hpsa driver updates
in [Kernel]
|
Prev: [PATCH 06/21] sched: drop preempt notifiers
Next: [Bug #14354] Bad corruption with 2.6.32-rc1 and upwards
From: James Bottomley on 16 Nov 2009 16:10 On Wed, 2009-11-11 at 10:50 -0600, Stephen M. Cameron wrote: > The following series implements hpsa scsi driver for HP Smart Arrays, > and some updates since the last time. > The first 5 patches in the series are already in Andrew Morton's tree. > > --- > > Andrew Morton (1): > avoid helpful cleanup patches. > > Stephen M. Cameron (16): > hpsa: fix typo that causes scsi status to be lost > hpsa: Make fill_cmd() return void > hpsa: Remove sendcmd, in no case are we required to poll for completions. > hpsa: Flush cache with interrupts still enabled. > hpsa: Retry driver initiated commands on unit attention > hpsa: decode unit attention condition and retry commands. > hpsa: Make hpsa_sdev_attrs static > hpsa: Allow device rescan to be triggered via sysfs. > Add thread to allow controllers to register for rescan for new devices > hpsa: Factor out some pci_unmap code > hpsa: Factor out command submission sequence > hpsa: Use shost_priv instead of accessing host->hostdata[0] directly. > hpsa: Allocate the correct amount of extra space for the scsi host > Fix use of unallocated memory for MSA2xxx enclosure device data. > hpsa: Fix vendor id check > Add hpsa driver for HP Smart Array controllers. Actually, it's pretty difficult to review a 17 patch series like this because the human mind (or at least mine) doesn't retain sufficient context from patch to patch. I ended up just pulling all 17 into a tree and reviewing the finished driver. That said: in hpsa.c: > static struct device_attribute *hpsa_shost_attrs[] = { > &dev_attr_rescan, > NULL, > }; We already have a host scan attribute which (admittedly using the transport class logic) you can plug into ... can't you just use it? It supplies user context, so you could dispense with all that scan thread stuff as well, I think. > static DEFINE_MUTEX(scan_mutex); > static LIST_HEAD(scan_q); > static int scan_thread(void *data); These names are too generic. We already have a scan_mutex at least defined at the top level. I know they're protected by static, but that doesn't necessarily help if they show up in a debug stack trace. All of this report luns stuff looks fairly identical to the report luns we do in scsi_scan.c ... barring the initial command, which could be translated. Wouldn't it be easier to have the generic code parse and do all of this? > static int hpsa_scsi_queue_command(struct scsi_cmnd *cmd, > void (*done)(struct scsi_cmnd *)) > { [...] > c = cmd_alloc(h); > spin_unlock_irqrestore(&h->lock, flags); > if (c == NULL) { /* trouble... */ > dev_err(&h->pdev->dev, "cmd_alloc returned NULL!\n"); > cmd->result = DID_NO_CONNECT << 16; > done(cmd); > return 0; > } I think you want to return SCSI_MLQUEUE_HOST_BUSY here, which will trigger a throttle and retry after either something frees or I/O pressure builds more. > static int hpsa_eh_device_reset_handler(struct scsi_cmnd *scsicmd) > { [...] > rc = hpsa_send_reset(h, dev->scsi3addr); > if (rc == 0 && wait_for_device_to_become_ready(h, dev->scsi3addr) == 0) > return SUCCESS; So the first thing we do after a device reset successful return is send a test unit ready to the failing device ... there's no real need for you to duplicate that, is there? The ioctl stuff looks like you could do it all with SG_IO now rather than rolling your own versions ... or is there some backward compatibility problem here? > static __devinit int hpsa_hard_reset_controller(struct pci_dev *pdev) > { [...] > set_current_state(TASK_UNINTERRUPTIBLE); > schedule_timeout(HZ >> 1); msleep(500) please .. This isn't the only place this occurs, could you replace all of them? in hpsa_cmd.h: > /* Unit Attentions ASC's as defined for the MSA2012sa */ > #define POWER_OR_RESET 0x29 > #define STATE_CHANGED 0x2a > #define UNIT_ATTENTION_CLEARED 0x2f > #define LUN_FAILED 0x3e > #define REPORT_LUNS_CHANGED 0x3f > > /* Unit Attentions ASCQ's as defined for the MSA2012sa */ > > /* These ASCQ's defined for ASC = POWER_OR_RESET */ > #define POWER_ON_RESET 0x00 > #define POWER_ON_REBOOT 0x01 > #define SCSI_BUS_RESET 0x02 > #define MSA_TARGET_RESET 0x03 > #define CONTROLLER_FAILOVER 0x04 > #define TRANSCEIVER_SE 0x05 > #define TRANSCEIVER_LVD 0x06 > > /* These ASCQ's defined for ASC = STATE_CHANGED */ > #define RESERVATION_PREEMPTED 0x03 > #define ASYM_ACCESS_CHANGED 0x06 > #define LUN_CAPACITY_CHANGED 0x09 Traditionally we've shied away from putting ASC/ASCQ values into defines ... but these all look to be global not hpsa local, so they should be in a common central file. Otherwise looks OK to a cursory glance. James -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Andrew Morton on 16 Nov 2009 16:40 On Mon, 16 Nov 2009 15:06:49 -0600 James Bottomley <James.Bottomley(a)suse.de> wrote: > Actually, it's pretty difficult to