmic_device_index = GET_BMIC_DRIVE_NUMBER(&rle->lunid[0]);
 
+	encl_dev->sas_address =
+		hpsa_get_enclosure_logical_identifier(h, scsi3addr);
+
 	if (encl_dev->target == -1 || encl_dev->lun == -1) {
 		rc = IO_OK;
 		goto out;
@@ -6571,6 +6625,17 @@ static int fill_cmd(struct CommandList *c, u8 cmd, struct ctlr_info *h,
 			c->Request.CDB[0] = HPSA_INQUIRY;
 			c->Request.CDB[4] = size & 0xFF;
 			break;
+		case RECEIVE_DIAGNOSTIC:
+			c->Request.CDBLen = 6;
+			c->Request.type_attr_dir =
+				TYPE_ATTR_DIR(cmd_type, ATTR_SIMPLE, XFER_READ);
+			c->Request.Timeout = 0;
+			c->Request.CDB[0] = cmd;
+			c->Request.CDB[1] = 1;
+			c->Request.CDB[2] = 1;
+			c->Request.CDB[3] = (size >> 8) & 0xFF;
+			c->Request.CDB[4] = size & 0xFF;
+			break;
 		case HPSA_REPORT_LOG:
 		case HPSA_REPORT_PHYS:
 			/* Talking to controller so It's a physical command
@@ -9508,7 +9573,7 @@ static int hpsa_add_sas_host(struct ctlr_info *h)
 	struct hpsa_sas_port *hpsa_sas_port;
 	struct hpsa_sas_phy *hpsa_sas_phy;
 
-	parent_dev = &h->scsi_host->shost_gendev;
+	parent_dev = &h->scsi_host->shost_dev;
 
 	hpsa_sas_node = hpsa_alloc_sas_node(parent_dev);
 	if (!hpsa_sas_node)
@@ -9599,7 +9664,7 @@ hpsa_sas_get_linkerrors(struct sas_phy *phy)
 static int
 hpsa_sas_get_enclosure_identifier(struct sas_rphy *rphy, u64 *identifier)
 {
-	*identifier = 0;
+	*identifier = rphy->identify.sas_address;
 	return 0;
 }
 

[PATCH 12/12] hpsa: bump driver version

[Don Brace]

Reviewed-by: Gerry Morong <gerry.morong@microsemi.com>
Reviewed-by: Scott Benesh <scott.benesh@microsemi.com>
Signed-off-by: Don Brace <don.brace@microsemi.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 5428a4c..e514489 100644
--- a/drivers/scsi/hpsa.c
+++ b/drivers/scsi/hpsa.c
@@ -60,7 +60,7 @@
  * HPSA_DRIVER_VERSION must be 3 byte values (0-255) separated by '.'
  * with an optional trailing '-' followed by a byte value (0-255).
  */
-#define HPSA_DRIVER_VERSION "3.4.20-0"
+#define HPSA_DRIVER_VERSION "3.4.20-125"
 #define DRIVER_NAME "HP HPSA Driver (v " HPSA_DRIVER_VERSION ")"
 #define HPSA "hpsa"
 

Re: [PATCH 00/12] hpsa updates

[Martin K. Petersen]

Don,

> hpsa: update discovery polling
>  - Correct corner case for RBODs with no initial volumes.

[...]

Series applied to 4.15/scsi-queue. Thanks!

-- 
Martin K. Petersen	Oracle Linux Engineering

Re: [PATCH 02/12] hpsa: cleanup sas_phy structures in sysfs when unloading

[James Bottomley]

On Fri, 2017-10-20 at 16:51 -0500, Don Brace wrote:
> From: Martin Wilck <mwilck@suse.de>
> 
> I am resubmitting this patch on behalf of Martin Wilck
> with his permission.
> 
> The original patch can be found here:
> https://www.spinics.net/lists/linux-scsi/msg102083.html
> 
> This patch did not help until Hannes's
> commit 9441284fbc39 ("scsi-fixup-kernel-warning-during-rmmod")
> was applied to the kernel.
> 
> --------------------------------------
> Original patch description from Martin:
> --------------------------------------
> 
> When the hpsa module is unloaded using rmmod, dangling
> symlinks remain under /sys/class/sas_phy. Fix this by
> calling sas_phy_delete() rather than sas_phy_free (which,
> according to comments, should not be called for PHYs that
> have been set up successfully, anyway).
> 
> Tested-by: Don Brace <don.brace@microsemi.com>
> Reviewed-by: Don Brace <don.brace@microsemi.com>
> Signed-off-by: Martin Wilck <mwilck@xxxxxxx>

Hey, what's this for a signoff?  The original patch had the correct
email address; how did it get mangled like this?

