[SCSI] esas2r: ATTO Technology ExpressSAS 6G SAS/SATA RAID Adapter Driver

This is a new driver for ATTO Technology's ExpressSAS series of hardware RAID
adapters.  It supports the following adapters:

    - ExpressSAS R60F
    - ExpressSAS R680
    - ExpressSAS R608
    - ExpressSAS R644

Signed-off-by: Bradley Grove <bgrove@attotech.com>
Signed-off-by: James Bottomley <JBottomley@Parallels.com>

1 files changed, 2032 insertions, 0 deletions

diff --git a/drivers/scsi/esas2r/esas2r_main.c b/drivers/scsi/esas2r/esas2r_main.c
new file mode 100644
index 000000000000..4abf1272e1eb
--- /dev/null
+++ b/drivers/scsi/esas2r/esas2r_main.c
@@ -0,0 +1,2032 @@
+/*
+ *  linux/drivers/scsi/esas2r/esas2r_main.c
+ *      For use with ATTO ExpressSAS R6xx SAS/SATA RAID controllers
+ *
+ *  Copyright (c) 2001-2013 ATTO Technology, Inc.
+ *  (mailto:linuxdrivers@attotech.com)
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version 2
+ * of the License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * NO WARRANTY
+ * THE PROGRAM IS PROVIDED ON AN "AS IS" BASIS, WITHOUT WARRANTIES OR
+ * CONDITIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED INCLUDING, WITHOUT
+ * LIMITATION, ANY WARRANTIES OR CONDITIONS OF TITLE, NON-INFRINGEMENT,
+ * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Each Recipient is
+ * solely responsible for determining the appropriateness of using and
+ * distributing the Program and assumes all risks associated with its
+ * exercise of rights under this Agreement, including but not limited to
+ * the risks and costs of program errors, damage to or loss of data,
+ * programs or equipment, and unavailability or interruption of operations.
+ *
+ * DISCLAIMER OF LIABILITY
+ * NEITHER RECIPIENT NOR ANY CONTRIBUTORS SHALL HAVE ANY LIABILITY FOR ANY
+ * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+ * DAMAGES (INCLUDING WITHOUT LIMITATION LOST PROFITS), HOWEVER CAUSED AND
+ * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR
+ * TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
+ * USE OR DISTRIBUTION OF THE PROGRAM OR THE EXERCISE OF ANY RIGHTS GRANTED
+ * HEREUNDER, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301,
+ * USA.
+ */
+
+#include "esas2r.h"
+
+MODULE_DESCRIPTION(ESAS2R_DRVR_NAME ": " ESAS2R_LONGNAME " driver");
+MODULE_AUTHOR("ATTO Technology, Inc.");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(ESAS2R_VERSION_STR);
+
+/* global definitions */
+
+static int found_adapters;
+struct esas2r_adapter *esas2r_adapters[MAX_ADAPTERS];
+
+#define ESAS2R_VDA_EVENT_PORT1       54414
+#define ESAS2R_VDA_EVENT_PORT2       54415
+#define ESAS2R_VDA_EVENT_SOCK_COUNT  2
+
+static struct esas2r_adapter *esas2r_adapter_from_kobj(struct kobject *kobj)
+{
+	struct device *dev = container_of(kobj, struct device, kobj);
+	struct Scsi_Host *host = class_to_shost(dev);
+
+	return (struct esas2r_adapter *)host->hostdata;
+}
+
+static ssize_t read_fw(struct file *file, struct kobject *kobj,
+		       struct bin_attribute *attr,
+		       char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+
+	return esas2r_read_fw(a, buf, off, count);
+}
+
+static ssize_t write_fw(struct file *file, struct kobject *kobj,
+			struct bin_attribute *attr,
+			char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+
+	return esas2r_write_fw(a, buf, off, count);
+}
+
+static ssize_t read_fs(struct file *file, struct kobject *kobj,
+		       struct bin_attribute *attr,
+		       char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+
+	return esas2r_read_fs(a, buf, off, count);
+}
+
+static ssize_t write_fs(struct file *file, struct kobject *kobj,
+			struct bin_attribute *attr,
+			char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+	int length = min(sizeof(struct esas2r_ioctl_fs), count);
+	int result = 0;
+
+	result = esas2r_write_fs(a, buf, off, count);
+
+	if (result < 0)
+		result = 0;
+
+	return length;
+}
+
+static ssize_t read_vda(struct file *file, struct kobject *kobj,
+			struct bin_attribute *attr,
+			char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+
+	return esas2r_read_vda(a, buf, off, count);
+}
+
+static ssize_t write_vda(struct file *file, struct kobject *kobj,
+			 struct bin_attribute *attr,
+			 char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+
+	return esas2r_write_vda(a, buf, off, count);
+}
+
+static ssize_t read_live_nvram(struct file *file, struct kobject *kobj,
+			       struct bin_attribute *attr,
+			       char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+	int length = min_t(size_t, sizeof(struct esas2r_sas_nvram), PAGE_SIZE);
+
+	memcpy(buf, a->nvram, length);
+	return length;
+}
+
+static ssize_t write_live_nvram(struct file *file, struct kobject *kobj,
+				struct bin_attribute *attr,
+				char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+	struct esas2r_request *rq;
+	int result = -EFAULT;
+
+	rq = esas2r_alloc_request(a);
+	if (rq == NULL)
+		return -ENOMEM;
+
+	if (esas2r_write_params(a, rq, (struct esas2r_sas_nvram *)buf))
+		result = count;
+
+	esas2r_free_request(a, rq);
+
+	return result;
+}
+
+static ssize_t read_default_nvram(struct file *file, struct kobject *kobj,
+				  struct bin_attribute *attr,
+				  char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+
+	esas2r_nvram_get_defaults(a, (struct esas2r_sas_nvram *)buf);
+
+	return sizeof(struct esas2r_sas_nvram);
+}
+
+static ssize_t read_hw(struct file *file, struct kobject *kobj,
+		       struct bin_attribute *attr,
+		       char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+	int length = min_t(size_t, sizeof(struct atto_ioctl), PAGE_SIZE);
+
+	if (!a->local_atto_ioctl)
+		return -ENOMEM;
+
+	if (handle_hba_ioctl(a, a->local_atto_ioctl) != IOCTL_SUCCESS)
+		return -ENOMEM;
+
+	memcpy(buf, a->local_atto_ioctl, length);
+
+	return length;
+}
+
+static ssize_t write_hw(struct file *file, struct kobject *kobj,
+			struct bin_attribute *attr,
+			char *buf, loff_t off, size_t count)
+{
+	struct esas2r_adapter *a = esas2r_adapter_from_kobj(kobj);
+	int length = min(sizeof(struct atto_ioctl), count);
+
+	if (!a->local_atto_ioctl) {
+		a->local_atto_ioctl = kzalloc(sizeof(struct atto_ioctl),
+					      GFP_KERNEL);
+		if (a->local_atto_ioctl == NULL) {
+			esas2r_log(ESAS2R_LOG_WARN,
+				   "write_hw kzalloc failed for %d bytes",
+				   sizeof(struct atto_ioctl));
+			return -ENOMEM;
+		}
+	}
+
+	memset(a->local_atto_ioctl, 0, sizeof(struct atto_ioctl));
+	memcpy(a->local_atto_ioctl, buf, length);
+
+	return length;
+}
+
+#define ESAS2R_RW_BIN_ATTR(_name) \
+	struct bin_attribute bin_attr_ ## _name = { \
+		.attr	= \
+		{ .name = __stringify(_name), .mode  = S_IRUSR | S_IWUSR }, \