review a 17 patch series like this > because the human mind (or at least mine) doesn't retain sufficient > context from patch to patch. I ended up just pulling all 17 into a tree > and reviewing the finished driver. I rolled all the patches together: From: "Stephen M. Cameron" <scameron(a)beardog.cce.hp.com> Add hpsa driver for HP Smart Array controllers. This driver supports a subset of HP Smart Array Controllers. It is a SCSI alternative to the cciss driver. [akpm(a)linux-foundation.org: avoid helpful cleanup patches] Signed-off-by: Stephen M. Cameron <scameron(a)beardog.cce.hp.com> Signed-off-by: Mike Miller <mikem(a)beardog.cce.hp.com> Cc: Jens Axboe <jens.axboe(a)oracle.com> Cc: James Bottomley <James.Bottomley(a)HansenPartnership.com> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- drivers/scsi/Kconfig | 10 drivers/scsi/Makefile | 1 drivers/scsi/hpsa.c | 3531 ++++++++++++++++++++++++++++++++++++++ drivers/scsi/hpsa.h | 273 ++ drivers/scsi/hpsa_cmd.h | 326 +++ 5 files changed, 4141 insertions(+) diff -puN drivers/scsi/Kconfig~scsi-add-hpsa-driver-for-hp-smart-array-controllers drivers/scsi/Kconfig --- a/drivers/scsi/Kconfig~scsi-add-hpsa-driver-for-hp-smart-array-controllers +++ a/drivers/scsi/Kconfig @@ -388,6 +388,16 @@ config BLK_DEV_3W_XXXX_RAID Please read the comments at the top of <file:drivers/scsi/3w-xxxx.c>. +config SCSI_HPSA + tristate "HP Smart Array SCSI driver" + depends on PCI && SCSI + help + This driver supports HP Smart Array Controllers (circa 2009). + It is a SCSI alternative to the cciss driver, which is a block + driver. Anyone wishing to use HP Smart Array controllers who + would prefer the devices be presented to linux as SCSI devices, + rather than as generic block devices should say Y here. + config SCSI_3W_9XXX tristate "3ware 9xxx SATA-RAID support" depends on PCI && SCSI diff -puN drivers/scsi/Makefile~scsi-add-hpsa-driver-for-hp-smart-array-controllers drivers/scsi/Makefile --- a/drivers/scsi/Makefile~scsi-add-hpsa-driver-for-hp-smart-array-controllers +++ a/drivers/scsi/Makefile @@ -91,6 +91,7 @@ obj-$(CONFIG_SCSI_BFA_FC) += bfa/ obj-$(CONFIG_SCSI_PAS16) += pas16.o obj-$(CONFIG_SCSI_T128) += t128.o obj-$(CONFIG_SCSI_DMX3191D) += dmx3191d.o +obj-$(CONFIG_SCSI_HPSA) += hpsa.o obj-$(CONFIG_SCSI_DTC3280) += dtc.o obj-$(CONFIG_SCSI_SYM53C8XX_2) += sym53c8xx_2/ obj-$(CONFIG_SCSI_ZALON) += zalon7xx.o diff -puN /dev/null drivers/scsi/hpsa.c --- /dev/null +++ a/drivers/scsi/hpsa.c @@ -0,0 +1,3531 @@ +/* + * Disk Array driver for HP Smart Array SAS controllers + * Copyright 2000, 2009 Hewlett-Packard Development Company, L.P. + * + * 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; version 2 of the License. + * + * 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, GOOD TITLE or + * NON INFRINGEMENT. See the GNU General Public License for more details. + * + * 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., 675 Mass Ave, Cambridge, MA 02139, USA. + * + * Questions/Comments/Bugfixes to iss_storagedev(a)hp.com + * + */ + +#include <linux/module.h> +#include <linux/interrupt.h> +#include <linux/types.h> +#include <linux/pci.h> +#include <linux/kernel.h> +#include <linux/slab.h> +#include <linux/delay.h> +#include <linux/fs.h> +#include <linux/timer.h> +#include <linux/seq_file.h> +#include <linux/init.h> +#include <linux/spinlock.h> +#include <linux/smp_lock.h> +#include <linux/compat.h> +#include <linux/blktrace_api.h> +#include <linux/uaccess.h> +#include <linux/io.h> +#include <linux/dma-mapping.h> +#include <linux/completion.h> +#include <linux/moduleparam.h> +#include <scsi/scsi.h> +#include <scsi/scsi_cmnd.h> +#include <scsi/scsi_device.h> +#include <scsi/scsi_host.h> +#include <linux/cciss_ioctl.h> +#include <linux/string.h> +#include <linux/bitmap.h> +#include <asm/atomic.h> +#include <linux/kthread.h> +#include "hpsa_cmd.h" +#include "hpsa.h" + +/* HPSA_DRIVER_VERSION must be 3 byte values (0-255) separated by '.' */ +#define HPSA_DRIVER_VERSION "1.0.0" +#define DRIVER_NAME "HP HPSA Driver (v " HPSA_DRIVER_VERSION ")" + +/* How long to wait (in milliseconds) for board to go into simple mode */ +#define MAX_CONFIG_WAIT 30000 +#define MAX_IOCTL_CONFIG_WAIT 1000 + +/*define how many times we will try a command because of bus resets */ +#define MAX_CMD_RETRIES 3 + +/* Embedded module documentation macros - see modules.h */ +MODULE_AUTHOR("Hewlett-Packard Company"); +MODULE_DESCRIPTION("Driver for HP Smart Array Controller version " \ + HPSA_DRIVER_VERSION); +MODULE_SUPPORTED_DEVICE("HP Smart Array Controllers"); +MODULE_VERSION(HPSA_DRIVER_VERSION); +MODULE_LICENSE("GPL"); + +static int hpsa_allow_any; +module_param(hpsa_allow_any, int, S_IRUGO|S_IWUSR); +MODULE_PARM_DESC(hpsa_allow_any, + "Allow hpsa driver to access unknown HP Smart Array hardware"); + +/* define the PCI info for the cards we can control */ +static const struct pci_device_id hpsa_pci_device_id[] = { + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSC, 