[PATCH v4 03/10] SCSI: esas2r: Add initialization functions

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Signed-off-by: Bradley Grove <bgrove@xxxxxxxxxxxx>
---
 drivers/scsi/esas2r/esas2r_init.c | 1773 +++++++++++++++++++++++++++++++++++++
 1 file changed, 1773 insertions(+)
 create mode 100644 drivers/scsi/esas2r/esas2r_init.c

diff --git a/drivers/scsi/esas2r/esas2r_init.c b/drivers/scsi/esas2r/esas2r_init.c
new file mode 100644
index 0000000..3a798e7
--- /dev/null
+++ b/drivers/scsi/esas2r/esas2r_init.c
@@ -0,0 +1,1773 @@
+/*
+ *  linux/drivers/scsi/esas2r/esas2r_init.c
+ *      For use with ATTO ExpressSAS R6xx SAS/SATA RAID controllers
+ *
+ *  Copyright (c) 2001-2013 ATTO Technology, Inc.
+ *  (mailto:linuxdrivers@xxxxxxxxxxxx)mpt3sas/mpt3sas_trigger_diag.
+ *
+ * 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"
+
+static bool esas2r_initmem_alloc(struct esas2r_adapter *a,
+				 struct esas2r_mem_desc *mem_desc,
+				 u32 align)
+{
+	mem_desc->esas2r_param = mem_desc->size + align;
+	mem_desc->virt_addr = NULL;
+	mem_desc->phys_addr = 0;
+	mem_desc->esas2r_data = dma_alloc_coherent(&a->pcid->dev,
+						   (size_t)mem_desc->
+						   esas2r_param,
+						   (dma_addr_t *)&mem_desc->
+						   phys_addr,
+						   GFP_KERNEL);
+
+	if (mem_desc->esas2r_data == NULL) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "failed to allocate %lu bytes of consistent memory!",
+			   (long
+			    unsigned
+			    int)mem_desc->esas2r_param);
+		return false;
+	}
+
+	mem_desc->virt_addr = PTR_ALIGN(mem_desc->esas2r_data, align);
+	mem_desc->phys_addr = ALIGN(mem_desc->phys_addr, align);
+	memset(mem_desc->virt_addr, 0, mem_desc->size);
+	return true;
+}
+
+static void esas2r_initmem_free(struct esas2r_adapter *a,
+				struct esas2r_mem_desc *mem_desc)
+{
+	if (mem_desc->virt_addr == NULL)
+		return;
+
+	/*
+	 * Careful!  phys_addr and virt_addr may have been adjusted from the
+	 * original allocation in order to return the desired alignment.  That
+	 * means we have to use the original address (in esas2r_data) and size
+	 * (esas2r_param) and calculate the original physical address based on
+	 * the difference between the requested and actual allocation size.
+	 */
+	if (mem_desc->phys_addr) {
+		int unalign = ((u8 *)mem_desc->virt_addr) -
+			      ((u8 *)mem_desc->esas2r_data);
+
+		dma_free_coherent(&a->pcid->dev,
+				  (size_t)mem_desc->esas2r_param,
+				  mem_desc->esas2r_data,
+				  (dma_addr_t)(mem_desc->phys_addr - unalign));
+	} else {
+		kfree(mem_desc->esas2r_data);
+	}
+
+	mem_desc->virt_addr = NULL;
+}
+
+static bool alloc_vda_req(struct esas2r_adapter *a,
+			  struct esas2r_request *rq)
+{
+	struct esas2r_mem_desc *memdesc = kzalloc(
+		sizeof(struct esas2r_mem_desc), GFP_KERNEL);
+
+	if (memdesc == NULL) {
+		esas2r_hdebug("could not alloc mem for vda request memdesc\n");
+		return false;
+	}
+
+	memdesc->size = sizeof(union atto_vda_req) +
+			ESAS2R_DATA_BUF_LEN;
+
+	if (!esas2r_initmem_alloc(a, memdesc, 256)) {
+		esas2r_hdebug("could not alloc mem for vda request\n");
+		kfree(memdesc);
+		return false;
+	}
+
+	a->num_vrqs++;
+	list_add(&memdesc->next_desc, &a->vrq_mds_head);
+
+	rq->vrq_md = memdesc;
+	rq->vrq = (union atto_vda_req *)memdesc->virt_addr;
+	rq->vrq->scsi.handle = a->num_vrqs;
+
+	return true;
+}
+
+static void esas2r_unmap_regions(struct esas2r_adapter *a)
+{
+	if (a->regs)
+		iounmap((void __iomem *)a->regs);
+
+	a->regs = NULL;
+
+	pci_release_region(a->pcid, 2);
+
+	if (a->data_window)
+		iounmap((void __iomem *)a->data_window);
+
+	a->data_window = NULL;
+
+	pci_release_region(a->pcid, 0);
+}
+
+static int esas2r_map_regions(struct esas2r_adapter *a)
+{
+	int error;
+
+	a->regs = NULL;
+	a->data_window = NULL;
+
+	error = pci_request_region(a->pcid, 2, a->name);
+	if (error != 0) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "pci_request_region(2) failed, error %d",
+			   error);
+
+		return error;
+	}
+
+	a->regs = (void __force *)ioremap(pci_resource_start(a->pcid, 2),
+					  pci_resource_len(a->pcid, 2));
+	if (a->regs == NULL) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "ioremap failed for regs mem region\n");
+		pci_release_region(a->pcid, 2);
+		return -EFAULT;
+	}
+
+	error = pci_request_region(a->pcid, 0, a->name);
+	if (error != 0) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "pci_request_region(2) failed, error %d",
+			   error);
+		esas2r_unmap_regions(a);
+		return error;
+	}
+
+	a->data_window = (void __force *)ioremap(pci_resource_start(a->pcid,
+								    0),
+						 pci_resource_len(a->pcid, 0));
+	if (a->data_window == NULL) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "ioremap failed for data_window mem region\n");
+		esas2r_unmap_regions(a);
+		return -EFAULT;
+	}
+
+	return 0;
+}
+
+static void esas2r_setup_interrupts(struct esas2r_adapter *a, int intr_mode)
+{
+	int i;
+
+	/* Set up interrupt mode based on the requested value */
+	switch (intr_mode) {
+	case INTR_MODE_LEGACY:
+use_legacy_interrupts:
+		a->intr_mode = INTR_MODE_LEGACY;
+		break;
+
+	case INTR_MODE_MSI:
+		i = pci_enable_msi(a->pcid);
+		if (i != 0) {
+			esas2r_log(ESAS2R_LOG_WARN,
+				   "failed to enable MSI for adapter %d, "
+				   "falling back to legacy interrupts "
+				   "(err=%d)", a->index,
+				   i);
+			goto use_legacy_interrupts;
+		}
+		a->intr_mode = INTR_MODE_MSI;
+		esas2r_lock_set_flags(&a->flags2, AF2_MSI_ENABLED);
+		break;
+
+
+	default:
+		esas2r_log(ESAS2R_LOG_WARN,
+			   "unknown interrupt_mode %d requested, "
+			   "falling back to legacy interrupt",
+			   interrupt_mode);
+		goto use_legacy_interrupts;
+	}
+}
+
+static void esas2r_claim_interrupts(struct esas2r_adapter *a)
+{
+	unsigned long flags = IRQF_DISABLED;
+
+	if (a->intr_mode == INTR_MODE_LEGACY)
+		flags |= IRQF_SHARED;
+
+	esas2r_log(ESAS2R_LOG_INFO,
+		   "esas2r_claim_interrupts irq=%d (%p, %s, %x)",
+		   a->pcid->irq, a, a->name, flags);
+
+	if (request_irq(a->pcid->irq,
+			(a->intr_mode ==
+			 INTR_MODE_LEGACY) ? esas2r_interrupt :
+			esas2r_msi_interrupt,
+			flags,
+			a->name,
+			a)) {
+		esas2r_log(ESAS2R_LOG_CRIT, "unable to request IRQ %02X",
+			   a->pcid->irq);
+		return;
+	}
+
+	esas2r_lock_set_flags(&a->flags2, AF2_IRQ_CLAIMED);