+		.size	= 0, \
+		.read	= read_ ## _name, \
+		.write	= write_ ## _name }
+
+ESAS2R_RW_BIN_ATTR(fw);
+ESAS2R_RW_BIN_ATTR(fs);
+ESAS2R_RW_BIN_ATTR(vda);
+ESAS2R_RW_BIN_ATTR(hw);
+ESAS2R_RW_BIN_ATTR(live_nvram);
+
+struct bin_attribute bin_attr_default_nvram = {
+	.attr	= { .name = "default_nvram", .mode = S_IRUGO },
+	.size	= 0,
+	.read	= read_default_nvram,
+	.write	= NULL
+};
+
+static struct scsi_host_template driver_template = {
+	.module				= THIS_MODULE,
+	.show_info			= esas2r_show_info,
+	.name				= ESAS2R_LONGNAME,
+	.release			= esas2r_release,
+	.info				= esas2r_info,
+	.ioctl				= esas2r_ioctl,
+	.queuecommand			= esas2r_queuecommand,
+	.eh_abort_handler		= esas2r_eh_abort,
+	.eh_device_reset_handler	= esas2r_device_reset,
+	.eh_bus_reset_handler		= esas2r_bus_reset,
+	.eh_host_reset_handler		= esas2r_host_reset,
+	.eh_target_reset_handler	= esas2r_target_reset,
+	.can_queue			= 128,
+	.this_id			= -1,
+	.sg_tablesize			= SCSI_MAX_SG_SEGMENTS,
+	.cmd_per_lun			=
+		ESAS2R_DEFAULT_CMD_PER_LUN,
+	.present			= 0,
+	.unchecked_isa_dma		= 0,
+	.use_clustering			= ENABLE_CLUSTERING,
+	.emulated			= 0,
+	.proc_name			= ESAS2R_DRVR_NAME,
+	.slave_configure		= esas2r_slave_configure,
+	.slave_alloc			= esas2r_slave_alloc,
+	.slave_destroy			= esas2r_slave_destroy,
+	.change_queue_depth		= esas2r_change_queue_depth,
+	.change_queue_type		= esas2r_change_queue_type,
+	.max_sectors			= 0xFFFF,
+};
+
+int sgl_page_size = 512;
+module_param(sgl_page_size, int, 0);
+MODULE_PARM_DESC(sgl_page_size,
+		 "Scatter/gather list (SGL) page size in number of S/G "
+		 "entries.  If your application is doing a lot of very large "
+		 "transfers, you may want to increase the SGL page size.  "
+		 "Default 512.");
+
+int num_sg_lists = 1024;
+module_param(num_sg_lists, int, 0);
+MODULE_PARM_DESC(num_sg_lists,
+		 "Number of scatter/gather lists.  Default 1024.");
+
+int sg_tablesize = SCSI_MAX_SG_SEGMENTS;
+module_param(sg_tablesize, int, 0);
+MODULE_PARM_DESC(sg_tablesize,
+		 "Maximum number of entries in a scatter/gather table.");
+
+int num_requests = 256;
+module_param(num_requests, int, 0);
+MODULE_PARM_DESC(num_requests,
+		 "Number of requests.  Default 256.");
+
+int num_ae_requests = 4;
+module_param(num_ae_requests, int, 0);
+MODULE_PARM_DESC(num_ae_requests,
+		 "Number of VDA asynchromous event requests.  Default 4.");
+
+int cmd_per_lun = ESAS2R_DEFAULT_CMD_PER_LUN;
+module_param(cmd_per_lun, int, 0);
+MODULE_PARM_DESC(cmd_per_lun,
+		 "Maximum number of commands per LUN.  Default "
+		 DEFINED_NUM_TO_STR(ESAS2R_DEFAULT_CMD_PER_LUN) ".");
+
+int can_queue = 128;
+module_param(can_queue, int, 0);
+MODULE_PARM_DESC(can_queue,
+		 "Maximum number of commands per adapter.  Default 128.");
+
+int esas2r_max_sectors = 0xFFFF;
+module_param(esas2r_max_sectors, int, 0);
+MODULE_PARM_DESC(esas2r_max_sectors,
+		 "Maximum number of disk sectors in a single data transfer.  "
+		 "Default 65535 (largest possible setting).");
+
+int interrupt_mode = 1;
+module_param(interrupt_mode, int, 0);
+MODULE_PARM_DESC(interrupt_mode,
+		 "Defines the interrupt mode to use.  0 for legacy"
+		 ", 1 for MSI.  Default is MSI (1).");
+
+static struct pci_device_id
+	esas2r_pci_table[] = {
+	{ ATTO_VENDOR_ID, 0x0049,	  ATTO_VENDOR_ID, 0x0049,
+	  0,
+	  0, 0 },
+	{ ATTO_VENDOR_ID, 0x0049,	  ATTO_VENDOR_ID, 0x004A,
+	  0,
+	  0, 0 },
+	{ ATTO_VENDOR_ID, 0x0049,	  ATTO_VENDOR_ID, 0x004B,
+	  0,
+	  0, 0 },
+	{ ATTO_VENDOR_ID, 0x0049,	  ATTO_VENDOR_ID, 0x004C,
+	  0,
+	  0, 0 },
+	{ ATTO_VENDOR_ID, 0x0049,	  ATTO_VENDOR_ID, 0x004D,
+	  0,
+	  0, 0 },
+	{ ATTO_VENDOR_ID, 0x0049,	  ATTO_VENDOR_ID, 0x004E,
+	  0,
+	  0, 0 },
+	{ 0,		  0,		  0,		  0,
+	  0,
+	  0, 0 }
+};
+
+MODULE_DEVICE_TABLE(pci, esas2r_pci_table);
+
+static int
+esas2r_probe(struct pci_dev *pcid, const struct pci_device_id *id);
+
+static void
+esas2r_remove(struct pci_dev *pcid);
+
+static struct pci_driver
+	esas2r_pci_driver = {
+	.name		= ESAS2R_DRVR_NAME,
+	.id_table	= esas2r_pci_table,
+	.probe		= esas2r_probe,
+	.remove		= esas2r_remove,
+	.suspend	= esas2r_suspend,
+	.resume		= esas2r_resume,
+};
+
+static int esas2r_probe(struct pci_dev *pcid,
+			const struct pci_device_id *id)
+{
+	struct Scsi_Host *host = NULL;
+	struct esas2r_adapter *a;
+	int err;
+
+	size_t host_alloc_size = sizeof(struct esas2r_adapter)
+				 + ((num_requests) +
+				    1) * sizeof(struct esas2r_request);
+
+	esas2r_log_dev(ESAS2R_LOG_DEBG, &(pcid->dev),
+		       "esas2r_probe() 0x%02x 0x%02x 0x%02x 0x%02x",
+		       pcid->vendor,
+		       pcid->device,
+		       pcid->subsystem_vendor,
+		       pcid->subsystem_device);
+
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pcid->dev),
+		       "before pci_enable_device() "
+		       "enable_cnt: %d",
+		       pcid->enable_cnt.counter);
+
+	err = pci_enable_device(pcid);
+	if (err != 0) {
+		esas2r_log_dev(ESAS2R_LOG_CRIT, &(pcid->dev),
+			       "pci_enable_device() FAIL (%d)",
+			       err);
+		return -ENODEV;
+	}
+
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pcid->dev),
+		       "pci_enable_device() OK");
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pcid->dev),
+		       "after pci_device_enable() enable_cnt: %d",
+		       pcid->enable_cnt.counter);
+
+	host = scsi_host_alloc(&driver_template, host_alloc_size);
+	if (host == NULL) {
+		esas2r_log(ESAS2R_LOG_CRIT, "scsi_host_alloc() FAIL");
+		return -ENODEV;
+	}
+
+	memset(host->hostdata, 0, host_alloc_size);
+
+	a = (struct esas2r_adapter *)host->hostdata;
+
+	esas2r_log(ESAS2R_LOG_INFO, "scsi_host_alloc() OK host: %p", host);
+
+	/* override max LUN and max target id */
+
+	host->max_id = ESAS2R_MAX_ID + 1;
+	host->max_lun = 255;
+
+	/* we can handle 16-byte CDbs */
+
+	host->max_cmd_len = 16;
+
+	host->can_queue = can_queue;
+	host->cmd_per_lun = cmd_per_lun;
+	host->this_id = host->max_id + 1;
+	host->max_channel = 0;
+	host->unique_id = found_adapters;
+	host->sg_tablesize = sg_tablesize;
+	host->max_sectors = esas2r_max_sectors;
+
+	/* set to bus master for BIOses that don't do it for us */
+
+	esas2r_log(ESAS2R_LOG_INFO, "pci_set_master() called");
+
+	pci_set_master(pcid);
+
+	if (!esas2r_init_adapter(host, pcid, found_adapters)) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "unable to initialize device at PCI bus %x:%x",
+			   pcid->bus->number,
+			   pcid->devfn);
+
+		esas2r_log_dev(ESAS2R_LOG_INFO, &(host->shost_gendev),
+			       "scsi_host_put() called");
+
+		scsi_host_put(host);
+
+		return 0;
+
+	}
+
+	esas2r_log(ESAS2R_LOG_INFO, "pci_set_drvdata(%p, %p) called", pcid,
+		   host->hostdata);
+
+	pci_set_drvdata(pcid, host);
+
+	esas2r_log(ESAS2R_LOG_INFO, "scsi_add_host() called");