0x103C, 0x3223}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSC, 0x103C, 0x3234}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSC, 0x103C, 0x323D}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSE, 0x103C, 0x3241}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSE, 0x103C, 0x3243}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSE, 0x103C, 0x3245}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSE, 0x103C, 0x3247}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSE, 0x103C, 0x3249}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSE, 0x103C, 0x324a}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSE, 0x103C, 0x324b}, + {PCI_VENDOR_ID_HP, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, + PCI_CLASS_STORAGE_RAID << 8, 0xffff << 8, 0}, + {0,} +}; + +MODULE_DEVICE_TABLE(pci, hpsa_pci_device_id); + +/* board_id = Subsystem Device ID & Vendor ID + * product = Marketing Name for the board + * access = Address of the struct of function pointers + */ +static struct board_type products[] = { + {0x3223103C, "Smart Array P800", &SA5_access}, + {0x3234103C, "Smart Array P400", &SA5_access}, + {0x323d103c, "Smart Array P700M", &SA5_access}, + {0x3241103C, "Smart Array P212", &SA5_access}, + {0x3243103C, "Smart Array P410", &SA5_access}, + {0x3245103C, "Smart Array P410i", &SA5_access}, + {0x3247103C, "Smart Array P411", &SA5_access}, + {0x3249103C, "Smart Array P812", &SA5_access}, + {0x324a103C, "Smart Array P712m", &SA5_access}, + {0x324b103C, "Smart Array P711m", &SA5_access}, + {0xFFFF103C, "Unknown Smart Array", &SA5_access}, +}; + +static int number_of_controllers; + +static irqreturn_t do_hpsa_intr(int irq, void *dev_id); +static int hpsa_ioctl(struct scsi_device *dev, int cmd, void *arg); +static void start_io(struct ctlr_info *h); + +#ifdef CONFIG_COMPAT +static int hpsa_compat_ioctl(struct scsi_device *dev, int cmd, void *arg); +#endif + +static void cmd_free(struct ctlr_info *h, struct CommandList *c); +static void cmd_special_free(struct ctlr_info *h, struct CommandList *c); +static struct CommandList *cmd_alloc(struct ctlr_info *h); +static struct CommandList *cmd_special_alloc(struct ctlr_info *h); +static void fill_cmd(struct CommandList *c, __u8 cmd, struct ctlr_info *h, + void *buff, size_t size, __u8 page_code, unsigned char *scsi3addr, + int cmd_type); + +static int hpsa_scsi_queue_command(struct scsi_cmnd *cmd, + void (*done)(struct scsi_cmnd *)); + +static int hpsa_eh_device_reset_handler(struct scsi_cmnd *scsicmd); +static int hpsa_slave_alloc(struct scsi_device *sdev); +static void hpsa_slave_destroy(struct scsi_device *sdev); + +static ssize_t raid_level_show(struct device *dev, + struct device_attribute *attr, char *buf); +static ssize_t lunid_show(struct device *dev, + struct device_attribute *attr, char *buf); +static ssize_t unique_id_show(struct device *dev, + struct device_attribute *attr, char *buf); +static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno); +static ssize_t host_store_rescan(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count); +static int check_for_unit_attention(struct ctlr_info *h, + struct CommandList *c); +static void check_ioctl_unit_attention(struct ctlr_info *h, + struct CommandList *c); + +DEVICE_ATTR(raid_level, S_IRUGO, raid_level_show, NULL); +DEVICE_ATTR(lunid, S_IRUGO, lunid_show, NULL); +DEVICE_ATTR(unique_id, S_IRUGO, unique_id_show, NULL); +DEVICE_ATTR(rescan, S_IWUSR, NULL, host_store_rescan); + +static struct device_attribute *hpsa_sdev_attrs[] = { + &dev_attr_raid_level, + &dev_attr_lunid, + &dev_attr_unique_id, + NULL, +}; + +static struct device_attribute *hpsa_shost_attrs[] = { + &dev_attr_rescan, + NULL, +}; + +static struct scsi_host_template hpsa_driver_template = { + .module = THIS_MODULE, + .name = "hpsa", + .proc_name = "hpsa", + .queuecommand = hpsa_scsi_queue_command, + .can_queue = 512, + .this_id = -1, + .sg_tablesize = MAXSGENTRIES, + .cmd_per_lun = 512, + .use_clustering = ENABLE_CLUSTERING, + .eh_device_reset_handler = hpsa_eh_device_reset_handler, + .ioctl = hpsa_ioctl, + .slave_alloc = hpsa_slave_alloc, + .slave_destroy = hpsa_slave_destroy, +#ifdef CONFIG_COMPAT + .compat_ioctl = hpsa_compat_ioctl, +#endif + .sdev_attrs = hpsa_sdev_attrs, + .shost_attrs = hpsa_shost_attrs, +}; + +static inline struct ctlr_info *sdev_to_hba(struct scsi_device *sdev) +{ + unsigned long *priv = shost_priv(sdev->host); + return (struct ctlr_info *) *priv; +} + +static struct task_struct *hpsa_scan_thread; +static DEFINE_MUTEX(scan_mutex); +static LIST_HEAD(scan_q); +static int scan_thread(void *data); + +/** + * add_to_scan_list() - add controller to rescan queue + * @h: Pointer to the controller. + * + * Adds the controller to the rescan queue if not already on the queue. + * + * returns 1 if added to the queue, 0 if skipped (could be on the + * queue already, or the controller could be initializing or shutting + * down). + **/ +static