James

RE: [PATCH 02/12] hpsa: cleanup sas_phy structures in sysfs when unloading

[Don Brace]

> -----Original Message-----
> On Fri, 2017-10-20 at 16:51 -0500, Don Brace wrote:
> > From: Martin Wilck <mwilck@suse.de>
> >
> > I am resubmitting this patch on behalf of Martin Wilck
> > with his permission.
> >
> > The original patch can be found here:
> > https://www.spinics.net/lists/linux-scsi/msg102083.html
> >
> > This patch did not help until Hannes's
> > commit 9441284fbc39 ("scsi-fixup-kernel-warning-during-rmmod")
> > was applied to the kernel.
> >
> > --------------------------------------
> > Original patch description from Martin:
> > --------------------------------------
> >
> > When the hpsa module is unloaded using rmmod, dangling
> > symlinks remain under /sys/class/sas_phy. Fix this by
> > calling sas_phy_delete() rather than sas_phy_free (which,
> > according to comments, should not be called for PHYs that
> > have been set up successfully, anyway).
> >
> > Tested-by: Don Brace <don.brace@microsemi.com>
> > Reviewed-by: Don Brace <don.brace@microsemi.com>
> > Signed-off-by: Martin Wilck <mwilck@xxxxxxx>
> 
> Hey, what's this for a signoff?  The original patch had the correct
> email address; how did it get mangled like this?
> 
> James

Unsure. I do not see the mangled e-mail anywhere in the e-mail
chain. I did not type the e-mail in like that. But it originated from me.
I may have pulled the patch from an archive that mangled it.

Do you want me to re-post the two patches? 

Thanks,
Don Brace
ESC - Smart Storage
Microsemi Corporation

Re: [PATCH 02/12] hpsa: cleanup sas_phy structures in sysfs when unloading

[Martin K. Petersen]

Don,

> Unsure. I do not see the mangled e-mail anywhere in the e-mail
> chain. I did not type the e-mail in like that. But it originated from
> me.  I may have pulled the patch from an archive that mangled it.
>
> Do you want me to re-post the two patches?

I'll fix them up. But it's a bad time to have to rebase...

-- 
Martin K. Petersen	Oracle Linux Engineering

[PATCH 00/12] hpsa updates

[Don Brace]

These patches are based on Linus's tree

These patches are for:
 - Multipath failover support in general.

The changes are:
 - update identify physical device structure
   - align with FW
 - stop getting enclosure info for externals
   - no BMIC support
 - update reset handler
   - update to match out of box driver
 - do not reset enclosures
   - reset can sometimes hang
 - rescan later if reset in progress
   - wait for devices to settle.
 - correct resets on retried commands
   - was not calling scsi_done on retried completion
 - correct queue depth for externals
   - Code not in correct function
 - separate monitor events from heartbeat worker
   - allows driver to check for changes more frequently
     without affecting controller lockup detection.
 - send ioaccel requests with 0 length down raid path
   - avoid hang issues for customers running older FW.
 - remove abort handler
   - align driver with our out of box driver
 - bump driver version
   - align version with out of box driver for multi-path changes

---

Don Brace (11):
      hpsa: update identify physical device structure
      hpsa: do not get enclosure info for external devices
      hpsa: update reset handler
      hpsa: do not reset enclosures
      hpsa: rescan later if reset in progress
      hpsa: correct resets on retried commands
      hpsa: cleanup reset handler
      hpsa: correct queue depth for externals
      hpsa: send ioaccel requests with 0 length down raid path
      hpsa: remove abort handler
      hpsa: bump driver version

Scott Teel (1):
      hpsa: separate monitor events from heartbeat worker


 drivers/scsi/hpsa.c     |  790 +++++++++--------------------------------------
 drivers/scsi/hpsa.h     |    3 
 drivers/scsi/hpsa_cmd.h |   20 +
 3 files changed, 164 insertions(+), 649 deletions(-)

--
Signature

[PATCH 00/12] hpsa updates

[Don Brace]

These patches are based on Linus's tree

These patches are for: 
 - Multipath failover support in general.

The changes are:
 - update identify physical device structure
   - align with FW
 - stop getting enclosure info for externals
   - no BMIC support
 - update reset handler
   - update to match out of box driver
 - do not reset enclosures
   - reset can sometimes hang
 - rescan later if reset in progress
   - wait for devices to settle.
 - correct resets on retried commands
   - was not calling scsi_done on retried completion
 - correct queue depth for externals
   - Code not in correct function
 - separate monitor events from heartbeat worker
   - allows driver to check for changes more frequently
     without affecting controller lockup detection.
 - send ioaccel requests with 0 length down raid path
   - avoid hang issues for customers running older FW.
 - remove abort handler
   - align driver with our out of box driver
 - bump driver version
   - align version with out of box driver for multi-path changes

---

Don Brace (11):
      hpsa: update identify physical device structure
      hpsa: do not get enclosure info for external devices
      hpsa: update reset handler
      hpsa: do not reset enclosures
      hpsa: rescan later if reset in progress
      hpsa: correct resets on retried commands
      hpsa: cleanup reset handler
      hpsa: correct queue depth for externals
      hpsa: send ioaccel requests with 0 length down raid path
      hpsa: remove abort handler
      hpsa: bump driver version

Scott Teel (1):
      hpsa: separate monitor events from heartbeat worker


 drivers/scsi/hpsa.c     |  790 +++++++++--------------------------------------
 drivers/scsi/hpsa.h     |    3 
 drivers/scsi/hpsa_cmd.h |   20 +
 3 files changed, 164 insertions(+), 649 deletions(-)

--
Signature