+	esas2r_log(ESAS2R_LOG_INFO,
+		   "claimed IRQ %d flags: 0x%lx",
+		   a->pcid->irq, flags);
+}
+
+int esas2r_init_adapter(struct Scsi_Host *host, struct pci_dev *pcid,
+			int index)
+{
+	struct esas2r_adapter *a;
+	u64 bus_addr = 0;
+	int i;
+	void *next_uncached;
+	struct esas2r_request *first_request, *last_request;
+
+	if (index >= MAX_ADAPTERS) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "tried to init invalid adapter index %u!",
+			   index);
+		return 0;
+	}
+
+	if (esas2r_adapters[index]) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "tried to init existing adapter index %u!",
+			   index);
+		return 0;
+	}
+
+	a = (struct esas2r_adapter *)host->hostdata;
+	memset(a, 0, sizeof(struct esas2r_adapter));
+	a->pcid = pcid;
+	a->host = host;
+
+	if (sizeof(dma_addr_t) > 4) {
+		const uint64_t required_mask = dma_get_required_mask
+						       (&pcid->dev);
+		if (required_mask > DMA_BIT_MASK(32)
+		    && !pci_set_dma_mask(pcid, DMA_BIT_MASK(64))
+		    && !pci_set_consistent_dma_mask(pcid,
+						    DMA_BIT_MASK(64))) {
+			esas2r_log_dev(ESAS2R_LOG_INFO,
+				       &(a->pcid->dev),
+				       "64-bit PCI addressing enabled\n");
+		} else if (!pci_set_dma_mask(pcid, DMA_BIT_MASK(32))
+			   && !pci_set_consistent_dma_mask(pcid,
+							   DMA_BIT_MASK(32))) {
+			esas2r_log_dev(ESAS2R_LOG_INFO,
+				       &(a->pcid->dev),
+				       "32-bit PCI addressing enabled\n");
+		} else {
+			esas2r_log(ESAS2R_LOG_CRIT,
+				   "failed to set DMA mask");
+			esas2r_kill_adapter(index);
+			return 0;
+		}
+	} else {
+		if (!pci_set_dma_mask(pcid, DMA_BIT_MASK(32))
+		    && !pci_set_consistent_dma_mask(pcid,
+						    DMA_BIT_MASK(32))) {
+			esas2r_log_dev(ESAS2R_LOG_INFO,
+				       &(a->pcid->dev),
+				       "32-bit PCI addressing enabled\n");
+		} else {
+			esas2r_log(ESAS2R_LOG_CRIT,
+				   "failed to set DMA mask");
+			esas2r_kill_adapter(index);
+			return 0;
+		}
+	}
+	esas2r_adapters[index] = a;
+	sprintf(a->name, ESAS2R_DRVR_NAME "_%02d", index);
+	esas2r_debug("new adapter %p, name %s", a, a->name);
+	spin_lock_init(&a->request_lock);
+	spin_lock_init(&a->fw_event_lock);
+	sema_init(&a->fm_api_semaphore, 1);
+	sema_init(&a->fs_api_semaphore, 1);
+	sema_init(&a->nvram_semaphore, 1);
+
+	esas2r_fw_event_off(a);
+	snprintf(a->fw_event_q_name, ESAS2R_KOBJ_NAME_LEN, "esas2r/%d",
+		 a->index);
+	a->fw_event_q = create_singlethread_workqueue(a->fw_event_q_name);
+
+	init_waitqueue_head(&a->buffered_ioctl_waiter);
+	init_waitqueue_head(&a->nvram_waiter);
+	init_waitqueue_head(&a->fm_api_waiter);
+	init_waitqueue_head(&a->fs_api_waiter);
+	init_waitqueue_head(&a->vda_waiter);
+
+	INIT_LIST_HEAD(&a->general_req.req_list);
+	INIT_LIST_HEAD(&a->active_list);
+	INIT_LIST_HEAD(&a->defer_list);
+	INIT_LIST_HEAD(&a->free_sg_list_head);
+	INIT_LIST_HEAD(&a->avail_request);
+	INIT_LIST_HEAD(&a->vrq_mds_head);
+	INIT_LIST_HEAD(&a->fw_event_list);
+
+	first_request = (struct esas2r_request *)((u8 *)(a + 1));
+
+	for (last_request = first_request, i = 1; i < num_requests;
+	     last_request++, i++) {
+		INIT_LIST_HEAD(&last_request->req_list);
+		list_add_tail(&last_request->comp_list, &a->avail_request);
+		if (!alloc_vda_req(a, last_request)) {
+			esas2r_log(ESAS2R_LOG_CRIT,
+				   "failed to allocate a VDA request!");
+			esas2r_kill_adapter(index);
+			return 0;
+		}
+	}
+
+	esas2r_debug("requests: %p to %p (%d, %d)", first_request,
+		     last_request,
+		     sizeof(*first_request),
+		     num_requests);
+
+	if (esas2r_map_regions(a) != 0) {
+		esas2r_log(ESAS2R_LOG_CRIT, "could not map PCI regions!");
+		esas2r_kill_adapter(index);
+		return 0;
+	}
+
+	a->index = index;
+
+	/* interrupts will be disabled until we are done with init */
+	atomic_inc(&a->dis_ints_cnt);
+	atomic_inc(&a->disable_cnt);
+	a->flags |= AF_CHPRST_PENDING
+		    | AF_DISC_PENDING
+		    | AF_FIRST_INIT
+		    | AF_LEGACY_SGE_MODE;
+
+	a->init_msg = ESAS2R_INIT_MSG_START;
+	a->max_vdareq_size = 128;
+	a->build_sgl = esas2r_build_sg_list_sge;
+
+	esas2r_setup_interrupts(a, interrupt_mode);
+
+	a->uncached_size = esas2r_get_uncached_size(a);
+	a->uncached = dma_alloc_coherent(&pcid->dev,
+					 (size_t)a->uncached_size,
+					 (dma_addr_t *)&bus_addr,
+					 GFP_KERNEL);
+	if (a->uncached == NULL) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "failed to allocate %d bytes of consistent memory!",
+			   a->uncached_size);
+		esas2r_kill_adapter(index);
+		return 0;
+	}
+
+	a->uncached_phys = bus_addr;
+
+	esas2r_debug("%d bytes uncached memory allocated @ %p (%x:%x)",
+		     a->uncached_size,
+		     a->uncached,
+		     upper_32_bits(bus_addr),
+		     lower_32_bits(bus_addr));
+	memset(a->uncached, 0, a->uncached_size);
+	next_uncached = a->uncached;
+
+	if (!esas2r_init_adapter_struct(a,
+					&next_uncached)) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "failed to initialize adapter structure (2)!");
+		esas2r_kill_adapter(index);
+		return 0;
+	}
+
+	tasklet_init(&a->tasklet,
+		     esas2r_adapter_tasklet,
+		     (unsigned long)a);
+
+	/*
+	 * Disable chip interrupts to prevent spurious interrupts
+	 * until we claim the IRQ.
+	 */
+	esas2r_disable_chip_interrupts(a);
+	esas2r_check_adapter(a);
+
+	if (!esas2r_init_adapter_hw(a, true))
+		esas2r_log(ESAS2R_LOG_CRIT, "failed to initialize hardware!");
+	else
+		esas2r_debug("esas2r_init_adapter ok");
+
+	esas2r_claim_interrupts(a);
+
+	if (a->flags2 & AF2_IRQ_CLAIMED)
+		esas2r_enable_chip_interrupts(a);
+
+	esas2r_lock_set_flags(&a->flags2, AF2_INIT_DONE);
+	if (!(a->flags & AF_DEGRADED_MODE))
+		esas2r_kickoff_timer(a);
+	esas2r_debug("esas2r_init_adapter done for %p (%d)",
+		     a, a->disable_cnt);
+
+	return 1;
+}
+
+static void esas2r_adapter_power_down(struct esas2r_adapter *a,
+				      int power_management)
+{
+	struct esas2r_mem_desc *memdesc, *next;
+
+	if ((a->flags2 & AF2_INIT_DONE)
+	    &&  (!(a->flags & AF_DEGRADED_MODE))) {
+		if (!power_management) {
+			del_timer_sync(&a->timer);
+			tasklet_kill(&a->tasklet);
+		}
+		esas2r_power_down(a);
+
+		/*
+		 * There are versions of firmware that do not handle the sync
+		 * cache command correctly.  Stall here to ensure that the
+		 * cache is lazily flushed.