+
+	err = scsi_add_host(host, &pcid->dev);
+
+	if (err) {
+		esas2r_log(ESAS2R_LOG_CRIT, "scsi_add_host returned %d", err);
+		esas2r_log_dev(ESAS2R_LOG_CRIT, &(host->shost_gendev),
+			       "scsi_add_host() FAIL");
+
+		esas2r_log_dev(ESAS2R_LOG_INFO, &(host->shost_gendev),
+			       "scsi_host_put() called");
+
+		scsi_host_put(host);
+
+		esas2r_log_dev(ESAS2R_LOG_INFO, &(host->shost_gendev),
+			       "pci_set_drvdata(%p, NULL) called",
+			       pcid);
+
+		pci_set_drvdata(pcid, NULL);
+
+		return -ENODEV;
+	}
+
+
+	esas2r_fw_event_on(a);
+
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(host->shost_gendev),
+		       "scsi_scan_host() called");
+
+	scsi_scan_host(host);
+
+	/* Add sysfs binary files */
+	if (sysfs_create_bin_file(&host->shost_dev.kobj, &bin_attr_fw))
+		esas2r_log_dev(ESAS2R_LOG_WARN, &(host->shost_gendev),
+			       "Failed to create sysfs binary file: fw");
+	else
+		a->sysfs_fw_created = 1;
+
+	if (sysfs_create_bin_file(&host->shost_dev.kobj, &bin_attr_fs))
+		esas2r_log_dev(ESAS2R_LOG_WARN, &(host->shost_gendev),
+			       "Failed to create sysfs binary file: fs");
+	else
+		a->sysfs_fs_created = 1;
+
+	if (sysfs_create_bin_file(&host->shost_dev.kobj, &bin_attr_vda))
+		esas2r_log_dev(ESAS2R_LOG_WARN, &(host->shost_gendev),
+			       "Failed to create sysfs binary file: vda");
+	else
+		a->sysfs_vda_created = 1;
+
+	if (sysfs_create_bin_file(&host->shost_dev.kobj, &bin_attr_hw))
+		esas2r_log_dev(ESAS2R_LOG_WARN, &(host->shost_gendev),
+			       "Failed to create sysfs binary file: hw");
+	else
+		a->sysfs_hw_created = 1;
+
+	if (sysfs_create_bin_file(&host->shost_dev.kobj, &bin_attr_live_nvram))
+		esas2r_log_dev(ESAS2R_LOG_WARN, &(host->shost_gendev),
+			       "Failed to create sysfs binary file: live_nvram");
+	else
+		a->sysfs_live_nvram_created = 1;
+
+	if (sysfs_create_bin_file(&host->shost_dev.kobj,
+				  &bin_attr_default_nvram))
+		esas2r_log_dev(ESAS2R_LOG_WARN, &(host->shost_gendev),
+			       "Failed to create sysfs binary file: default_nvram");
+	else
+		a->sysfs_default_nvram_created = 1;
+
+	found_adapters++;
+
+	return 0;
+}
+
+static void esas2r_remove(struct pci_dev *pdev)
+{
+	struct Scsi_Host *host;
+	int index;
+
+	if (pdev == NULL) {
+		esas2r_log(ESAS2R_LOG_WARN, "esas2r_remove pdev==NULL");
+		return;
+	}
+
+	host = pci_get_drvdata(pdev);
+
+	if (host == NULL) {
+		/*
+		 * this can happen if pci_set_drvdata was already called
+		 * to clear the host pointer.  if this is the case, we
+		 * are okay; this channel has already been cleaned up.
+		 */
+
+		return;
+	}
+
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev),
+		       "esas2r_remove(%p) called; "
+		       "host:%p", pdev,
+		       host);
+
+	index = esas2r_cleanup(host);
+
+	if (index < 0)
+		esas2r_log_dev(ESAS2R_LOG_WARN, &(pdev->dev),
+			       "unknown host in %s",
+			       __func__);
+
+	found_adapters--;
+
+	/* if this was the last adapter, clean up the rest of the driver */
+
+	if (found_adapters == 0)
+		esas2r_cleanup(NULL);
+}
+
+static int __init esas2r_init(void)
+{
+	int i;
+
+	esas2r_log(ESAS2R_LOG_INFO, "%s called", __func__);
+
+	/* verify valid parameters */
+
+	if (can_queue < 1) {
+		esas2r_log(ESAS2R_LOG_WARN,
+			   "warning: can_queue must be at least 1, value "
+			   "forced.");
+		can_queue = 1;
+	} else if (can_queue > 2048) {
+		esas2r_log(ESAS2R_LOG_WARN,
+			   "warning: can_queue must be no larger than 2048, "
+			   "value forced.");
+		can_queue = 2048;
+	}
+
+	if (cmd_per_lun < 1) {
+		esas2r_log(ESAS2R_LOG_WARN,
+			   "warning: cmd_per_lun must be at least 1, value "
+			   "forced.");
+		cmd_per_lun = 1;
+	} else if (cmd_per_lun > 2048) {
+		esas2r_log(ESAS2R_LOG_WARN,
+			   "warning: cmd_per_lun must be no larger than "
+			   "2048, value forced.");
+		cmd_per_lun = 2048;
+	}
+
+	if (sg_tablesize < 32) {
+		esas2r_log(ESAS2R_LOG_WARN,
+			   "warning: sg_tablesize must be at least 32, "
+			   "value forced.");
+		sg_tablesize = 32;
+	}
+
+	if (esas2r_max_sectors < 1) {
+		esas2r_log(ESAS2R_LOG_WARN,
+			   "warning: esas2r_max_sectors must be at least "
+			   "1, value forced.");
+		esas2r_max_sectors = 1;
+	} else if (esas2r_max_sectors > 0xffff) {
+		esas2r_log(ESAS2R_LOG_WARN,
+			   "warning: esas2r_max_sectors must be no larger "
+			   "than 0xffff, value forced.");
+		esas2r_max_sectors = 0xffff;
+	}
+
+	sgl_page_size &= ~(ESAS2R_SGL_ALIGN - 1);
+
+	if (sgl_page_size < SGL_PG_SZ_MIN)
+		sgl_page_size = SGL_PG_SZ_MIN;
+	else if (sgl_page_size > SGL_PG_SZ_MAX)
+		sgl_page_size = SGL_PG_SZ_MAX;
+
+	if (num_sg_lists < NUM_SGL_MIN)
+		num_sg_lists = NUM_SGL_MIN;
+	else if (num_sg_lists > NUM_SGL_MAX)
+		num_sg_lists = NUM_SGL_MAX;
+
+	if (num_requests < NUM_REQ_MIN)
+		num_requests = NUM_REQ_MIN;
+	else if (num_requests > NUM_REQ_MAX)
+		num_requests = NUM_REQ_MAX;
+
+	if (num_ae_requests < NUM_AE_MIN)
+		num_ae_requests = NUM_AE_MIN;
+	else if (num_ae_requests > NUM_AE_MAX)
+		num_ae_requests = NUM_AE_MAX;
+
+	/* set up other globals */
+
+	for (i = 0; i < MAX_ADAPTERS; i++)
+		esas2r_adapters[i] = NULL;
+
+	/* initialize */
+
+	driver_template.module = THIS_MODULE;
+
+	if (pci_register_driver(&esas2r_pci_driver) != 0)
+		esas2r_log(ESAS2R_LOG_CRIT, "pci_register_driver FAILED");
+	else
+		esas2r_log(ESAS2R_LOG_INFO, "pci_register_driver() OK");
+
+	if (!found_adapters) {
+		pci_unregister_driver(&esas2r_pci_driver);
+		esas2r_cleanup(NULL);
+
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "driver will not be loaded because no ATTO "
+			   "%s devices were found",
+			   ESAS2R_DRVR_NAME);
+		return -1;
+	} else {
+		esas2r_log(ESAS2R_LOG_INFO, "found %d adapters",
+			   found_adapters);
+	}
+
+	return 0;
+}
+
+/* Handle ioctl calls to "/proc/scsi/esas2r/ATTOnode" */
+static const struct file_operations esas2r_proc_fops = {
+	.compat_ioctl	= esas2r_proc_ioctl,
+	.unlocked_ioctl = esas2r_proc_ioctl,
+};
+
+static struct Scsi_Host *esas2r_proc_host;
+static int esas2r_proc_major;
+
+long esas2r_proc_ioctl(struct file *fp, unsigned int cmd, unsigned long arg)
+{
+	return esas2r_ioctl_handler(esas2r_proc_host->hostdata,
+				    (int)cmd, (void __user *)arg);
+}
+
+static void __exit esas2r_exit(void)
+{
+	esas2r_log(ESAS2R_LOG_INFO, "%s called", __func__);
+
+	if (esas2r_proc_major > 0) {
+		esas2r_log(ESAS2R_LOG_INFO, "unregister proc");
+
+		remove_proc_entry(ATTONODE_NAME,
+				  esas2r_proc_host->hostt->proc_dir);
+		unregister_chrdev(esas2r_proc_major, ESAS2R_DRVR_NAME);
+
+		esas2r_proc_major = 0;
+	}
+
+	esas2r_log(ESAS2R_LOG_INFO, "pci_unregister_driver() called");
+
+	pci_unregister_driver(&esas2r_pci_driver);
+}
+
+int esas2r_show_info(struct seq_file *m, struct Scsi_Host *sh)