int add_to_scan_list(struct ctlr_info *h) +{ + struct ctlr_info *test_h; + int found = 0; + int ret = 0; + + if (h->busy_initializing) + return 0; + + if (!mutex_trylock(&h->busy_shutting_down)) + return 0; + + mutex_lock(&scan_mutex); + list_for_each_entry(test_h, &scan_q, scan_list) { + if (test_h == h) { + found = 1; + break; + } + } + if (!found && !h->busy_scanning) { + INIT_COMPLETION(h->scan_wait); + list_add_tail(&h->scan_list, &scan_q); + ret = 1; + } + mutex_unlock(&scan_mutex); + mutex_unlock(&h->busy_shutting_down); + + return ret; +} + +/** + * remove_from_scan_list() - remove controller from rescan queue + * @h: Pointer to the controller. + * + * Removes the controller from the rescan queue if present. Blocks if + * the controller is currently conducting a rescan. The controller + * can be in one of three states: + * 1. Doesn't need a scan + * 2. On the scan list, but not scanning yet (we remove it) + * 3. Busy scanning (and not on the list). In this case we want to wait for + * the scan to complete to make sure the scanning thread for this + * controller is completely idle. + **/ +static void remove_from_scan_list(struct ctlr_info *h) +{ + struct ctlr_info *test_h, *tmp_h; + + mutex_lock(&scan_mutex); + list_for_each_entry_safe(test_h, tmp_h, &scan_q, scan_list) { + if (test_h == h) { /* state 2. */ + list_del(&h->scan_list); + complete_all(&h->scan_wait); + mutex_unlock(&scan_mutex); + return; + } + } + if (h->busy_scanning) { /* state 3. */ + mutex_unlock(&scan_mutex); + wait_for_completion(&h->scan_wait); + } else { /* state 1, nothing to do. */ + mutex_unlock(&scan_mutex); + } +} + +/* scan_thread() - kernel thread used to rescan controllers + * @data: Ignored. + * + * A kernel thread used scan for drive topology changes on + * controllers. The thread processes only one controller at a time + * using a queue. Controllers are added to the queue using + * add_to_scan_list() and removed from the queue either after done + * processing or using remove_from_scan_list(). + * + * returns 0. + **/ +static int scan_thread(__attribute__((unused)) void *data) +{ + struct ctlr_info *h; + int host_no; + + while (1) { + set_current_state(TASK_INTERRUPTIBLE); + schedule(); + if (kthread_should_stop()) + break; + + while (1) { + mutex_lock(&scan_mutex); + if (list_empty(&scan_q)) { + mutex_unlock(&scan_mutex); + break; + } + h = list_entry(scan_q.next, struct ctlr_info, + scan_list); + list_del(&h->scan_list); + h->busy_scanning = 1; + mutex_unlock(&scan_mutex); + host_no = h->scsi_host ? h->scsi_host->host_no : -1; + hpsa_update_scsi_devices(h, host_no); + complete_all(&h->scan_wait); + mutex_lock(&scan_mutex); + h->busy_scanning = 0; + mutex_unlock(&scan_mutex); + } + } + return 0; +} + +static int check_for_unit_attention(struct ctlr_info *h, + struct CommandList *c) +{ + if (c->err_info->SenseInfo[2] != UNIT_ATTENTION) + return 0; + + switch (c->err_info->SenseInfo[12]) { + case STATE_CHANGED: + dev_warn(&h->pdev->dev, "hpsa%d: a state change " + "detected, command retried\n", h->ctlr); + break; + case LUN_FAILED: + dev_warn(&h->pdev->dev, "hpsa%d: LUN failure " + "detected, action required\n", h->ctlr); + break; + case REPORT_LUNS_CHANGED: + dev_warn(&h->pdev->dev, "hpsa%d: report LUN data " + "changed\n", h->ctlr); + /* + * Here, we could call add_to_scan_list and wake up the scan thread, + * except that it's quite likely that we will get more than one + * REPORT_LUNS_CHANGED condition in quick succession, which means + * that those which occur after the first one will likely happen + * *during* the scan_thread's rescan. And the rescan code is not + * robust enough to restart in the middle, undoing what it has already + * done, and it's not clear that it's even possible to do this, since + * part of what it does is notify the SCSI mid layer, which starts + * doing it's own i/o to read partition tables and so on, and the + * driver doesn't have visibility to know what might need undoing. + * In any event, if possible, it is horribly complicated to get right + * so we just don't do it for now. + * + * Note: this REPORT_LUNS_CHANGED condition only occurs on the MSA2012. + */ + break; + case POWER_OR_RESET: + dev_warn(&h->pdev->dev, "hpsa%d: a power on " + "or device reset detected\n", h->ctlr); + break; + case UNIT_ATTENTION_CLEARED: + dev_warn(&h->pdev->dev, "hpsa%d: unit attention " + "cleared by another initiator\n", h->ctlr); + break; + default: + dev_warn(&h->pdev->dev, "hpsa%d: unknown " + "unit attention detected\n", h->ctlr); + break; + } + return 1; +} + +static ssize_t host_store_rescan(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct ctlr_info *h; + struct Scsi_Host *shost = class_to_shost(dev); + unsigned long *priv = shost_priv(shost); + h = (struct ctlr_info *) *priv; + if (add_to_scan_list(h)) { + wake_up_process(hpsa_scan_thread); + wait_for_completion_interruptible(&h->scan_wait); + } + return count; +} + +/* Enqueuing and