+		 */
+		mdelay(500);
+		esas2r_debug("chip halted");
+	}
+
+	/* Remove sysfs binary files */
+	if (a->sysfs_fw_created) {
+		sysfs_remove_bin_file(&a->host->shost_dev.kobj, &bin_attr_fw);
+		a->sysfs_fw_created = 0;
+	}
+
+	if (a->sysfs_fs_created) {
+		sysfs_remove_bin_file(&a->host->shost_dev.kobj, &bin_attr_fs);
+		a->sysfs_fs_created = 0;
+	}
+
+	if (a->sysfs_vda_created) {
+		sysfs_remove_bin_file(&a->host->shost_dev.kobj, &bin_attr_vda);
+		a->sysfs_vda_created = 0;
+	}
+
+	if (a->sysfs_hw_created) {
+		sysfs_remove_bin_file(&a->host->shost_dev.kobj, &bin_attr_hw);
+		a->sysfs_hw_created = 0;
+	}
+
+	if (a->sysfs_live_nvram_created) {
+		sysfs_remove_bin_file(&a->host->shost_dev.kobj,
+				      &bin_attr_live_nvram);
+		a->sysfs_live_nvram_created = 0;
+	}
+
+	if (a->sysfs_default_nvram_created) {
+		sysfs_remove_bin_file(&a->host->shost_dev.kobj,
+				      &bin_attr_default_nvram);
+		a->sysfs_default_nvram_created = 0;
+	}
+
+	/* Clean up interrupts */
+	if (a->flags2 & AF2_IRQ_CLAIMED) {
+		esas2r_log_dev(ESAS2R_LOG_INFO,
+			       &(a->pcid->dev),
+			       "free_irq(%d) called", a->pcid->irq);
+
+		free_irq(a->pcid->irq, a);
+		esas2r_debug("IRQ released");
+		esas2r_lock_clear_flags(&a->flags2, AF2_IRQ_CLAIMED);
+	}
+
+	if (a->flags2 & AF2_MSI_ENABLED) {
+		pci_disable_msi(a->pcid);
+		esas2r_lock_clear_flags(&a->flags2, AF2_MSI_ENABLED);
+		esas2r_debug("MSI disabled");
+	}
+
+	if (a->inbound_list_md.virt_addr)
+		esas2r_initmem_free(a, &a->inbound_list_md);
+
+	if (a->outbound_list_md.virt_addr)
+		esas2r_initmem_free(a, &a->outbound_list_md);
+
+	list_for_each_entry_safe(memdesc, next, &a->free_sg_list_head,
+				 next_desc) {
+		esas2r_initmem_free(a, memdesc);
+	}
+
+	/* Following frees everything allocated via alloc_vda_req */
+	list_for_each_entry_safe(memdesc, next, &a->vrq_mds_head, next_desc) {
+		esas2r_initmem_free(a, memdesc);
+		list_del(&memdesc->next_desc);
+		kfree(memdesc);
+	}
+
+	kfree(a->first_ae_req);
+	a->first_ae_req = NULL;
+
+	kfree(a->sg_list_mds);
+	a->sg_list_mds = NULL;
+
+	kfree(a->req_table);
+	a->req_table = NULL;
+
+	if (a->regs) {
+		esas2r_unmap_regions(a);
+		a->regs = NULL;
+		a->data_window = NULL;
+		esas2r_debug("regions unmapped");
+	}
+}
+
+/* Release/free allocated resources for specified adapters. */
+void esas2r_kill_adapter(int i)
+{
+	struct esas2r_adapter *a = esas2r_adapters[i];
+
+	if (a) {
+		unsigned long flags;
+		struct workqueue_struct *wq;
+		esas2r_debug("killing adapter %p [%d] ", a, i);
+		esas2r_fw_event_off(a);
+		esas2r_adapter_power_down(a, 0);
+		if (esas2r_buffered_ioctl &&
+		    (a->pcid == esas2r_buffered_ioctl_pcid)) {
+			dma_free_coherent(&a->pcid->dev,
+					  (size_t)esas2r_buffered_ioctl_size,
+					  esas2r_buffered_ioctl,
+					  esas2r_buffered_ioctl_addr);
+			esas2r_buffered_ioctl = NULL;
+		}
+
+		if (a->vda_buffer) {
+			dma_free_coherent(&a->pcid->dev,
+					  (size_t)VDA_MAX_BUFFER_SIZE,
+					  a->vda_buffer,
+					  (dma_addr_t)a->ppvda_buffer);
+			a->vda_buffer = NULL;
+		}
+		if (a->fs_api_buffer) {
+			dma_free_coherent(&a->pcid->dev,
+					  (size_t)a->fs_api_buffer_size,
+					  a->fs_api_buffer,
+					  (dma_addr_t)a->ppfs_api_buffer);
+			a->fs_api_buffer = NULL;
+		}
+
+		kfree(a->local_atto_ioctl);
+		a->local_atto_ioctl = NULL;
+
+		spin_lock_irqsave(&a->fw_event_lock, flags);
+		wq = a->fw_event_q;
+		a->fw_event_q = NULL;
+		spin_unlock_irqrestore(&a->fw_event_lock, flags);
+		if (wq)
+			destroy_workqueue(wq);
+
+		if (a->uncached) {
+			dma_free_coherent(&a->pcid->dev,
+					  (size_t)a->uncached_size,
+					  a->uncached,
+					  (dma_addr_t)a->uncached_phys);
+			a->uncached = NULL;
+			esas2r_debug("uncached area freed");
+		}
+
+		esas2r_log_dev(ESAS2R_LOG_INFO,
+			       &(a->pcid->dev),
+			       "pci_disable_device() called.  msix_enabled: %d "
+			       "msi_enabled: %d irq: %d pin: %d",
+			       a->pcid->msix_enabled,
+			       a->pcid->msi_enabled,
+			       a->pcid->irq,
+			       a->pcid->pin);
+
+		esas2r_log_dev(ESAS2R_LOG_INFO,
+			       &(a->pcid->dev),
+			       "before pci_disable_device() enable_cnt: %d",
+			       a->pcid->enable_cnt.counter);
+
+		pci_disable_device(a->pcid);
+		esas2r_log_dev(ESAS2R_LOG_INFO,
+			       &(a->pcid->dev),
+			       "after pci_disable_device() enable_cnt: %d",
+			       a->pcid->enable_cnt.counter);
+
+		esas2r_log_dev(ESAS2R_LOG_INFO,
+			       &(a->pcid->dev),
+			       "pci_set_drv_data(%p, NULL) called",
+			       a->pcid);
+
+		pci_set_drvdata(a->pcid, NULL);
+		esas2r_adapters[i] = NULL;
+
+		if (a->flags2 & AF2_INIT_DONE) {
+			esas2r_lock_clear_flags(&a->flags2,
+						AF2_INIT_DONE);
+
+			esas2r_lock_set_flags(&a->flags,
+					      AF_DEGRADED_MODE);
+
+			esas2r_log_dev(ESAS2R_LOG_INFO,
+				       &(a->host->shost_gendev),
+				       "scsi_remove_host() called");
+
+			scsi_remove_host(a->host);
+
+			esas2r_log_dev(ESAS2R_LOG_INFO,
+				       &(a->host->shost_gendev),
+				       "scsi_host_put() called");
+
+			scsi_host_put(a->host);
+		}
+	}
+}
+
+int esas2r_cleanup(struct Scsi_Host *host)
+{
+	struct esas2r_adapter *a = (struct esas2r_adapter *)host->hostdata;
+	int index;
+
+	if (host == NULL) {
+		int i;
+
+		esas2r_debug("esas2r_cleanup everything");
+		for (i = 0; i < MAX_ADAPTERS; i++)
+			esas2r_kill_adapter(i);
+		return -1;
+	}
+
+	esas2r_debug("esas2r_cleanup called for host %p", host);
+	index = a->index;
+	esas2r_kill_adapter(index);
+	return index;
+}
+
+int esas2r_suspend(struct pci_dev *pdev, pm_message_t state)
+{
+	struct Scsi_Host *host = pci_get_drvdata(pdev);
+	u32 device_state;
+	struct esas2r_adapter *a = (struct esas2r_adapter *)host->hostdata;
+
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev), "suspending adapter()");
+	if (!a)
+		return -ENODEV;
+
+	esas2r_adapter_power_down(a, 1);
+	device_state = pci_choose_state(pdev, state);
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev),
+		       "pci_save_state() called");
+	pci_save_state(pdev);
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev),
+		       "pci_disable_device() called");
+	pci_disable_device(pdev);
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev),
+		       "pci_set_power_state() called");
+	pci_set_power_state(pdev, device_state);
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev), "esas2r_suspend(): 0");
+	return 0;
+}
+
+int esas2r_resume(struct pci_dev *pdev)
+{
+	struct Scsi_Host *host = pci_get_drvdata(pdev);
+	struct esas2r_adapter *a = (struct esas2r_adapter *)host->hostdata;
+	int rez;
+
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev), "resuming adapter()");
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev),
+		       "pci_set_power_state(PCI_D0) "
+		       "called");
+	pci_set_power_state(pdev, PCI_D0);
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev),
+		       "pci_enable_wake(PCI_D0, 0) "
+		       "called");
+	pci_enable_wake(pdev, PCI_D0, 0);
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev),
+		       "pci_restore_state() called");
+	pci_restore_state(pdev);
+	esas2r_log_dev(ESAS2R_LOG_INFO, &(pdev->dev),
+		       "pci_enable_device() called");
+	rez = pci_enable_device(pdev);
+	pci_set_master(pdev);
+
+	if (!a) {
+		rez = -ENODEV;
+		goto error_exit;
+	}
+
+	if (esas2r_map_regions(a) != 0) {
+		esas2r_log(ESAS2R_LOG_CRIT, "could not re-map PCI regions!");
+		rez = -ENOMEM;
+		goto error_exit;
+	}
+
+	/* Set up interupt mode */
+	esas2r_setup_interrupts(a, a->intr_mode);
+
+	/*
+	 * Disable chip interrupts to prevent spurious interrupts until we
+	 * claim the IRQ.