+{
+	struct esas2r_adapter *a = (struct esas2r_adapter *)sh->hostdata;
+
+	struct esas2r_target *t;
+	int dev_count = 0;
+
+	esas2r_log(ESAS2R_LOG_DEBG, "esas2r_show_info (%p,%d)", m, sh->host_no);
+
+	seq_printf(m, ESAS2R_LONGNAME "\n"
+		   "Driver version: "ESAS2R_VERSION_STR "\n"
+		   "Flash version: %s\n"
+		   "Firmware version: %s\n"
+		   "Copyright "ESAS2R_COPYRIGHT_YEARS "\n"
+		   "http://www.attotech.com\n"
+		   "\n",
+		   a->flash_rev,
+		   a->fw_rev[0] ? a->fw_rev : "(none)");
+
+
+	seq_printf(m, "Adapter information:\n"
+		   "--------------------\n"
+		   "Model: %s\n"
+		   "SAS address: %02X%02X%02X%02X:%02X%02X%02X%02X\n",
+		   esas2r_get_model_name(a),
+		   a->nvram->sas_addr[0],
+		   a->nvram->sas_addr[1],
+		   a->nvram->sas_addr[2],
+		   a->nvram->sas_addr[3],
+		   a->nvram->sas_addr[4],
+		   a->nvram->sas_addr[5],
+		   a->nvram->sas_addr[6],
+		   a->nvram->sas_addr[7]);
+
+	seq_puts(m, "\n"
+		   "Discovered devices:\n"
+		   "\n"
+		   "   #  Target ID\n"
+		   "---------------\n");
+
+	for (t = a->targetdb; t < a->targetdb_end; t++)
+		if (t->buffered_target_state == TS_PRESENT) {
+			seq_printf(m, " %3d   %3d\n",
+				   ++dev_count,
+				   (u16)(uintptr_t)(t - a->targetdb));
+		}
+
+	if (dev_count == 0)
+		seq_puts(m, "none\n");
+
+	seq_puts(m, "\n");
+	return 0;
+
+}
+
+int esas2r_release(struct Scsi_Host *sh)
+{
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(sh->shost_gendev),
+		       "esas2r_release() called");
+
+	esas2r_cleanup(sh);
+	if (sh->irq)
+		free_irq(sh->irq, NULL);
+	scsi_unregister(sh);
+	return 0;
+}
+
+const char *esas2r_info(struct Scsi_Host *sh)
+{
+	struct esas2r_adapter *a = (struct esas2r_adapter *)sh->hostdata;
+	static char esas2r_info_str[512];
+
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(sh->shost_gendev),
+		       "esas2r_info() called");
+
+	/*
+	 * if we haven't done so already, register as a char driver
+	 * and stick a node under "/proc/scsi/esas2r/ATTOnode"
+	 */
+
+	if (esas2r_proc_major <= 0) {
+		esas2r_proc_host = sh;
+
+		esas2r_proc_major = register_chrdev(0, ESAS2R_DRVR_NAME,
+						    &esas2r_proc_fops);
+
+		esas2r_log_dev(ESAS2R_LOG_DEBG, &(sh->shost_gendev),
+			       "register_chrdev (major %d)",
+			       esas2r_proc_major);
+
+		if (esas2r_proc_major > 0) {
+			struct proc_dir_entry *pde;
+
+			pde = proc_create(ATTONODE_NAME, 0,
+					  sh->hostt->proc_dir,
+					  &esas2r_proc_fops);
+
+			if (!pde) {
+				esas2r_log_dev(ESAS2R_LOG_WARN,
+					       &(sh->shost_gendev),
+					       "failed to create_proc_entry");
+				esas2r_proc_major = -1;
+			}
+		}
+	}
+
+	sprintf(esas2r_info_str,
+		ESAS2R_LONGNAME " (bus 0x%02X, device 0x%02X, IRQ 0x%02X)"
+		" driver version: "ESAS2R_VERSION_STR "  firmware version: "
+		"%s\n",
+		a->pcid->bus->number, a->pcid->devfn, a->pcid->irq,
+		a->fw_rev[0] ? a->fw_rev : "(none)");
+
+	return esas2r_info_str;
+}
+
+/* Callback for building a request scatter/gather list */
+static u32 get_physaddr_from_sgc(struct esas2r_sg_context *sgc, u64 *addr)
+{
+	u32 len;
+
+	if (likely(sgc->cur_offset == sgc->exp_offset)) {
+		/*
+		 * the normal case: caller used all bytes from previous call, so
+		 * expected offset is the same as the current offset.
+		 */
+
+		if (sgc->sgel_count < sgc->num_sgel) {
+			/* retrieve next segment, except for first time */
+			if (sgc->exp_offset > (u8 *)0) {
+				/* advance current segment */
+				sgc->cur_sgel = sg_next(sgc->cur_sgel);
+				++(sgc->sgel_count);
+			}
+
+
+			len = sg_dma_len(sgc->cur_sgel);
+			(*addr) = sg_dma_address(sgc->cur_sgel);
+
+			/* save the total # bytes returned to caller so far */
+			sgc->exp_offset += len;
+
+		} else {
+			len = 0;
+		}
+	} else if (sgc->cur_offset < sgc->exp_offset) {
+		/*
+		 * caller did not use all bytes from previous call. need to
+		 * compute the address based on current segment.
+		 */
+
+		len = sg_dma_len(sgc->cur_sgel);
+		(*addr) = sg_dma_address(sgc->cur_sgel);
+
+		sgc->exp_offset -= len;
+
+		/* calculate PA based on prev segment address and offsets */
+		*addr = *addr +
+			(sgc->cur_offset - sgc->exp_offset);
+
+		sgc->exp_offset += len;
+
+		/* re-calculate length based on offset */
+		len = lower_32_bits(
+			sgc->exp_offset - sgc->cur_offset);
+	} else {   /* if ( sgc->cur_offset > sgc->exp_offset ) */
+		   /*
+		    * we don't expect the caller to skip ahead.
+		    * cur_offset will never exceed the len we return
+		    */
+		len = 0;
+	}
+
+	return len;
+}
+
+int esas2r_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd)
+{
+	struct esas2r_adapter *a =
+		(struct esas2r_adapter *)cmd->device->host->hostdata;
+	struct esas2r_request *rq;
+	struct esas2r_sg_context sgc;
+	unsigned bufflen;
+
+	/* Assume success, if it fails we will fix the result later. */
+	cmd->result = DID_OK << 16;
+
+	if (unlikely(a->flags & AF_DEGRADED_MODE)) {
+		cmd->result = DID_NO_CONNECT << 16;
+		cmd->scsi_done(cmd);
+		return 0;
+	}
+
+	rq = esas2r_alloc_request(a);
+	if (unlikely(rq == NULL)) {
+		esas2r_debug("esas2r_alloc_request failed");
+		return SCSI_MLQUEUE_HOST_BUSY;
+	}
+
+	rq->cmd = cmd;
+	bufflen = scsi_bufflen(cmd);
+
+	if (likely(bufflen != 0)) {
+		if (cmd->sc_data_direction == DMA_TO_DEVICE)
+			rq->vrq->scsi.flags |= cpu_to_le32(FCP_CMND_WRD);
+		else if (cmd->sc_data_direction == DMA_FROM_DEVICE)
+			rq->vrq->scsi.flags |= cpu_to_le32(FCP_CMND_RDD);
+	}
+
+	memcpy(rq->vrq->scsi.cdb, cmd->cmnd, cmd->cmd_len);
+	rq->vrq->scsi.length = cpu_to_le32(bufflen);
+	rq->target_id = cmd->device->id;
+	rq->vrq->scsi.flags |= cpu_to_le32(cmd->device->lun);
+	rq->sense_buf = cmd->sense_buffer;
+	rq->sense_len = SCSI_SENSE_BUFFERSIZE;
+
+	esas2r_sgc_init(&sgc, a, rq, NULL);
+
+	sgc.length = bufflen;
+	sgc.cur_offset = NULL;
+
+	sgc.cur_sgel = scsi_sglist(cmd);
+	sgc.exp_offset = NULL;
+	sgc.num_sgel = scsi_dma_map(cmd);
+	sgc.sgel_count = 0;
+
+	if (unlikely(sgc.num_sgel < 0)) {
+		esas2r_free_request(a, rq);
+		return SCSI_MLQUEUE_HOST_BUSY;
+	}
+
+	sgc.get_phys_addr = (PGETPHYSADDR)get_physaddr_from_sgc;
+
+	if (unlikely(!esas2r_build_sg_list(a, rq, &sgc))) {
+		scsi_dma_unmap(cmd);
+		esas2r_free_request(a, rq);
+		return SCSI_MLQUEUE_HOST_BUSY;
+	}
+
+	esas2r_debug("start request %p to %d:%d\n", rq, (int)cmd->device->id,
+		     (int)cmd->device->lun);
+
+	esas2r_start_request(a, rq);
+
+	return 0;
+}
+
+static void complete_task_management_request(struct esas2r_adapter *a,
+					     struct esas2r_request *rq)
+{
+	(*rq->task_management_status_ptr) = rq->req_stat;
+	esas2r_free_request(a, rq);
+}
+
+/**
+ * Searches the specified queue for the specified queue for the command
+ * to abort.