dequeuing functions for cmdlists. */ +static inline void addQ(struct hlist_head *list, struct CommandList *c) +{ + hlist_add_head(&c->list, list); +} + +static void enqueue_cmd_and_start_io(struct ctlr_info *h, + struct CommandList *c) +{ + unsigned long flags; + spin_lock_irqsave(&h->lock, flags); + addQ(&h->reqQ, c); + h->Qdepth++; + start_io(h); + spin_unlock_irqrestore(&h->lock, flags); +} + +static inline void removeQ(struct CommandList *c) +{ + if (WARN_ON(hlist_unhashed(&c->list))) + return; + hlist_del_init(&c->list); +} + +static inline int is_hba_lunid(unsigned char scsi3addr[]) +{ + return memcmp(scsi3addr, RAID_CTLR_LUNID, 8) == 0; +} + +static inline int is_logical_dev_addr_mode(unsigned char scsi3addr[]) +{ + return (scsi3addr[3] & 0xC0) == 0x40; +} + +static const char *raid_label[] = { "0", "4", "1(1+0)", "5", "5+1", "ADG", + "UNKNOWN" +}; +#define RAID_UNKNOWN (ARRAY_SIZE(raid_label) - 1) + +static ssize_t raid_level_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + ssize_t l = 0; + int rlevel; + struct ctlr_info *h; + struct scsi_device *sdev; + struct hpsa_scsi_dev_t *hdev; + unsigned long flags; + + sdev = to_scsi_device(dev); + h = sdev_to_hba(sdev); + spin_lock_irqsave(&h->lock, flags); + hdev = sdev->hostdata; + if (!hdev) { + spin_unlock_irqrestore(&h->lock, flags); + return -ENODEV; + } + + /* Is this even a logical drive? */ + if (!is_logical_dev_addr_mode(hdev->scsi3addr)) { + spin_unlock_irqrestore(&h->lock, flags); + l = snprintf(buf, PAGE_SIZE, "N/A\n"); + return l; + } + + rlevel = hdev->raid_level; + spin_unlock_irqrestore(&h->lock, flags); + if (rlevel < 0 || rlevel > RAID_UNKNOWN) + rlevel = RAID_UNKNOWN; + l = snprintf(buf, PAGE_SIZE, "RAID %s\n", raid_label[rlevel]); + return l; +} + +static ssize_t lunid_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct ctlr_info *h; + struct scsi_device *sdev; + struct hpsa_scsi_dev_t *hdev; + unsigned long flags; + unsigned char lunid[8]; + + sdev = to_scsi_device(dev); + h = sdev_to_hba(sdev); + spin_lock_irqsave(&h->lock, flags); + hdev = sdev->hostdata; + if (!hdev) { + spin_unlock_irqrestore(&h->lock, flags); + return -ENODEV; + } + memcpy(lunid, hdev->scsi3addr, sizeof(lunid)); + spin_unlock_irqrestore(&h->lock, flags); + return snprintf(buf, 20, "0x%02x%02x%02x%02x%02x%02x%02x%02x\n", + lunid[0], lunid[1], lunid[2], lunid[3], + lunid[4], lunid[5], lunid[6], lunid[7]); +} + +static ssize_t unique_id_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct ctlr_info *h; + struct scsi_device *sdev; + struct hpsa_scsi_dev_t *hdev; + unsigned long flags; + unsigned char sn[16]; + + sdev = to_scsi_device(dev); + h = sdev_to_hba(sdev); + spin_lock_irqsave(&h->lock, flags); + hdev = sdev->hostdata; + if (!hdev) { + spin_unlock_irqrestore(&h->lock, flags); + return -ENODEV; + } + memcpy(sn, hdev->device_id, sizeof(sn)); + spin_unlock_irqrestore(&h->lock, flags); + return snprintf(buf, 16 * 2 + 2, + "%02X%02X%02X%02X%02X%02X%02X%02X" + "%02X%02X%02X%02X%02X%02X%02X%02X\n", + sn[0], sn[1], sn[2], sn[3], + sn[4], sn[5], sn[6], sn[7], + sn[8], sn[9], sn[10], sn[11], + sn[12], sn[13], sn[14], sn[15]); +} + +static int hpsa_find_target_lun(struct ctlr_info *h, + unsigned char scsi3addr[], int bus, int *target, int *lun) +{ + /* finds an unused bus, target, lun for a new physical device + * assumes h->devlock is held + */ + int i, found = 0; + DECLARE_BITMAP(lun_taken, HPSA_MAX_SCSI_DEVS_PER_HBA); + + memset(&lun_taken[0], 0, HPSA_MAX_SCSI_DEVS_PER_HBA >> 3); + + for (i = 0; i < h->ndevices; i++) { + if (h->dev[i]->bus == bus && h->dev[i]->target != -1) + set_bit(h->dev[i]->target, lun_taken); + } + + for (i = 0; i < HPSA_MAX_SCSI_DEVS_PER_HBA; i++) { + if (!test_bit(i, lun_taken)) { + /* *bus = 1; */ + *target = i; + *lun = 0; + found = 1; + break; + } + } + return !found; +} + +/* Add an entry into h->dev[] array. */ +static int hpsa_scsi_add_entry(struct ctlr_info *h, int hostno, + struct hpsa_scsi_dev_t *device, + struct hpsa_scsi_dev_t *added[], int *nadded) +{ + /* assumes h->devlock is held */ + int n = h->ndevices; + int i; + unsigned char addr1[8], addr2[8]; + struct hpsa_scsi_dev_t *sd; + + if (n >= HPSA_MAX_SCSI_DEVS_PER_HBA) { + dev_err(&h->pdev->dev, "too many devices, some will be " + "inaccessible.\n"); + return -1; + } + + /* physical devices do not have lun or target assigned until now. */ + if (device->lun != -1) + /* Logical device, lun is already assigned. */ + goto lun_assigned; + + /* If this device a non-zero lun of a multi-lun device + * byte 4 of the 8-byte LUN addr will contain the logical + * unit no, zero otherise. + */ + if (device->scsi3addr[4] == 0) { + /* This is not a non-zero lun of a multi-lun device */ + if (hpsa_find_target_lun(h, device->scsi3addr, + device->bus, &device->target, &device->lun) != 0) + return -1; + goto lun_assigned; + } + + /* This is a non-zero lun of a multi-lun device. + * Search through our list and find the device which + * has the same 8 byte LUN address, excepting byte 4. + * Assign the same bus and target for this new LUN. + * Use the logical unit number from the firmware. + */ + memcpy(addr1, device->scsi3addr, 8); + addr1[4] = 0; + for (i = 0; i < n; i++) { + sd = h->dev[i]; + memcpy(addr2, sd->scsi3addr, 8); + addr2[4] = 0; + /* differ only in byte 4? */ + if (memcmp(addr1, addr2, 8) == 0) { + device->bus = sd->bus; + device->target = sd->target; + device->lun = device->scsi3addr[4]; + break; + } + } + if (device->lun == -1) { + dev_warn(&h->pdev->dev, "physical device with no LUN=0," + " suspect firmware bug or unsupported hardware " + "configuration.