+	 */
+	esas2r_disable_chip_interrupts(a);
+	if (!esas2r_power_up(a, true)) {
+		esas2r_debug("yikes, esas2r_power_up failed");
+		rez = -ENOMEM;
+		goto error_exit;
+	}
+
+	esas2r_claim_interrupts(a);
+
+	if (a->flags2 & AF2_IRQ_CLAIMED) {
+		/*
+		 * Now that system interrupt(s) are claimed, we can enable
+		 * chip interrupts.
+		 */
+		esas2r_enable_chip_interrupts(a);
+		esas2r_kickoff_timer(a);
+	} else {
+		esas2r_debug("yikes, unable to claim IRQ");
+		esas2r_log(ESAS2R_LOG_CRIT, "could not re-claim IRQ!");
+		rez = -ENOMEM;
+		goto error_exit;
+	}
+
+error_exit:
+	esas2r_log_dev(ESAS2R_LOG_CRIT, &(pdev->dev), "esas2r_resume(): %d",
+		       rez);
+	return rez;
+}
+
+bool esas2r_set_degraded_mode(struct esas2r_adapter *a, char *error_str)
+{
+	esas2r_lock_set_flags(&a->flags, AF_DEGRADED_MODE);
+	esas2r_log(ESAS2R_LOG_CRIT,
+		   "setting adapter to degraded mode: %s\n", error_str);
+	return false;
+}
+
+u32 esas2r_get_uncached_size(struct esas2r_adapter *a)
+{
+	return sizeof(struct esas2r_sas_nvram)
+	       + ALIGN(ESAS2R_DISC_BUF_LEN, 8)
+	       + ALIGN(sizeof(u32), 8) /* outbound list copy pointer */
+	       + 8
+	       + (num_sg_lists * (u16)sgl_page_size)
+	       + ALIGN((num_requests + num_ae_requests + 1 +
+			ESAS2R_LIST_EXTRA) *
+		       sizeof(struct esas2r_inbound_list_source_entry),
+		       8)
+	       + ALIGN((num_requests + num_ae_requests + 1 +
+			ESAS2R_LIST_EXTRA) *
+		       sizeof(struct atto_vda_ob_rsp), 8)
+	       + 256; /* VDA request and buffer align */
+}
+
+static void esas2r_init_pci_cfg_space(struct esas2r_adapter *a)
+{
+	int pcie_cap_reg;
+
+	pcie_cap_reg = pci_find_capability(a->pcid, PCI_CAP_ID_EXP);
+	if (0xffff && pcie_cap_reg) {
+		u16 devcontrol;
+
+		pci_read_config_word(a->pcid, pcie_cap_reg + PCI_EXP_DEVCTL,
+				     &devcontrol);
+
+		if ((devcontrol & PCI_EXP_DEVCTL_READRQ) > 0x2000) {
+			esas2r_log(ESAS2R_LOG_INFO,
+				   "max read request size > 512B");
+
+			devcontrol &= ~PCI_EXP_DEVCTL_READRQ;
+			devcontrol |= 0x2000;
+			pci_write_config_word(a->pcid,
+					      pcie_cap_reg + PCI_EXP_DEVCTL,
+					      devcontrol);
+		}
+	}
+}
+
+/*
+ * Determine the organization of the uncached data area and
+ * finish initializing the adapter structure
+ */
+bool esas2r_init_adapter_struct(struct esas2r_adapter *a,
+				void **uncached_area)
+{
+	u32 i;
+	u8 *high;
+	struct esas2r_inbound_list_source_entry *element;
+	struct esas2r_request *rq;
+	struct esas2r_mem_desc *sgl;
+
+	spin_lock_init(&a->sg_list_lock);
+	spin_lock_init(&a->mem_lock);
+	spin_lock_init(&a->queue_lock);
+
+	a->targetdb_end = &a->targetdb[ESAS2R_MAX_TARGETS];
+
+	if (!alloc_vda_req(a, &a->general_req)) {
+		esas2r_hdebug(
+			"failed to allocate a VDA request for the general req!");
+		return false;
+	}
+
+	/* allocate requests for asynchronous events */
+	a->first_ae_req =
+		kzalloc(num_ae_requests * sizeof(struct esas2r_request),
+			GFP_KERNEL);
+
+	if (a->first_ae_req == NULL) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "failed to allocate memory for asynchronous events");
+		return false;
+	}
+
+	/* allocate the S/G list memory descriptors */
+	a->sg_list_mds = kzalloc(
+		num_sg_lists * sizeof(struct esas2r_mem_desc), GFP_KERNEL);
+
+	if (a->sg_list_mds == NULL) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "failed to allocate memory for s/g list descriptors");
+		return false;
+	}
+
+	/* allocate the request table */
+	a->req_table =
+		kzalloc((num_requests + num_ae_requests +
+			 1) * sizeof(struct esas2r_request *), GFP_KERNEL);
+
+	if (a->req_table == NULL) {
+		esas2r_log(ESAS2R_LOG_CRIT,
+			   "failed to allocate memory for the request table");
+		return false;
+	}
+
+	/* initialize PCI configuration space */
+	esas2r_init_pci_cfg_space(a);
+
+	/*
+	 * the thunder_stream boards all have a serial flash part that has a
+	 * different base address on the AHB bus.