+ *
+ * @param [in] a
+ * @param [in] abort_request
+ * @param [in] cmd
+ * t
+ * @return 0 on failure, 1 if command was not found, 2 if command was found
+ */
+static int esas2r_check_active_queue(struct esas2r_adapter *a,
+				     struct esas2r_request **abort_request,
+				     struct scsi_cmnd *cmd,
+				     struct list_head *queue)
+{
+	bool found = false;
+	struct esas2r_request *ar = *abort_request;
+	struct esas2r_request *rq;
+	struct list_head *element, *next;
+
+	list_for_each_safe(element, next, queue) {
+
+		rq = list_entry(element, struct esas2r_request, req_list);
+
+		if (rq->cmd == cmd) {
+
+			/* Found the request.  See what to do with it. */
+			if (queue == &a->active_list) {
+				/*
+				 * We are searching the active queue, which
+				 * means that we need to send an abort request
+				 * to the firmware.
+				 */
+				ar = esas2r_alloc_request(a);
+				if (ar == NULL) {
+					esas2r_log_dev(ESAS2R_LOG_WARN,
+						       &(a->host->shost_gendev),
+						       "unable to allocate an abort request for cmd %p",
+						       cmd);
+					return 0; /* Failure */
+				}
+
+				/*
+				 * Task management request must be formatted
+				 * with a lock held.
+				 */
+				ar->sense_len = 0;
+				ar->vrq->scsi.length = 0;
+				ar->target_id = rq->target_id;
+				ar->vrq->scsi.flags |= cpu_to_le32(
+					(u8)le32_to_cpu(rq->vrq->scsi.flags));
+
+				memset(ar->vrq->scsi.cdb, 0,
+				       sizeof(ar->vrq->scsi.cdb));
+
+				ar->vrq->scsi.flags |= cpu_to_le32(
+					FCP_CMND_TRM);
+				ar->vrq->scsi.u.abort_handle =
+					rq->vrq->scsi.handle;
+			} else {
+				/*
+				 * The request is pending but not active on
+				 * the firmware.  Just free it now and we'll
+				 * report the successful abort below.
+				 */
+				list_del_init(&rq->req_list);
+				esas2r_free_request(a, rq);
+			}
+
+			found = true;
+			break;
+		}
+
+	}
+
+	if (!found)
+		return 1;       /* Not found */
+
+	return 2;               /* found */
+
+
+}
+
+int esas2r_eh_abort(struct scsi_cmnd *cmd)
+{
+	struct esas2r_adapter *a =
+		(struct esas2r_adapter *)cmd->device->host->hostdata;
+	struct esas2r_request *abort_request = NULL;
+	unsigned long flags;
+	struct list_head *queue;
+	int result;
+
+	esas2r_log(ESAS2R_LOG_INFO, "eh_abort (%p)", cmd);
+
+	if (a->flags & AF_DEGRADED_MODE) {
+		cmd->result = DID_ABORT << 16;
+
+		scsi_set_resid(cmd, 0);
+
+		cmd->scsi_done(cmd);
+
+		return 0;
+	}
+
+	spin_lock_irqsave(&a->queue_lock, flags);
+
+	/*
+	 * Run through the defer and active queues looking for the request
+	 * to abort.
+	 */
+
+	queue = &a->defer_list;
+
+check_active_queue:
+
+	result = esas2r_check_active_queue(a, &abort_request, cmd, queue);
+
+	if (!result) {
+		spin_unlock_irqrestore(&a->queue_lock, flags);
+		return FAILED;
+	} else if (result == 2 && (queue == &a->defer_list)) {
+		queue = &a->active_list;
+		goto check_active_queue;
+	}
+
+	spin_unlock_irqrestore(&a->queue_lock, flags);
+
+	if (abort_request) {
+		u8 task_management_status = RS_PENDING;
+
+		/*
+		 * the request is already active, so we need to tell
+		 * the firmware to abort it and wait for the response.
+		 */
+
+		abort_request->comp_cb = complete_task_management_request;
+		abort_request->task_management_status_ptr =
+			&task_management_status;
+
+		esas2r_start_request(a, abort_request);
+
+		if (atomic_read(&a->disable_cnt) == 0)
+			esas2r_do_deferred_processes(a);
+
+		while (task_management_status == RS_PENDING)
+			msleep(10);
+
+		/*
+		 * Once we get here, the original request will have been
+		 * completed by the firmware and the abort request will have
+		 * been cleaned up.  we're done!
+		 */
+
+		return SUCCESS;
+	}
+
+	/*
+	 * If we get here, either we found the inactive request and
+	 * freed it, or we didn't find it at all.  Either way, success!
+	 */
+
+	cmd->result = DID_ABORT << 16;
+
+	scsi_set_resid(cmd, 0);
+
+	cmd->scsi_done(cmd);
+
+	return SUCCESS;
+}
+
+static int esas2r_host_bus_reset(struct scsi_cmnd *cmd, bool host_reset)
+{
+	struct esas2r_adapter *a =
+		(struct esas2r_adapter *)cmd->device->host->hostdata;
+
+	if (a->flags & AF_DEGRADED_MODE)
+		return FAILED;
+
+	if (host_reset)
+		esas2r_reset_adapter(a);
+	else
+		esas2r_reset_bus(a);
+
+	/* above call sets the AF_OS_RESET flag.  wait for it to clear. */
+
+	while (a->flags & AF_OS_RESET) {
+		msleep(10);
+
+		if (a->flags & AF_DEGRADED_MODE)
+			return FAILED;
+	}
+
+	if (a->flags & AF_DEGRADED_MODE)
+		return FAILED;
+
+	return SUCCESS;
+}
+
+int esas2r_host_reset(struct scsi_cmnd *cmd)
+{
+	esas2r_log(ESAS2R_LOG_INFO, "host_reset (%p)", cmd);
+
+	return esas2r_host_bus_reset(cmd, true);
+}
+
+int esas2r_bus_reset(struct scsi_cmnd *cmd)
+{
+	esas2r_log(ESAS2R_LOG_INFO, "bus_reset (%p)", cmd);
+
+	return esas2r_host_bus_reset(cmd, false);
+}
+
+static int esas2r_dev_targ_reset(struct scsi_cmnd *cmd, bool target_reset)
+{
+	struct esas2r_adapter *a =
+		(struct esas2r_adapter *)cmd->device->host->hostdata;
+	struct esas2r_request *rq;
+	u8 task_management_status = RS_PENDING;
+	bool completed;
+
+	if (a->flags & AF_DEGRADED_MODE)
+		return FAILED;
+
+retry:
+	rq = esas2r_alloc_request(a);
+	if (rq == NULL) {
+		if (target_reset) {
+			esas2r_log(ESAS2R_LOG_CRIT,
+				   "unable to allocate a request for a "
+				   "target reset (%d)!",
+				   cmd->device->id);
+		} else {
+			esas2r_log(ESAS2R_LOG_CRIT,
+				   "unable to allocate a request for a "
+				   "device reset (%d:%d)!",
+				   cmd->device->id,
+				   cmd->device->lun);
+		}
+
+
+		return FAILED;
+	}
+
+	rq->target_id = cmd->device->id;
+	rq->vrq->scsi.flags |= cpu_to_le32(cmd->device->lun);
+	rq->req_stat = RS_PENDING;
+
+	rq->comp_cb = complete_task_management_request;
+	rq->task_management_status_ptr = &task_management_status;
+
+	if (target_reset) {
+		esas2r_debug("issuing target reset (%p) to id %d", rq,
+			     cmd->device->id);
+		completed = esas2r_send_task_mgmt(a, rq, 0x20);
+	} else {
+		esas2r_debug("issuing device reset (%p) to id %d lun %d", rq,
+			     cmd->device->id, cmd->device->lun);
+		completed = esas2r_send_task_mgmt(a, rq, 0x10);
+	}
+
+	if (completed) {
+		/* Task management cmd completed right away, need to free it. */
+
+		esas2r_free_request(a, rq);
+	} else {
+		/*
+		 * Wait for firmware to complete the request.  Completion
+		 * callback will free it.
+		 */
+		while (task_management_status == RS_PENDING)
+			msleep(10);
+	}
+
+	if (a->flags & AF_DEGRADED_MODE)
+		return FAILED;
+
+	if (task_management_status == RS_BUSY) {
+		/*
+		 * Busy, probably because we are flashing.  Wait a bit and
+		 * try again.
+		 */
+		msleep(100);
+		goto retry;
+	}
+
+	return SUCCESS;
+}
+
+int esas2r_device_reset(struct scsi_cmnd *cmd)
+{
+	esas2r_log(ESAS2R_LOG_INFO, "device_reset (%p)", cmd);
+
+	return esas2r_dev_targ_reset(cmd, false);
+
+}
+
+int esas2r_target_reset(struct scsi_cmnd *cmd)
+{
+	esas2r_log(ESAS2R_LOG_INFO, "target_reset (%p)", cmd);
+
+	return esas2r_dev_targ_reset(cmd, true);
+}
+
+int esas2r_change_queue_depth(struct scsi_device *dev, int depth, int reason)
+{
+	esas2r_log(ESAS2R_LOG_INFO, "change_queue_depth %p, %d", dev, depth);
+
+	scsi_adjust_queue_depth(dev, scsi_get_tag_type(dev), depth);
+
+	return dev->queue_depth;
+}
+
+int esas2r_change_queue_type(struct scsi_device *dev, int type)
+{
+	esas2r_log(ESAS2R_LOG_INFO, "change_queue_type %p, %d", dev, type);
+
+	if (dev->tagged_supported) {
+		scsi_set_tag_type(dev, type);
+
+		if (type)
+			scsi_activate_tcq(dev, dev->queue_depth);
+		else
+			scsi_deactivate_tcq(dev, dev->queue_depth);
+	} else {
+		type = 0;
+	}
+
+	return type;
+}
+
+int esas2r_slave_alloc(struct scsi_device *dev)
+{
+	return 0;
+}
+
+int esas2r_slave_configure(struct scsi_device *dev)
+{
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(dev->sdev_gendev),
+		       "esas2r_slave_configure()");
+
+	if (dev->tagged_supported) {
+		scsi_set_tag_type(dev, MSG_SIMPLE_TAG);
+		scsi_activate_tcq(dev, cmd_per_lun);
+	} else {
+		scsi_set_tag_type(dev, 0);
+		scsi_deactivate_tcq(dev, cmd_per_lun);
+	}
+
+	return 0;
+}
+
+void esas2r_slave_destroy(struct scsi_device *dev)
+{
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(dev->sdev_gendev),
+		       "esas2r_slave_destroy()");
+}
+
+void esas2r_log_request_failure(struct esas2r_adapter *a,
+				struct esas2r_request *rq)
+{
+	u8 reqstatus = rq->req_stat;
+
+	if (reqstatus == RS_SUCCESS)
+		return;
+
+	if (rq->vrq->scsi.function == VDA_FUNC_SCSI) {
+		if (reqstatus == RS_SCSI_ERROR) {
+			if (rq->func_rsp.scsi_rsp.sense_len >= 13) {
+				esas2r_log(ESAS2R_LOG_WARN,
+					   "request failure - SCSI error %x ASC:%x ASCQ:%x CDB:%x",
+					   rq->sense_buf[2], rq->sense_buf[12],
+					   rq->sense_buf[13],
+					   rq->vrq->scsi.cdb[0]);
+			} else {
+				esas2r_log(ESAS2R_LOG_WARN,
+					   "request failure - SCSI error CDB:%x\n",
+					   rq->vrq->scsi.cdb[0]);
+			}
+		} else if ((rq->vrq->scsi.cdb[0] != INQUIRY
+			    && rq->vrq->scsi.cdb[0] != REPORT_LUNS)
+			   || (reqstatus != RS_SEL
+			       && reqstatus != RS_SEL2)) {
+			if ((reqstatus == RS_UNDERRUN) &&
+			    (rq->vrq->scsi.cdb[0] == INQUIRY)) {
+				/* Don't log inquiry underruns */
+			} else {
+				esas2r_log(ESAS2R_LOG_WARN,
+					   "request failure - cdb:%x reqstatus:%d target:%d",
+					   rq->vrq->scsi.cdb[0], reqstatus,
+					   rq->target_id);
+			}
+		}
+	}
+}
+
+void esas2r_wait_request(struct esas2r_adapter *a, struct esas2r_request *rq)
+{
+	u32 starttime;
+	u32 timeout;
+
+	starttime = jiffies_to_msecs(jiffies);
+	timeout = rq->timeout ? rq->timeout : 5000;
+
+	while (true) {
+		esas2r_polled_interrupt(a);
+
+		if (rq->req_stat != RS_STARTED)
+			break;
+
+		schedule_timeout_interruptible(msecs_to_jiffies(100));
+
+		if ((jiffies_to_msecs(jiffies) - starttime) > timeout) {
+			esas2r_hdebug("request TMO");
+			esas2r_bugon();
+
+			rq->req_stat = RS_TIMEOUT;
+
+			esas2r_local_reset_adapter(a);
+			return;
+		}
+	}
+}
+
+u32 esas2r_map_data_window(struct esas2r_adapter *a, u32 addr_lo)
+{
+	u32 offset = addr_lo & (MW_DATA_WINDOW_SIZE - 1);
+	u32 base = addr_lo & -(signed int)MW_DATA_WINDOW_SIZE;
+
+	if (a->window_base != base) {
+		esas2r_write_register_dword(a, MVR_PCI_WIN1_REMAP,
+					    base | MVRPW1R_ENABLE);
+		esas2r_flush_register_dword(a, MVR_PCI_WIN1_REMAP);
+		a->window_base = base;
+	}
+
+	return offset;
+}
+
+/* Read a block of data from chip memory */
+bool esas2r_read_mem_block(struct esas2r_adapter *a,
+			   void *to,
+			   u32 from,
+			   u32 size)
+{
+	u8 *end = (u8 *)to;
+
+	while (size) {
+		u32 len;
+		u32 offset;
+		u32 iatvr;
+
+		iatvr = (from & -(signed int)MW_DATA_WINDOW_SIZE);
+
+		esas2r_map_data_window(a, iatvr);
+
+		offset = from & (MW_DATA_WINDOW_SIZE - 1);
+		len = size;
+
+		if (len > MW_DATA_WINDOW_SIZE - offset)
+			len = MW_DATA_WINDOW_SIZE - offset;
+
+		from += len;
+		size -= len;
+
+		while (len--) {
+			*end++ = esas2r_read_data_byte(a, offset);
+			offset++;
+		}
+	}
+
+	return true;
+}
+
+void esas2r_nuxi_mgt_data(u8 function, void *data)
+{
+	struct atto_vda_grp_info *g;
+	struct atto_vda_devinfo *d;
+	struct atto_vdapart_info *p;
+	struct atto_vda_dh_info *h;
+	struct atto_vda_metrics_info *m;
+	struct atto_vda_schedule_info *s;
+	struct atto_vda_buzzer_info *b;
+	u8 i;
+
+	switch (function) {
+	case VDAMGT_BUZZER_INFO:
+	case VDAMGT_BUZZER_SET:
+
+		b = (struct atto_vda_buzzer_info *)data;
+
+		b->duration = le32_to_cpu(b->duration);
+		break;
+
+	case VDAMGT_SCHEDULE_INFO:
+	case VDAMGT_SCHEDULE_EVENT:
+
+		s = (struct atto_vda_schedule_info *)data;
+
+		s->id = le32_to_cpu(s->id);
+
+		break;
+
+	case VDAMGT_DEV_INFO:
+	case VDAMGT_DEV_CLEAN:
+	case VDAMGT_DEV_PT_INFO:
+	case VDAMGT_DEV_FEATURES:
+	case VDAMGT_DEV_PT_FEATURES:
+	case VDAMGT_DEV_OPERATION:
+
+		d = (struct atto_vda_devinfo *)data;
+
+		d->capacity = le64_to_cpu(d->capacity);
+		d->block_size = le32_to_cpu(d->block_size);
+		d->ses_dev_index = le16_to_cpu(d->ses_dev_index);
+		d->target_id = le16_to_cpu(d->target_id);
+		d->lun = le16_to_cpu(d->lun);
+		d->features = le16_to_cpu(d->features);
+		break;
+
+	case VDAMGT_GRP_INFO:
+	case VDAMGT_GRP_CREATE:
+	case VDAMGT_GRP_DELETE:
+	case VDAMGT_ADD_STORAGE:
+	case VDAMGT_MEMBER_ADD:
+	case VDAMGT_GRP_COMMIT:
+	case VDAMGT_GRP_REBUILD:
+	case VDAMGT_GRP_COMMIT_INIT:
+	case VDAMGT_QUICK_RAID:
+	case VDAMGT_GRP_FEATURES:
+	case VDAMGT_GRP_COMMIT_INIT_AUTOMAP:
+	case VDAMGT_QUICK_RAID_INIT_AUTOMAP:
+	case VDAMGT_SPARE_LIST:
+	case VDAMGT_SPARE_ADD:
+	case VDAMGT_SPARE_REMOVE:
+	case VDAMGT_LOCAL_SPARE_ADD:
+	case VDAMGT_GRP_OPERATION:
+
+		g = (struct atto_vda_grp_info *)data;
+
+		g->capacity = le64_to_cpu(g->capacity);
+		g->block_size = le32_to_cpu(g->block_size);
+		g->interleave = le32_to_cpu(g->interleave);
+		g->features = le16_to_cpu(g->features);
+
+		for (i = 0; i < 32; i++)
+			g->members[i] = le16_to_cpu(g->members[i]);
+
+		break;
+
+	case VDAMGT_PART_INFO:
+	case VDAMGT_PART_MAP:
+	case VDAMGT_PART_UNMAP:
+	case VDAMGT_PART_AUTOMAP:
+	case VDAMGT_PART_SPLIT:
+	case VDAMGT_PART_MERGE:
+
+		p = (struct atto_vdapart_info *)data;
+
+		p->part_size = le64_to_cpu(p->part_size);
+		p->start_lba = le32_to_cpu(p->start_lba);
+		p->block_size = le32_to_cpu(p->block_size);
+		p->target_id = le16_to_cpu(p->target_id);
+		break;
+
+	case VDAMGT_DEV_HEALTH_REQ:
+
+		h = (struct atto_vda_dh_info *)data;
+
+		h->med_defect_cnt = le32_to_cpu(h->med_defect_cnt);
+		h->info_exc_cnt = le32_to_cpu(h->info_exc_cnt);
+		break;
+
+	case VDAMGT_DEV_METRICS:
+
+		m = (struct atto_vda_metrics_info *)data;
+
+		for (i = 0; i < 32; i++)
+			m->dev_indexes[i] = le16_to_cpu(m->dev_indexes[i]);
+
+		break;
+
+	default:
+		break;
+	}
+}
+
+void esas2r_nuxi_cfg_data(u8 function, void *data)
+{
+	struct atto_vda_cfg_init *ci;
+
+	switch (function) {
+	case VDA_CFG_INIT:
+	case VDA_CFG_GET_INIT:
+	case VDA_CFG_GET_INIT2:
+
+		ci = (struct atto_vda_cfg_init *)data;
+
+		ci->date_time.year = le16_to_cpu(ci->date_time.year);
+		ci->sgl_page_size = le32_to_cpu(ci->sgl_page_size);
+		ci->vda_version = le32_to_cpu(ci->vda_version);
+		ci->epoch_time = le32_to_cpu(ci->epoch_time);
+		ci->ioctl_tunnel = le32_to_cpu(ci->ioctl_tunnel);
+		ci->num_targets_backend = le32_to_cpu(ci->num_targets_backend);
+		break;
+
+	default:
+		break;
+	}
+}
+
+void esas2r_nuxi_ae_data(union atto_vda_ae *ae)
+{
+	struct atto_vda_ae_raid *r = &ae->raid;
+	struct atto_vda_ae_lu *l = &ae->lu;
+
+	switch (ae->hdr.bytype) {
+	case VDAAE_HDR_TYPE_RAID:
+
+		r->dwflags = le32_to_cpu(r->dwflags);
+		break;
+
+	case VDAAE_HDR_TYPE_LU:
+
+		l->dwevent = le32_to_cpu(l->dwevent);
+		l->wphys_target_id = le16_to_cpu(l->wphys_target_id);
+		l->id.tgtlun.wtarget_id = le16_to_cpu(l->id.tgtlun.wtarget_id);
+
+		if (l->hdr.bylength >= offsetof(struct atto_vda_ae_lu, id)
+		    + sizeof(struct atto_vda_ae_lu_tgt_lun_raid)) {
+			l->id.tgtlun_raid.dwinterleave
+				= le32_to_cpu(l->id.tgtlun_raid.dwinterleave);
+			l->id.tgtlun_raid.dwblock_size
+				= le32_to_cpu(l->id.tgtlun_raid.dwblock_size);
+		}
+
+		break;
+
+	case VDAAE_HDR_TYPE_DISK:
+	default:
+		break;
+	}
+}
+
+void esas2r_free_request(struct esas2r_adapter *a, struct esas2r_request *rq)
+{
+	unsigned long flags;
+
+	esas2r_rq_destroy_request(rq, a);
+	spin_lock_irqsave(&a->request_lock, flags);
+	list_add(&rq->comp_list, &a->avail_request);
+	spin_unlock_irqrestore(&a->request_lock, flags);
+}
+
+struct esas2r_request *esas2r_alloc_request(struct esas2r_adapter *a)
+{
+	struct esas2r_request *rq;
+	unsigned long flags;
+
+	spin_lock_irqsave(&a->request_lock, flags);
+
+	if (unlikely(list_empty(&a->avail_request))) {
+		spin_unlock_irqrestore(&a->request_lock, flags);
+		return NULL;
+	}
+
+	rq = list_first_entry(&a->avail_request, struct esas2r_request,
+			      comp_list);
+	list_del(&rq->comp_list);
+	spin_unlock_irqrestore(&a->request_lock, flags);
+	esas2r_rq_init_request(rq, a);
+
+	return rq;
+
+}
+
+void esas2r_complete_request_cb(struct esas2r_adapter *a,
+				struct esas2r_request *rq)
+{
+	esas2r_debug("completing request %p\n", rq);
+
+	scsi_dma_unmap(rq->cmd);
+
+	if (unlikely(rq->req_stat != RS_SUCCESS)) {
+		esas2r_debug("[%x STATUS %x:%x (%x)]", rq->target_id,
+			     rq->req_stat,
+			     rq->func_rsp.scsi_rsp.scsi_stat,
+			     rq->cmd);
+
+		rq->cmd->result =
+			((esas2r_req_status_to_error(rq->req_stat) << 16)
+			 | (rq->func_rsp.scsi_rsp.scsi_stat & STATUS_MASK));
+
+		if (rq->req_stat == RS_UNDERRUN)
+			scsi_set_resid(rq->cmd,
+				       le32_to_cpu(rq->func_rsp.scsi_rsp.
+						   residual_length));
+		else
+			scsi_set_resid(rq->cmd, 0);
+	}
+
+	rq->cmd->scsi_done(rq->cmd);
+
+	esas2r_free_request(a, rq);
+}
+
+/* Run tasklet to handle stuff outside of interrupt context. */
+void esas2r_adapter_tasklet(unsigned long context)
+{
+	struct esas2r_adapter *a = (struct esas2r_adapter *)context;
+
+	if (unlikely(a->flags2 & AF2_TIMER_TICK)) {
+		esas2r_lock_clear_flags(&a->flags2, AF2_TIMER_TICK);
+		esas2r_timer_tick(a);
+	}
+
+	if (likely(a->flags2 & AF2_INT_PENDING)) {
+		esas2r_lock_clear_flags(&a->flags2, AF2_INT_PENDING);
+		esas2r_adapter_interrupt(a);
+	}
+
+	if (esas2r_is_tasklet_pending(a))
+		esas2r_do_tasklet_tasks(a);
+
+	if (esas2r_is_tasklet_pending(a)
+	    || (a->flags2 & AF2_INT_PENDING)
+	    || (a->flags2 & AF2_TIMER_TICK)) {
+		esas2r_lock_clear_flags(&a->flags, AF_TASKLET_SCHEDULED);
+		esas2r_schedule_tasklet(a);
+	} else {
+		esas2r_lock_clear_flags(&a->flags, AF_TASKLET_SCHEDULED);
+	}
+}
+
+static void esas2r_timer_callback(unsigned long context);
+
+void esas2r_kickoff_timer(struct esas2r_adapter *a)
+{
+	init_timer(&a->timer);
+
+	a->timer.function = esas2r_timer_callback;
+	a->timer.data = (unsigned long)a;
+	a->timer.expires = jiffies +
+			   msecs_to_jiffies(100);
+
+	add_timer(&a->timer);
+}
+
+static void esas2r_timer_callback(unsigned long context)
+{
+	struct esas2r_adapter *a = (struct esas2r_adapter *)context;
+
+	esas2r_lock_set_flags(&a->flags2, AF2_TIMER_TICK);
+
+	esas2r_schedule_tasklet(a);
+
+	esas2r_kickoff_timer(a);
+}
+
+/*
+ * Firmware events need to be handled outside of interrupt context
+ * so we schedule a delayed_work to handle them.