+	 */
+	if ((a->pcid->subsystem_vendor == ATTO_VENDOR_ID)
+	    && (a->pcid->subsystem_device & ATTO_SSDID_TBT))
+		a->flags2 |= AF2_THUNDERBOLT;
+
+	if (a->flags2 & AF2_THUNDERBOLT)
+		a->flags2 |= AF2_SERIAL_FLASH;
+
+	if (a->pcid->subsystem_device == ATTO_TLSH_1068)
+		a->flags2 |= AF2_THUNDERLINK;
+
+	/* Uncached Area */
+	high = (u8 *)*uncached_area;
+
+	/* initialize the scatter/gather table pages */
+
+	for (i = 0, sgl = a->sg_list_mds; i < num_sg_lists; i++, sgl++) {
+		sgl->size = sgl_page_size;
+
+		list_add_tail(&sgl->next_desc, &a->free_sg_list_head);
+
+		if (!esas2r_initmem_alloc(a, sgl, ESAS2R_SGL_ALIGN)) {
+			/* Allow the driver to load if the minimum count met. */
+			if (i < NUM_SGL_MIN)
+				return false;
+			break;
+		}
+	}
+
+	/* compute the size of the lists */
+	a->list_size = num_requests + ESAS2R_LIST_EXTRA;
+
+	/* allocate the inbound list */
+	a->inbound_list_md.size = a->list_size *
+				  sizeof(struct
+					 esas2r_inbound_list_source_entry);
+
+	if (!esas2r_initmem_alloc(a, &a->inbound_list_md, ESAS2R_LIST_ALIGN)) {
+		esas2r_hdebug("failed to allocate IB list");
+		return false;
+	}
+
+	/* allocate the outbound list */
+	a->outbound_list_md.size = a->list_size *
+				   sizeof(struct atto_vda_ob_rsp);
+
+	if (!esas2r_initmem_alloc(a, &a->outbound_list_md,
+				  ESAS2R_LIST_ALIGN)) {
+		esas2r_hdebug("failed to allocate IB list");
+		return false;
+	}
+
+	/* allocate the NVRAM structure */
+	a->nvram = (struct esas2r_sas_nvram *)high;
+	high += sizeof(struct esas2r_sas_nvram);
+
+	/* allocate the discovery buffer */
+	a->disc_buffer = high;
+	high += ESAS2R_DISC_BUF_LEN;
+	high = PTR_ALIGN(high, 8);
+
+	/* allocate the outbound list copy pointer */
+	a->outbound_copy = (u32 volatile *)high;
+	high += sizeof(u32);
+
+	if (!(a->flags & AF_NVR_VALID))
+		esas2r_nvram_set_defaults(a);
+
+	/* update the caller's uncached memory area pointer */
+	*uncached_area = (void *)high;
+
+	/* initialize the allocated memory */
+	if (a->flags & AF_FIRST_INIT) {
+		memset(a->req_table, 0,
+		       (num_requests + num_ae_requests +
+			1) * sizeof(struct esas2r_request *));
+
+		esas2r_targ_db_initialize(a);
+
+		/* prime parts of the inbound list */
+		element =
+			(struct esas2r_inbound_list_source_entry *)a->
+			inbound_list_md.
+			virt_addr;
+
+		for (i = 0; i < a->list_size; i++) {
+			element->address = 0;
+			element->reserved = 0;
+			element->length = cpu_to_le32(HWILSE_INTERFACE_F0
+						      | (sizeof(union
+								atto_vda_req)
+							 /
+							 sizeof(u32)));
+			element++;
+		}
+
+		/* init the AE requests */
+		for (rq = a->first_ae_req, i = 0; i < num_ae_requests; rq++,
+		     i++) {
+			INIT_LIST_HEAD(&rq->req_list);
+			if (!alloc_vda_req(a, rq)) {
+				esas2r_hdebug(
+					"failed to allocate a VDA request!");
+				return false;
+			}
+
+			esas2r_rq_init_request(rq, a);
+
+			/* override the completion function */
+			rq->comp_cb = esas2r_ae_complete;
+		}
+	}
+
+	return true;
+}
+
+/* This code will verify that the chip is operational. */
+bool esas2r_check_adapter(struct esas2r_adapter *a)
+{
+	u32 starttime;
+	u32 doorbell;
+	u64 ppaddr;
+	u32 dw;
+
+	/*
+	 * if the chip reset detected flag is set, we can bypass a bunch of
+	 * stuff.
+	 */
+	if (a->flags & AF_CHPRST_DETECTED)
+		goto skip_chip_reset;
+
+	/*
+	 * BEFORE WE DO ANYTHING, disable the chip interrupts!  the boot driver
+	 * may have left them enabled or we may be recovering from a fault.
+	 */
+	esas2r_write_register_dword(a, MU_INT_MASK_OUT, ESAS2R_INT_DIS_MASK);
+	esas2r_flush_register_dword(a, MU_INT_MASK_OUT);
+
+	/*
+	 * wait for the firmware to become ready by forcing an interrupt and
+	 * waiting for a response.
+	 */
+	starttime = jiffies_to_msecs(jiffies);
+
+	while (true) {
+		esas2r_force_interrupt(a);
+		doorbell = esas2r_read_register_dword(a, MU_DOORBELL_OUT);
+		if (doorbell == 0xFFFFFFFF) {
+			/*
+			 * Give the firmware up to two seconds to enable
+			 * register access after a reset.
+			 */
+			if ((jiffies_to_msecs(jiffies) - starttime) > 2000)
+				return esas2r_set_degraded_mode(a,
+								"unable to access registers");
+		} else if (doorbell & DRBL_FORCE_INT) {
+			u32 ver = (doorbell & DRBL_FW_VER_MSK);
+
+			/*
+			 * This driver supports version 0 and version 1 of
+			 * the API
+			 */
+			esas2r_write_register_dword(a, MU_DOORBELL_OUT,
+						    doorbell);
+
+			if (ver == DRBL_FW_VER_0) {
+				esas2r_lock_set_flags(&a->flags,
+						      AF_LEGACY_SGE_MODE);
+
+				a->max_vdareq_size = 128;
+				a->build_sgl = esas2r_build_sg_list_sge;
+			} else if (ver == DRBL_FW_VER_1) {
+				esas2r_lock_clear_flags(&a->flags,
+							AF_LEGACY_SGE_MODE);
+
+				a->max_vdareq_size = 1024;
+				a->build_sgl = esas2r_build_sg_list_prd;
+			} else {
+				return esas2r_set_degraded_mode(a,
+								"unknown firmware version");
+			}
+			break;
+		}
+
+		schedule_timeout_interruptible(msecs_to_jiffies(100));
+
+		if ((jiffies_to_msecs(jiffies) - starttime) > 180000) {
+			esas2r_hdebug("FW ready TMO");
+			esas2r_bugon();
+
+			return esas2r_set_degraded_mode(a,
+							"firmware start has timed out");
+		}
+	}
+
+	/* purge any asynchronous events since we will repost them later */
+	esas2r_write_register_dword(a, MU_DOORBELL_IN, DRBL_MSG_IFC_DOWN);
+	starttime = jiffies_to_msecs(jiffies);
+
+	while (true) {
+		doorbell = esas2r_read_register_dword(a, MU_DOORBELL_OUT);
+		if (doorbell & DRBL_MSG_IFC_DOWN) {
+			esas2r_write_register_dword(a, MU_DOORBELL_OUT,
+						    doorbell);
+			break;
+		}
+
+		schedule_timeout_interruptible(msecs_to_jiffies(50));
+
+		if ((jiffies_to_msecs(jiffies) - starttime) > 3000) {
+			esas2r_hdebug("timeout waiting for interface down");
+			break;
+		}
+	}
+skip_chip_reset:
+	/*
+	 * first things first, before we go changing any of these registers
+	 * disable the communication lists.
+	 */
+	dw = esas2r_read_register_dword(a, MU_IN_LIST_CONFIG);
+	dw &= ~MU_ILC_ENABLE;
+	esas2r_write_register_dword(a, MU_IN_LIST_CONFIG, dw);
+	dw = esas2r_read_register_dword(a, MU_OUT_LIST_CONFIG);
+	dw &= ~MU_OLC_ENABLE;
+	esas2r_write_register_dword(a, MU_OUT_LIST_CONFIG, dw);
+
+	/* configure the communication list addresses */
+	ppaddr = a->inbound_list_md.phys_addr;
+	esas2r_write_register_dword(a, MU_IN_LIST_ADDR_LO,
+				    lower_32_bits(ppaddr));
+	esas2r_write_register_dword(a, MU_IN_LIST_ADDR_HI,
+				    upper_32_bits(ppaddr));
+	ppaddr = a->outbound_list_md.phys_addr;
+	esas2r_write_register_dword(a, MU_OUT_LIST_ADDR_LO,
+				    lower_32_bits(ppaddr));
+	esas2r_write_register_dword(a, MU_OUT_LIST_ADDR_HI,
+				    upper_32_bits(ppaddr));
+	ppaddr = a->uncached_phys +
+		 ((u8 *)a->outbound_copy - a->uncached);
+	esas2r_write_register_dword(a, MU_OUT_LIST_COPY_PTR_LO,
+				    lower_32_bits(ppaddr));
+	esas2r_write_register_dword(a, MU_OUT_LIST_COPY_PTR_HI,
+				    upper_32_bits(ppaddr));
+
+	/* reset the read and write pointers */
+	*a->outbound_copy =
+		a->last_write =
+			a->last_read = a->list_size - 1;
+	esas2r_lock_set_flags(&a->flags, AF_COMM_LIST_TOGGLE);
+	esas2r_write_register_dword(a, MU_IN_LIST_WRITE, MU_ILW_TOGGLE |
+				    a->last_write);
+	esas2r_write_register_dword(a, MU_OUT_LIST_COPY, MU_OLC_TOGGLE |
+				    a->last_write);
+	esas2r_write_register_dword(a, MU_IN_LIST_READ, MU_ILR_TOGGLE |
+				    a->last_write);
+	esas2r_write_register_dword(a, MU_OUT_LIST_WRITE,
+				    MU_OLW_TOGGLE | a->last_write);
+
+	/* configure the interface select fields */
+	dw = esas2r_read_register_dword(a, MU_IN_LIST_IFC_CONFIG);
+	dw &= ~(MU_ILIC_LIST | MU_ILIC_DEST);
+	esas2r_write_register_dword(a, MU_IN_LIST_IFC_CONFIG,
+				    (dw | MU_ILIC_LIST_F0 | MU_ILIC_DEST_DDR));
+	dw = esas2r_read_register_dword(a, MU_OUT_LIST_IFC_CONFIG);
+	dw &= ~(MU_OLIC_LIST | MU_OLIC_SOURCE);
+	esas2r_write_register_dword(a, MU_OUT_LIST_IFC_CONFIG,
+				    (dw | MU_OLIC_LIST_F0 |
+				     MU_OLIC_SOURCE_DDR));
+
+	/* finish configuring the communication lists */
+	dw = esas2r_read_register_dword(a, MU_IN_LIST_CONFIG);
+	dw &= ~(MU_ILC_ENTRY_MASK | MU_ILC_NUMBER_MASK);
+	dw |= MU_ILC_ENTRY_4_DW | MU_ILC_DYNAMIC_SRC
+	      | (a->list_size << MU_ILC_NUMBER_SHIFT);
+	esas2r_write_register_dword(a, MU_IN_LIST_CONFIG, dw);
+	dw = esas2r_read_register_dword(a, MU_OUT_LIST_CONFIG);
+	dw &= ~(MU_OLC_ENTRY_MASK | MU_OLC_NUMBER_MASK);
+	dw |= MU_OLC_ENTRY_4_DW | (a->list_size << MU_OLC_NUMBER_SHIFT);
+	esas2r_write_register_dword(a, MU_OUT_LIST_CONFIG, dw);
+
+	/*
+	 * notify the firmware that we're done setting up the communication
+	 * list registers.  wait here until the firmware is done configuring
+	 * its lists.  it will signal that it is done by enabling the lists.
+	 */
+	esas2r_write_register_dword(a, MU_DOORBELL_IN, DRBL_MSG_IFC_INIT);
+	starttime = jiffies_to_msecs(jiffies);
+
+	while (true) {
+		doorbell = esas2r_read_register_dword(a, MU_DOORBELL_OUT);
+		if (doorbell & DRBL_MSG_IFC_INIT) {
+			esas2r_write_register_dword(a, MU_DOORBELL_OUT,
+						    doorbell);
+			break;
+		}
+
+		schedule_timeout_interruptible(msecs_to_jiffies(100));
+
+		if ((jiffies_to_msecs(jiffies) - starttime) > 3000) {
+			esas2r_hdebug(
+				"timeout waiting for communication list init");
+			esas2r_bugon();
+			return esas2r_set_degraded_mode(a,
+							"timeout waiting for communication list init");
+		}
+	}
+
+	/*
+	 * flag whether the firmware supports the power down doorbell.  we
+	 * determine this by reading the inbound doorbell enable mask.
+	 */
+	doorbell = esas2r_read_register_dword(a, MU_DOORBELL_IN_ENB);
+	if (doorbell & DRBL_POWER_DOWN)
+		esas2r_lock_set_flags(&a->flags2, AF2_VDA_POWER_DOWN);
+	else
+		esas2r_lock_clear_flags(&a->flags2, AF2_VDA_POWER_DOWN);
+
+	/*
+	 * enable assertion of outbound queue and doorbell interrupts in the
+	 * main interrupt cause register.
+	 */
+	esas2r_write_register_dword(a, MU_OUT_LIST_INT_MASK, MU_OLIS_MASK);
+	esas2r_write_register_dword(a, MU_DOORBELL_OUT_ENB, DRBL_ENB_MASK);
+	return true;
+}
+
+/* Process the initialization message just completed and format the next one. */
+static bool esas2r_format_init_msg(struct esas2r_adapter *a,
+				   struct esas2r_request *rq)
+{
+	u32 msg = a->init_msg;
+	struct atto_vda_cfg_init *ci;
+
+	a->init_msg = 0;
+
+	switch (msg) {
+	case ESAS2R_INIT_MSG_START:
+	case ESAS2R_INIT_MSG_REINIT:
+	{
+		struct timeval now;
+		do_gettimeofday(&now);
+		esas2r_hdebug("CFG init");
+		esas2r_build_cfg_req(a,
+				     rq,
+				     VDA_CFG_INIT,
+				     0,
+				     NULL);
+		ci = (struct atto_vda_cfg_init *)&rq->vrq->cfg.data.init;
+		ci->sgl_page_size = sgl_page_size;
+		ci->epoch_time = now.tv_sec;
+		rq->flags |= RF_FAILURE_OK;
+		a->init_msg = ESAS2R_INIT_MSG_INIT;
+		break;
+	}
+
+	case ESAS2R_INIT_MSG_INIT:
+		if (rq->req_stat == RS_SUCCESS) {
+			u32 major;
+			u32 minor;
+
+			a->fw_version = le16_to_cpu(
+				rq->func_rsp.cfg_rsp.vda_version);
+			a->fw_build = rq->func_rsp.cfg_rsp.fw_build;
+			major = LOBYTE(rq->func_rsp.cfg_rsp.fw_release);
+			minor = HIBYTE(rq->func_rsp.cfg_rsp.fw_release);
+			a->fw_version += (major << 16) + (minor << 24);
+		} else {
+			esas2r_hdebug("FAILED");
+		}
+
+		/*
+		 * the 2.71 and earlier releases of R6xx firmware did not error
+		 * unsupported config requests correctly.
+		 */
+
+		if ((a->flags2 & AF2_THUNDERBOLT)
+		    || (be32_to_cpu(a->fw_version) >
+			be32_to_cpu(0x47020052))) {
+			esas2r_hdebug("CFG get init");
+			esas2r_build_cfg_req(a,
+					     rq,
+					     VDA_CFG_GET_INIT2,
+					     sizeof(struct atto_vda_cfg_init),
+					     NULL);
+
+			rq->vrq->cfg.sg_list_offset = offsetof(
+				struct atto_vda_cfg_req,
+				data.sge);
+			rq->vrq->cfg.data.prde.ctl_len =
+				cpu_to_le32(sizeof(struct atto_vda_cfg_init));
+			rq->vrq->cfg.data.prde.address = cpu_to_le64(
+				rq->vrq_md->phys_addr +
+				sizeof(union atto_vda_req));
+			rq->flags |= RF_FAILURE_OK;
+			a->init_msg = ESAS2R_INIT_MSG_GET_INIT;
+			break;
+		}
+
+	case ESAS2R_INIT_MSG_GET_INIT:
+		if (msg == ESAS2R_INIT_MSG_GET_INIT) {
+			ci = (struct atto_vda_cfg_init *)rq->data_buf;
+			if (rq->req_stat == RS_SUCCESS) {
+				a->num_targets_backend =
+					le32_to_cpu(ci->num_targets_backend);
+				a->ioctl_tunnel =
+					le32_to_cpu(ci->ioctl_tunnel);
+			} else {
+				esas2r_hdebug("FAILED");
+			}
+		}
+	/* fall through */
+
+	default:
+		rq->req_stat = RS_SUCCESS;
+		return false;
+	}
+	return true;
+}
+
+/*
+ * Perform initialization messages via the request queue.  Messages are
+ * performed with interrupts disabled.