+ */
+
+static void
+esas2r_free_fw_event(struct esas2r_fw_event_work *fw_event)
+{
+	unsigned long flags;
+	struct esas2r_adapter *a = fw_event->a;
+
+	spin_lock_irqsave(&a->fw_event_lock, flags);
+	list_del(&fw_event->list);
+	kfree(fw_event);
+	spin_unlock_irqrestore(&a->fw_event_lock, flags);
+}
+
+void
+esas2r_fw_event_off(struct esas2r_adapter *a)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&a->fw_event_lock, flags);
+	a->fw_events_off = 1;
+	spin_unlock_irqrestore(&a->fw_event_lock, flags);
+}
+
+void
+esas2r_fw_event_on(struct esas2r_adapter *a)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&a->fw_event_lock, flags);
+	a->fw_events_off = 0;
+	spin_unlock_irqrestore(&a->fw_event_lock, flags);
+}
+
+static void esas2r_add_device(struct esas2r_adapter *a, u16 target_id)
+{
+	int ret;
+	struct scsi_device *scsi_dev;
+
+	scsi_dev = scsi_device_lookup(a->host, 0, target_id, 0);
+
+	if (scsi_dev) {
+		esas2r_log_dev(
+			ESAS2R_LOG_WARN,
+			&(scsi_dev->
+			  sdev_gendev),
+			"scsi device already exists at id %d", target_id);
+
+		scsi_device_put(scsi_dev);
+	} else {
+		esas2r_log_dev(
+			ESAS2R_LOG_INFO,
+			&(a->host->
+			  shost_gendev),
+			"scsi_add_device() called for 0:%d:0",
+			target_id);
+
+		ret = scsi_add_device(a->host, 0, target_id, 0);
+		if (ret) {
+			esas2r_log_dev(
+				ESAS2R_LOG_CRIT,
+				&(a->host->
+				  shost_gendev),
+				"scsi_add_device failed with %d for id %d",
+				ret, target_id);
+		}
+	}
+}
+
+static void esas2r_remove_device(struct esas2r_adapter *a, u16 target_id)
+{
+	struct scsi_device *scsi_dev;
+
+	scsi_dev = scsi_device_lookup(a->host, 0, target_id, 0);
+
+	if (scsi_dev) {
+		scsi_device_set_state(scsi_dev, SDEV_OFFLINE);
+
+		esas2r_log_dev(
+			ESAS2R_LOG_INFO,
+			&(scsi_dev->
+			  sdev_gendev),
+			"scsi_remove_device() called for 0:%d:0",
+			target_id);
+
+		scsi_remove_device(scsi_dev);
+
+		esas2r_log_dev(
+			ESAS2R_LOG_INFO,
+			&(scsi_dev->
+			  sdev_gendev),
+			"scsi_device_put() called");
+
+		scsi_device_put(scsi_dev);
+	} else {
+		esas2r_log_dev(
+			ESAS2R_LOG_WARN,
+			&(a->host->shost_gendev),
+			"no target found at id %d",
+			target_id);
+	}
+}
+
+/*
+ * Sends a firmware asynchronous event to anyone who happens to be
+ * listening on the defined ATTO VDA event ports.
+ */
+static void esas2r_send_ae_event(struct esas2r_fw_event_work *fw_event)
+{
+	struct esas2r_vda_ae *ae = (struct esas2r_vda_ae *)fw_event->data;
+	char *type;
+
+	switch (ae->vda_ae.hdr.bytype) {
+	case VDAAE_HDR_TYPE_RAID:
+		type = "RAID group state change";
+		break;
+
+	case VDAAE_HDR_TYPE_LU:
+		type = "Mapped destination LU change";
+		break;
+
+	case VDAAE_HDR_TYPE_DISK:
+		type = "Physical disk inventory change";
+		break;
+
+	case VDAAE_HDR_TYPE_RESET:
+		type = "Firmware reset";
+		break;
+
+	case VDAAE_HDR_TYPE_LOG_INFO:
+		type = "Event Log message (INFO level)";
+		break;
+
+	case VDAAE_HDR_TYPE_LOG_WARN:
+		type = "Event Log message (WARN level)";
+		break;
+
+	case VDAAE_HDR_TYPE_LOG_CRIT:
+		type = "Event Log message (CRIT level)";
+		break;
+
+	case VDAAE_HDR_TYPE_LOG_FAIL:
+		type = "Event Log message (FAIL level)";
+		break;
+
+	case VDAAE_HDR_TYPE_NVC:
+		type = "NVCache change";
+		break;
+
+	case VDAAE_HDR_TYPE_TLG_INFO:
+		type = "Time stamped log message (INFO level)";
+		break;
+
+	case VDAAE_HDR_TYPE_TLG_WARN:
+		type = "Time stamped log message (WARN level)";
+		break;
+
+	case VDAAE_HDR_TYPE_TLG_CRIT:
+		type = "Time stamped log message (CRIT level)";
+		break;
+
+	case VDAAE_HDR_TYPE_PWRMGT:
+		type = "Power management";
+		break;
+
+	case VDAAE_HDR_TYPE_MUTE:
+		type = "Mute button pressed";
+		break;
+
+	case VDAAE_HDR_TYPE_DEV:
+		type = "Device attribute change";
+		break;
+
+	default:
+		type = "Unknown";
+		break;
+	}
+
+	esas2r_log(ESAS2R_LOG_WARN,
+		   "An async event of type \"%s\" was received from the firmware.  The event contents are:",
+		   type);
+	esas2r_log_hexdump(ESAS2R_LOG_WARN, &ae->vda_ae,
+			   ae->vda_ae.hdr.bylength);
+
+}
+
+static void
+esas2r_firmware_event_work(struct work_struct *work)
+{
+	struct esas2r_fw_event_work *fw_event =
+		container_of(work, struct esas2r_fw_event_work, work.work);
+
+	struct esas2r_adapter *a = fw_event->a;
+
+	u16 target_id = *(u16 *)&fw_event->data[0];
+
+	if (a->fw_events_off)
+		goto done;
+
+	switch (fw_event->type) {
+	case fw_event_null:
+		break; /* do nothing */
+
+	case fw_event_lun_change:
+		esas2r_remove_device(a, target_id);
+		esas2r_add_device(a, target_id);
+		break;
+
+	case fw_event_present:
+		esas2r_add_device(a, target_id);
+		break;
+
+	case fw_event_not_present:
+		esas2r_remove_device(a, target_id);
+		break;
+
+	case fw_event_vda_ae:
+		esas2r_send_ae_event(fw_event);
+		break;
+	}
+
+done:
+	esas2r_free_fw_event(fw_event);
+}
+
+void esas2r_queue_fw_event(struct esas2r_adapter *a,
+			   enum fw_event_type type,
+			   void *data,
+			   int data_sz)
+{
+	struct esas2r_fw_event_work *fw_event;
+	unsigned long flags;
+
+	fw_event = kzalloc(sizeof(struct esas2r_fw_event_work), GFP_ATOMIC);
+	if (!fw_event) {
+		esas2r_log(ESAS2R_LOG_WARN,
+			   "esas2r_queue_fw_event failed to alloc");
+		return;
+	}
+
+	if (type == fw_event_vda_ae) {
+		struct esas2r_vda_ae *ae =
+			(struct esas2r_vda_ae *)fw_event->data;
+
+		ae->signature = ESAS2R_VDA_EVENT_SIG;
+		ae->bus_number = a->pcid->bus->number;
+		ae->devfn = a->pcid->devfn;
+		memcpy(&ae->vda_ae, data, sizeof(ae->vda_ae));
+	} else {
+		memcpy(fw_event->data, data, data_sz);
+	}
+
+	fw_event->type = type;
+	fw_event->a = a;
+
+	spin_lock_irqsave(&a->fw_event_lock, flags);
+	list_add_tail(&fw_event->list, &a->fw_event_list);
+	INIT_DELAYED_WORK(&fw_event->work, esas2r_firmware_event_work);
+	queue_delayed_work_on(
+		smp_processor_id(), a->fw_event_q, &fw_event->work,
+		msecs_to_jiffies(1));
+	spin_unlock_irqrestore(&a->fw_event_lock, flags);
+}
+
+void esas2r_target_state_changed(struct esas2r_adapter *a, u16 targ_id,
+				 u8 state)
+{
+	if (state == TS_LUN_CHANGE)
+		esas2r_queue_fw_event(a, fw_event_lun_change, &targ_id,
+				      sizeof(targ_id));
+	else if (state == TS_PRESENT)
+		esas2r_queue_fw_event(a, fw_event_present, &targ_id,
+				      sizeof(targ_id));
+	else if (state == TS_NOT_PRESENT)
+		esas2r_queue_fw_event(a, fw_event_not_present, &targ_id,
+				      sizeof(targ_id));
+}
+
+/* Translate status to a Linux SCSI mid-layer error code */
+int esas2r_req_status_to_error(u8 req_stat)
+{
+	switch (req_stat) {
+	case RS_OVERRUN:
+	case RS_UNDERRUN:
+	case RS_SUCCESS:
+	/*
+	 * NOTE: SCSI mid-layer wants a good status for a SCSI error, because
+	 *       it will check the scsi_stat value in the completion anyway.
+	 */
+	case RS_SCSI_ERROR:
+		return DID_OK;
+
+	case RS_SEL:
+	case RS_SEL2:
+		return DID_NO_CONNECT;
+
+	case RS_RESET:
+		return DID_RESET;
+
+	case RS_ABORTED:
+		return DID_ABORT;
+
+	case RS_BUSY:
+		return DID_BUS_BUSY;
+	}
+
+	/* everything else is just an error. */
+
+	return DID_ERROR;
+}
+
+module_init(esas2r_init);
+module_exit(esas2r_exit);