+ */
+bool esas2r_init_msgs(struct esas2r_adapter *a)
+{
+	bool success = true;
+	struct esas2r_request *rq = &a->general_req;
+
+	esas2r_rq_init_request(rq, a);
+	rq->comp_cb = esas2r_dummy_complete;
+
+	if (a->init_msg == 0)
+		a->init_msg = ESAS2R_INIT_MSG_REINIT;
+
+	while (a->init_msg) {
+		if (esas2r_format_init_msg(a, rq)) {
+			unsigned long flags;
+			while (true) {
+				spin_lock_irqsave(&a->queue_lock, flags);
+				esas2r_start_vda_request(a, rq);
+				spin_unlock_irqrestore(&a->queue_lock, flags);
+				esas2r_wait_request(a, rq);
+				if (rq->req_stat != RS_PENDING)
+					break;
+			}
+		}
+
+		if (rq->req_stat == RS_SUCCESS
+		    || ((rq->flags & RF_FAILURE_OK)
+			&& rq->req_stat != RS_TIMEOUT))
+			continue;
+
+		esas2r_log(ESAS2R_LOG_CRIT, "init message %x failed (%x, %x)",
+			   a->init_msg, rq->req_stat, rq->flags);
+		a->init_msg = ESAS2R_INIT_MSG_START;
+		success = false;
+		break;
+	}
+
+	esas2r_rq_destroy_request(rq, a);
+	return success;
+}
+
+/* Initialize the adapter chip */
+bool esas2r_init_adapter_hw(struct esas2r_adapter *a, bool init_poll)
+{
+	bool rslt = false;
+	struct esas2r_request *rq;
+	u32 i;
+
+	if (a->flags & AF_DEGRADED_MODE)
+		goto exit;
+
+	if (!(a->flags & AF_NVR_VALID)) {
+		if (!esas2r_nvram_read_direct(a))
+			esas2r_log(ESAS2R_LOG_WARN,
+				   "invalid/missing NVRAM parameters");
+	}
+
+	if (!esas2r_init_msgs(a)) {
+		esas2r_set_degraded_mode(a, "init messages failed");
+		goto exit;
+	}
+
+	/* The firmware is ready. */
+	esas2r_lock_clear_flags(&a->flags, AF_DEGRADED_MODE);
+	esas2r_lock_clear_flags(&a->flags, AF_CHPRST_PENDING);
+
+	/* Post all the async event requests */
+	for (i = 0, rq = a->first_ae_req; i < num_ae_requests; i++, rq++)
+		esas2r_start_ae_request(a, rq);
+
+	if (!a->flash_rev[0])
+		esas2r_read_flash_rev(a);
+
+	if (!a->image_type[0])
+		esas2r_read_image_type(a);
+
+	if (a->fw_version == 0)
+		a->fw_rev[0] = 0;
+	else
+		sprintf(a->fw_rev, "%1d.%02d",
+			(int)LOBYTE(HIWORD(a->fw_version)),
+			(int)HIBYTE(HIWORD(a->fw_version)));
+
+	esas2r_hdebug("firmware revision: %s", a->fw_rev);
+
+	if ((a->flags & AF_CHPRST_DETECTED)
+	    && (a->flags & AF_FIRST_INIT)) {
+		esas2r_enable_chip_interrupts(a);
+		return true;
+	}
+
+	/* initialize discovery */
+	esas2r_disc_initialize(a);
+
+	/*
+	 * wait for the device wait time to expire here if requested.  this is
+	 * usually requested during initial driver load and possibly when
+	 * resuming from a low power state.  deferred device waiting will use
+	 * interrupts.  chip reset recovery always defers device waiting to
+	 * avoid being in a TASKLET too long.
+	 */
+	if (init_poll) {
+		u32 currtime = a->disc_start_time;
+		u32 nexttick = 100;
+		u32 deltatime;
+
+		/*
+		 * Block Tasklets from getting scheduled and indicate this is
+		 * polled discovery.
+		 */
+		esas2r_lock_set_flags(&a->flags, AF_TASKLET_SCHEDULED);
+		esas2r_lock_set_flags(&a->flags, AF_DISC_POLLED);
+
+		/*
+		 * Temporarily bring the disable count to zero to enable
+		 * deferred processing.  Note that the count is already zero
+		 * after the first initialization.
+		 */
+		if (a->flags & AF_FIRST_INIT)
+			atomic_dec(&a->disable_cnt);
+
+		while (a->flags & AF_DISC_PENDING) {
+			schedule_timeout_interruptible(msecs_to_jiffies(100));
+
+			/*
+			 * Determine the need for a timer tick based on the
+			 * delta time between this and the last iteration of
+			 * this loop.  We don't use the absolute time because
+			 * then we would have to worry about when nexttick
+			 * wraps and currtime hasn't yet.
+			 */
+			deltatime = jiffies_to_msecs(jiffies) - currtime;
+			currtime += deltatime;
+
+			/*
+			 * Process any waiting discovery as long as the chip is
+			 * up.  If a chip reset happens during initial polling,
+			 * we have to make sure the timer tick processes the
+			 * doorbell indicating the firmware is ready.
+			 */
+			if (!(a->flags & AF_CHPRST_PENDING))
+				esas2r_disc_check_for_work(a);
+
+			/* Simulate a timer tick. */
+			if (nexttick <= deltatime) {
+
+				/* Time for a timer tick */
+				nexttick += 100;
+				esas2r_timer_tick(a);
+			}
+
+			if (nexttick > deltatime)
+				nexttick -= deltatime;
+
+			/* Do any deferred processing */
+			if (esas2r_is_tasklet_pending(a))
+				esas2r_do_tasklet_tasks(a);
+
+		}
+
+		if (a->flags & AF_FIRST_INIT)
+			atomic_inc(&a->disable_cnt);
+
+		esas2r_lock_clear_flags(&a->flags, AF_DISC_POLLED);
+		esas2r_lock_clear_flags(&a->flags, AF_TASKLET_SCHEDULED);
+	}
+
+
+	esas2r_targ_db_report_changes(a);
+
+	/*
+	 * For cases where (a) the initialization messages processing may
+	 * handle an interrupt for a port event and a discovery is waiting, but
+	 * we are not waiting for devices, or (b) the device wait time has been
+	 * exhausted but there is still discovery pending, start any leftover
+	 * discovery in interrupt driven mode.
+	 */
+	esas2r_disc_start_waiting(a);
+
+	/* Enable chip interrupts */
+	a->int_mask = ESAS2R_INT_STS_MASK;
+	esas2r_enable_chip_interrupts(a);
+	esas2r_enable_heartbeat(a);
+	rslt = true;
+
+exit:
+	/*
+	 * Regardless of whether initialization was successful, certain things
+	 * need to get done before we exit.
+	 */
+
+	if ((a->flags & AF_CHPRST_DETECTED)
+	    && (a->flags & AF_FIRST_INIT)) {
+		/*
+		 * Reinitialization was performed during the first
+		 * initialization.  Only clear the chip reset flag so the
+		 * original device polling is not cancelled.
+		 */
+		if (!rslt)
+			esas2r_lock_clear_flags(&a->flags, AF_CHPRST_PENDING);
+	} else {
+		/* First initialization or a subsequent re-init is complete. */
+		if (!rslt) {
+			esas2r_lock_clear_flags(&a->flags, AF_CHPRST_PENDING);
+			esas2r_lock_clear_flags(&a->flags, AF_DISC_PENDING);
+		}
+
+
+		/* Enable deferred processing after the first initialization. */
+		if (a->flags & AF_FIRST_INIT) {
+			esas2r_lock_clear_flags(&a->flags, AF_FIRST_INIT);
+
+			if (atomic_dec_return(&a->disable_cnt) == 0)
+				esas2r_do_deferred_processes(a);
+		}
+	}
+
+	return rslt;
+}
+
+void esas2r_reset_adapter(struct esas2r_adapter *a)
+{
+	esas2r_lock_set_flags(&a->flags, AF_OS_RESET);
+	esas2r_local_reset_adapter(a);
+	esas2r_schedule_tasklet(a);
+}
+
+void esas2r_reset_chip(struct esas2r_adapter *a)
+{
+	if (!esas2r_is_adapter_present(a))
+		return;
+
+	/*
+	 * Before we reset the chip, save off the VDA core dump.  The VDA core
+	 * dump is located in the upper 512KB of the onchip SRAM.  Make sure
+	 * to not overwrite a previous crash that was saved.
+	 */
+	if ((a->flags2 & AF2_COREDUMP_AVAIL)
+	    && !(a->flags2 & AF2_COREDUMP_SAVED)
+	    && a->fw_coredump_buff) {
+		esas2r_read_mem_block(a,
+				      a->fw_coredump_buff,
+				      MW_DATA_ADDR_SRAM + 0x80000,
+				      ESAS2R_FWCOREDUMP_SZ);
+
+		esas2r_lock_set_flags(&a->flags2, AF2_COREDUMP_SAVED);
+	}
+
+	esas2r_lock_clear_flags(&a->flags2, AF2_COREDUMP_AVAIL);
+
+	/* Reset the chip */
+	if (a->pcid->revision == MVR_FREY_B2)
+		esas2r_write_register_dword(a, MU_CTL_STATUS_IN_B2,
+					    MU_CTL_IN_FULL_RST2);
+	else
+		esas2r_write_register_dword(a, MU_CTL_STATUS_IN,
+					    MU_CTL_IN_FULL_RST);
+
+
+	/* Stall a little while to let the reset condition clear */
+	mdelay(10);
+}
+
+static void esas2r_power_down_notify_firmware(struct esas2r_adapter *a)
+{
+	u32 starttime;
+	u32 doorbell;
+
+	esas2r_write_register_dword(a, MU_DOORBELL_IN, DRBL_POWER_DOWN);
+	starttime = jiffies_to_msecs(jiffies);
+
+	while (true) {
+		doorbell = esas2r_read_register_dword(a, MU_DOORBELL_OUT);
+		if (doorbell & DRBL_POWER_DOWN) {
+			esas2r_write_register_dword(a, MU_DOORBELL_OUT,
+						    doorbell);
+			break;
+		}
+
+		schedule_timeout_interruptible(msecs_to_jiffies(100));
+
+		if ((jiffies_to_msecs(jiffies) - starttime) > 30000) {
+			esas2r_hdebug("Timeout waiting for power down");
+			break;
+		}
+	}
+}
+
+/*
+ * Perform power management processing including managing device states, adapter
+ * states, interrupts, and I/O.
+ */
+void esas2r_power_down(struct esas2r_adapter *a)
+{
+	esas2r_lock_set_flags(&a->flags, AF_POWER_MGT);
+	esas2r_lock_set_flags(&a->flags, AF_POWER_DOWN);
+
+	if (!(a->flags & AF_DEGRADED_MODE)) {
+		u32 starttime;
+		u32 doorbell;
+
+		/*
+		 * We are currently running OK and will be reinitializing later.
+		 * increment the disable count to coordinate with
+		 * esas2r_init_adapter.  We don't have to do this in degraded
+		 * mode since we never enabled interrupts in the first place.
+		 */
+		esas2r_disable_chip_interrupts(a);
+		esas2r_disable_heartbeat(a);
+
+		/* wait for any VDA activity to clear before continuing */
+		esas2r_write_register_dword(a, MU_DOORBELL_IN,
+					    DRBL_MSG_IFC_DOWN);
+		starttime = jiffies_to_msecs(jiffies);
+
+		while (true) {
+			doorbell =
+				esas2r_read_register_dword(a, MU_DOORBELL_OUT);
+			if (doorbell & DRBL_MSG_IFC_DOWN) {
+				esas2r_write_register_dword(a, MU_DOORBELL_OUT,
+							    doorbell);
+				break;
+			}
+
+			schedule_timeout_interruptible(msecs_to_jiffies(100));
+
+			if ((jiffies_to_msecs(jiffies) - starttime) > 3000) {
+				esas2r_hdebug(
+					"timeout waiting for interface down");
+				break;
+			}
+		}
+
+		/*
+		 * For versions of firmware that support it tell them the driver
+		 * is powering down.
+		 */
+		if (a->flags2 & AF2_VDA_POWER_DOWN)
+			esas2r_power_down_notify_firmware(a);
+	}
+
+	/* Suspend I/O processing. */
+	esas2r_lock_set_flags(&a->flags, AF_OS_RESET);
+	esas2r_lock_set_flags(&a->flags, AF_DISC_PENDING);
+	esas2r_lock_set_flags(&a->flags, AF_CHPRST_PENDING);
+
+	esas2r_process_adapter_reset(a);
+
+	/* Remove devices now that I/O is cleaned up. */
+	a->prev_dev_cnt = esas2r_targ_db_get_tgt_cnt(a);
+	esas2r_targ_db_remove_all(a, false);
+}
+
+/*
+ * Perform power management processing including managing device states, adapter
+ * states, interrupts, and I/O.
+ */
+bool esas2r_power_up(struct esas2r_adapter *a, bool init_poll)
+{
+	bool ret;
+
+	esas2r_lock_clear_flags(&a->flags, AF_POWER_DOWN);
+	esas2r_init_pci_cfg_space(a);
+	esas2r_lock_set_flags(&a->flags, AF_FIRST_INIT);
+	atomic_inc(&a->disable_cnt);
+
+	/* reinitialize the adapter */
+	ret = esas2r_check_adapter(a);
+	if (!esas2r_init_adapter_hw(a, init_poll))
+		ret = false;
+
+	/* send the reset asynchronous event */
+	esas2r_send_reset_ae(a, true);
+
+	/* clear this flag after initialization. */
+	esas2r_lock_clear_flags(&a->flags, AF_POWER_MGT);
+	return ret;
+}
+
+bool esas2r_is_adapter_present(struct esas2r_adapter *a)
+{
+	if (a->flags & AF_NOT_PRESENT)
+		return false;
+
+	if (esas2r_read_register_dword(a, MU_DOORBELL_OUT) == 0xFFFFFFFF) {
+		esas2r_lock_set_flags(&a->flags, AF_NOT_PRESENT);
+
+		return false;
+	}
+	return true;
+}
+
+const char *esas2r_get_model_name(struct esas2r_adapter *a)
+{
+	switch (a->pcid->subsystem_device) {
+	case ATTO_ESAS_R680:
+		return "ATTO ExpressSAS R680";
+
+	case ATTO_ESAS_R608:
+		return "ATTO ExpressSAS R608";
+
+	case ATTO_ESAS_R60F:
+		return "ATTO ExpressSAS R60F";
+
+	case ATTO_ESAS_R6F0:
+		return "ATTO ExpressSAS R6F0";
+
+	case ATTO_ESAS_R644:
+		return "ATTO ExpressSAS R644";
+
+	case ATTO_ESAS_R648:
+		return "ATTO ExpressSAS R648";
+
+	case ATTO_TSSC_3808:
+		return "ATTO ThunderStream SC 3808D";
+
+	case ATTO_TSSC_3808E:
+		return "ATTO ThunderStream SC 3808E";
+
+	case ATTO_TLSH_1068:
+		return "ATTO ThunderLink SH 1068";
+	}
+
+	return "ATTO SAS Controller";
+}
+
+const char *esas2r_get_model_name_short(struct esas2r_adapter *a)
+{
+	switch (a->pcid->subsystem_device) {
+	case ATTO_ESAS_R680:
+		return "R680";
+
+	case ATTO_ESAS_R608:
+		return "R608";
+
+	case ATTO_ESAS_R60F:
+		return "R60F";
+
+	case ATTO_ESAS_R6F0:
+		return "R6F0";
+
+	case ATTO_ESAS_R644:
+		return "R644";
+
+	case ATTO_ESAS_R648:
+		return "R648";
+
+	case ATTO_TSSC_3808:
+		return "SC 3808D";
+
+	case ATTO_TSSC_3808E:
+		return "SC 3808E";
+
+	case ATTO_TLSH_1068:
+		return "SH 1068";
+	}
+
+	return "unknown";
+}
-- 
1.8.1.4

--
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html




[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [SCSI Target Devel]     [Linux SCSI Target Infrastructure]     [Kernel Newbies]     [IDE]     [Security]     [Git]     [Netfilter]     [Bugtraq]     [Yosemite News]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux ATA RAID]     [Linux IIO]     [Samba]     [Device Mapper]
  Powered by Linux