\n"); + return -1; + } + +lun_assigned: + + h->dev[n] = device; + h->ndevices++; + added[*nadded] = device; + (*nadded)++; + + /* initially, (before registering with scsi layer) we don't + * know our hostno and we don't want to print anything first + * time anyway (the scsi layer's inquiries will show that info) + */ + /* if (hostno != -1) */ + dev_info(&h->pdev->dev, "%s device c%db%dt%dl%d added.\n", + scsi_device_type(device->devtype), hostno, + device->bus, device->target, device->lun); + return 0; +} + +/* Remove an entry from h->dev[] array. */ +static void hpsa_scsi_remove_entry(struct ctlr_info *h, int hostno, int entry, + struct hpsa_scsi_dev_t *removed[], int *nremoved) +{ + /* assumes h->devlock is held */ + int i; + struct hpsa_scsi_dev_t *sd; + + if (entry < 0 || entry >= HPSA_MAX_SCSI_DEVS_PER_HBA) + BUG(); + + sd = h->dev[entry]; + removed[*nremoved] = h->dev[entry]; + (*nremoved)++; + + for (i = entry; i < h->ndevices-1; i++) + h->dev[i] = h->dev[i+1]; + h->ndevices--; + dev_info(&h->pdev->dev, "%s device c%db%dt%dl%d removed.\n", + scsi_device_type(sd->devtype), hostno, sd->bus, sd->target, + sd->lun); +} + +#define SCSI3ADDR_EQ(a, b) ( \ + (a)[7] == (b)[7] && \ + (a)[6] == (b)[6] && \ + (a)[5] == (b)[5] && \ + (a)[4] == (b)[4] && \ + (a)[3] == (b)[3] && \ + (a)[2] == (b)[2] && \ + (a)[1] == (b)[1] && \ + (a)[0] == (b)[0]) + +static void fixup_botched_add(struct ctlr_info *h, + struct hpsa_scsi_dev_t *added) +{ + /* called when scsi_add_device fails in order to re-adjust + * h->dev[] to match the mid layer's view. + */ + unsigned long flags; + int i, j; + + spin_lock_irqsave(&h->lock, flags); + for (i = 0; i < h->ndevices; i++) { + if (h->dev[i] == added) { + for (j = i; j < h->ndevices-1; j++) + h->dev[j] = h->dev[j+1]; + h->ndevices--; + break; + } + } + spin_unlock_irqrestore(&h->lock, flags); + kfree(added); +} + +static inline int device_is_the_same(struct hpsa_scsi_dev_t *dev1, + struct hpsa_scsi_dev_t *dev2) +{ + if ((is_logical_dev_addr_mode(dev1->scsi3addr) || + (dev1->lun != -1 && dev2->lun != -1)) && + dev1->devtype != 0x0C) + return (memcmp(dev1, dev2, sizeof(*dev1)) == 0); + + /* we compare everything except lun and target as these + * are not yet assigned. Compare parts likely + * to differ first + */ + if (memcmp(dev1->scsi3addr, dev2->scsi3addr, + sizeof(dev1->scsi3addr)) != 0) + return 0; + if (memcmp(dev1->device_id, dev2->device_id, + sizeof(dev1->device_id)) != 0) + return 0; + if (memcmp(dev1->model, dev2->model, sizeof(dev1->model)) != 0) + return 0; + if (memcmp(dev1->vendor, dev2->vendor, sizeof(dev1->vendor)) != 0) + return 0; + if (memcmp(dev1->revision, dev2->revision, sizeof(dev1->revision)) != 0) + return 0; + if (dev1->devtype != dev2->devtype) + return 0; + if (dev1->raid_level != dev2->raid_level) + return 0; + if (dev1->bus != dev2->bus) + return 0; + return 1; +} + +/* Find needle in haystack. If exact match found, return DEVICE_SAME, + * and return needle location in *index. If scsi3addr matches, but not + * vendor, model, serial num, etc. return DEVICE_CHANGED, and return needle + * location in *index. If needle not found, return DEVICE_NOT_FOUND. + */ +static int hpsa_scsi_find_entry(struct hpsa_scsi_dev_t *needle, + struct hpsa_scsi_dev_t *haystack[], int haystack_size, + int *index) +{ + int i; +#define DEVICE_NOT_FOUND 0 +#define DEVICE_CHANGED 1 +#define DEVICE_SAME 2 + for (i = 0; i < haystack_size; i++) { + if (SCSI3ADDR_EQ(needle->scsi3addr, haystack[i]->scsi3addr)) { + *index = i; + if (device_is_the_same(needle, haystack[i])) + return DEVICE_SAME; + else + return DEVICE_CHANGED; + } + } + *index = -1; + return DEVICE_NOT_FOUND; +} + +static int adjust_hpsa_scsi_table(struct ctlr_info *h, int hostno, + struct hpsa_scsi_dev_t *sd[], int nsds) +{ + /* sd contains scsi3 addresses and devtypes, and inquiry + * data. This function takes what's in sd to be the current + * reality and updates h->dev[] to reflect that reality. + */ + int i, entry, device_change, changes = 0; + struct hpsa_scsi_dev_t *csd; + unsigned long flags; + struct hpsa_scsi_dev_t **added, **removed; + int nadded, nremoved; + struct Scsi_Host *sh = NULL; + + added = kzalloc(sizeof(*added) * HPSA_MAX_SCSI_DEVS_PER_HBA, + GFP_KERNEL); + removed = kzalloc(sizeof(*removed) * HPSA_MAX_SCSI_DEVS_PER_HBA, + GFP_KERNEL); + + if (!added || !removed) { + dev_warn(&h->pdev->dev, "out of memory in " + "adjust_hpsa_scsi_table\n"); + goto free_and_out; + } + + spin_lock_irqsave(&h->devlock, flags); + + /* find any devices in h->dev[] that are not in + * sd[] and remove them from h->dev[], and for any + * devices which have changed, remove the old device + * info and add the new device info. + */ + i = 0; + nremoved = 0; + nadded = 0; + while (i < h->ndevices) { + csd = h->dev[i]; + device_change = hpsa_scsi_find_entry(csd, sd, nsds, &entry); + if (device_change == DEVICE_NOT_FOUND) { + changes++; + hpsa_scsi_remove_entry(h, hostno, i, + removed, &nremoved); + continue; /* remove ^^^, hence i not incremented */ + } else if (device_change == DEVICE_CHANGED) { + changes++; + hpsa_scsi_remove_entry(h, hostno, i, + removed, &nremoved); + (void) hpsa_scsi_add_entry(h, hostno, sd[entry], + added, &nadded); + /* add can't fail, we just removed one. */ + sd[entry] = NULL; /* prevent it from being freed */ + } + i++; + } + + /* Now, make sure every device listed in sd[] is also + * listed in h->dev[], adding them if they aren't found + */ + + for (i = 0; i < nsds; i++) { + if (!sd[i]) /* if already added above. */ + continue; + device_change = hpsa_scsi_find_entry(sd[i], h->dev, + h->ndevices, &entry); + if (device_change == DEVICE_NOT_FOUND) { + changes++; + if (hpsa_scsi_add_entry(h, hostno, sd[i], + added, &nadded) != 0) + break; + sd[i] = NULL; /* prevent from being freed later. */ + } else if (device_change == DEVICE_CHANGED) { + /* should never happen... */ + changes++; + dev_warn(&h->pdev->dev, + "device unexpectedly changed.\n"); + /* but if it does happen, we just ignore that device */ + } + } + spin_unlock_irqrestore(&h->devlock, flags); + + /* Don't notify scsi mid layer of any changes the first time through + * (or if there are no changes) scsi_scan_host will do it later the + * first time through. + */ + if (hostno == -1 || !changes) + goto free_and_out; + + sh = h->scsi_host; + /* Notify scsi mid layer of any removed devices */ + for (i = 0; i < nremoved; i++) { + struct scsi_device *sdev = + scsi_device_lookup(sh, removed[i]->bus, + removed[i]->target, removed[i]->lun); + if (sdev != NULL) { + scsi_remove_device(sdev); + scsi_device_put(sdev); + } else { + /* We don't expect to get here. + * future cmds to this device will get selection + * timeout as if the device was gone. + */ + dev_warn(&h->pdev->dev, "didn't find c%db%dt%dl%d " + " for removal.", hostno, removed[i]->bus, + removed[i]->target, removed[i]->lun); + } + kfree(removed[i]); + removed[i] = NULL; + } + + /* Notify scsi mid layer of any added devices */ + for (i = 0; i < nadded; i++) { + if (scsi_add_device(sh, added[i]->bus, + added[i]->target, added[i]->lun) == 0) + continue; + dev_warn(&h->pdev->dev, "scsi_add_device c%db%dt%dl%d failed, " + "device not added.\n", hostno, added[i]->bus, + added[i]->target, added[i]->lun); + /* now we have to remove it from h->dev, + * since it didn't get added to scsi mid layer + */ + fixup_botched_add(h, added[i]); + } + +free_and_out: + kfree(added); + kfree(removed); + return 0; +} + +/* + * Lookup bus/target/lun and retrun corresponding struct hpsa_scsi_dev_t * + * Assume's h->devlock is held. + */ +static struct hpsa_scsi_dev_t *lookup_hpsa_scsi_dev(struct ctlr_info *h, + int bus, int target, int lun) +{ + int i; + struct hpsa_scsi_dev_t *sd; + + for (i = 0; i < h->ndevices; i++) { + sd = h->dev[i]; + if (sd->bus == bus && sd->target == target && sd->lun == lun) + return sd; + } + return NULL; +} + +/* link sdev->hostdata to our per-device structure. */ +static int hpsa_slave_alloc(struct scsi_device *sdev) +{ + struct hpsa_scsi_dev_t *sd; + unsigned long flags; + struct ctlr_info *h; + + h = sdev_to_hba(sdev); + spin_lock_irqsave(&h->devlock, flags); + sd = lookup_hpsa_scsi_dev(h, sdev_channel(sdev), + sdev_id(sdev), sdev->lun); + if (sd != NULL) + sdev->hostdata = sd; + spin_unlock_irqrestore(&h->devlock, flags); + return 0; +} + +static void hpsa_slave_destroy(struct scsi_device *sdev) +{ + return; /* nothing to do. */ +} + +static void hpsa_scsi_setup(struct ctlr_info *h) +{ + h->ndevices = 0; + h->scsi_host = NULL; + spin_lock_init(&h->devlock); + return; +} + +static void complete_scsi_command(struct CommandList *cp, + int timeout, __u32 tag) +{ + struct scsi_cmnd *cmd; + struct ctlr_info *h; + struct ErrorInfo *ei; + + unsigned char sense_key; + unsigned char asc; /* additional sense code */ + unsigned char ascq; /* additional sense code qualifier */ + + ei = cp->err_info; + cmd = (struct scsi_cmnd *) cp->scsi_cmd; + h = cp->h; + + scsi_dma_unmap(cmd); /* undo the DMA mappings */ + + cmd->result = (DID_OK << 16); /* host byte */ + cmd->result |= (COMMAND_COMPLETE << 8); /* msg byte */ + cmd->result |= (ei->ScsiStatus); + + /* copy the sense data whether we need to or not. */ + memcpy(cmd->sense_buffer, ei->SenseInfo, + ei->SenseLen > SCSI_SENSE_BUFFERSIZE ? + SCSI_SENSE_BUFFERSIZE : + ei->SenseLen); + scsi_set_resid(cmd, ei->ResidualCnt); + + if (ei->CommandStatus == 0) { + cmd->scsi_done(cmd); + cmd_free(h, cp); + return; + } + + /* an error has occurred */ + switch (ei->CommandStatus) { + + case CMD_TARGET_STATUS: + if (ei->ScsiStatus) { + /* Get sense key */ + sense_key = 0xf & ei->SenseInfo[2]; + /* Get additional sense code */ + asc = ei->SenseInfo[12]; + /* Get addition sense code qualifier */ + ascq = ei->SenseInfo[13]; + } + + if (ei->ScsiStatus == SAM_STAT_CHECK_CONDITION) { + if (check_for_unit_attention(h, cp)) { + cmd->result = DID_SOFT_ERROR << 16; + break; + } + if (sense_key == ILLEGAL_REQUEST) { + /* If ASC/ASCQ indicate Logical Unit + * Not Supported condition, + */ + if ((asc == 0x25) && (ascq == 0x0)) { + dev_warn(&h->pdev->dev, "cp %p " + "has check condition\n", cp); + break; + } + } + + if (sense_key == NOT_READY) { + /* If Sense is Not Ready, Logical Unit + * Not ready, Manual Intervention + * required + */ + if ((asc == 0x04) && (ascq == 0x03)) { + cmd->result = DID_NO_CONNECT << 16; + dev_warn(&h->pdev->dev, "cp %p " + "has check condition: unit " + "not ready, manual " + "intervention required\n", cp); + break; + } + } + + + /* Must be some other type of check condition */ + cmd->result |= (ei->ScsiStatus << 1); + dev_warn(&h->pdev->dev, "cp %p has check condition: " + "unknown type: " + "Sense: 0x%x, ASC: 0x%x, ASCQ: 0x%x, " + "Returning result: 0x%x, " + "cmd=[%02x %02x %02x %02x %02x " + "%02x %02x %02x %02x %02x]\n", + cp, sense_key, asc, ascq, + cmd->result, + cmd->cmnd[0], cmd->cmnd[1], + cmd->cmnd[2], cmd->cmnd[3], + cmd->cmnd[4], cmd->cmnd[5], + cmd->cmnd[6], cmd->cmnd[7], + cmd->cmnd[8], cmd->cmnd[9]); + break; + } + + + /* Problem was not a check condition + * Pass it up to the upper layers... + */ + if (ei->ScsiStatus) { + + cmd->result |= (ei->ScsiStatus << 1); + dev_warn(&h->pdev->dev, "cp %p has status 0x%x " + "Sense: 0x%x, ASC: 0x%x, ASCQ: 0x%x, " + "Returning result: 0x%x\n", + cp, ei->ScsiStatus, + sense_key, asc, ascq, + cmd->result); + } else { /* scsi status is zero??? How??? */ + dev_warn(&h->pdev->dev, "cp %p SCSI status was 0. " + "Returning no connection.\n", cp), + + /* Ordinarily, this case should never happen, + * but there is a bug in some released firmware + * revisions that allows it to happen if, for + * example, a 4100 backplane loses power and + * the tape drive is in it. We assume that + * it's a fatal error of some kind because we + * can't show that it wasn't. We will make it + * look like selection timeout since that is + * the most common reason for this to occur, + * and it's severe enough. + */ + + cmd->result = DID_NO_CONNECT << 16; + } + break; + + case CMD_DATA_UNDERRUN: /* let mid layer handle it. */ + break; + case CMD_DATA_OVERRUN: + dev_warn(&h->pdev->dev, "cp %p has" + " completed with data overrun " + "reported\n", cp); + break; + case CMD_INVALID: { + /* print_bytes(cp, sizeof(*cp), 1, 0); + print_cmd(cp); */ + /* We get CMD_INVALID if you address a non-existent device + * instead of a selection timeout (no response). You will + * see this if you yank out a drive, then try to access it. + * This is kind of a shame because it means that any other + * CMD_INVALID (e.g. driver bug) will get interpreted as a + * missing target. */ + cmd->result = DID_NO_CONNECT << 16; + } + break; + case CMD_PROTOCOL_ERR: + dev_warn(&h->pdev->dev, "cp %p has " + "protocol error \n", cp); + break; + case CMD_HARDWARE_ERR: + cmd->result = DID_ERROR << 16; + dev_warn(&h->pdev->dev, "cp %p had hardware error\n", cp); + break; + case CMD_CONNECTION_LOST: + cmd->result = DID_ERROR << 16; + dev_warn(&h->pdev->dev, "cp %p had connection lost\n", cp); + break; + case CMD_ABORTED: + cmd->result = DID_ABORT << 16; + dev_warn(&h->pdev->dev, "cp %p was aborted with status 0x%x\n", + cp, ei->ScsiStatus); + break; + case CMD_ABORT_FAILED: + cmd->result = DID_ERROR << 16; + dev_warn(&h->pdev->dev, "cp %p reports abort failed\n", cp); + break; + case CMD_UNSOLICITED_ABORT: + cmd->result = DID_ABORT << 16; + dev_warn(&h->pdev->dev, "cp %p aborted do to an unsolicited " + "abort\n", cp); + break; + case CMD_TIMEOUT: + cmd->result = DID_TIME_OUT << 16; + dev_warn(&h->pdev->dev, "cp %p timedout\n", cp); + break; + default: + cmd->result = DID_ERROR << 16; + dev_warn(&h->pdev->dev, "cp %p returned unknown status %x\n", + cp, ei->CommandStatus); + } + cmd->scsi_done(cmd); + cmd_free(h, cp); +} + +static int hpsa_scsi_detect(struct ctlr_info *h) +{ + struct Scsi_Host *sh; + int error; + + sh = scsi_host_alloc(&hpsa_driver_template, sizeof(h)); + if (sh == NULL) + goto fail; + + sh->io_port = 0; + sh->n_io_port = 0; + sh->this_id = -1; + sh->max_channel = 3; + sh->max_cmd_len = MAX_COMMAND_SIZE; + sh->max_lun = HPSA_MAX_LUN; + sh->max_id = HPSA_MAX_LUN; + h->scsi_host = sh; + sh->hostdata[0] = (unsigned long) h; + sh->irq = h->intr[SIMPLE_MODE_INT]; + sh->unique_id = sh->irq; + error = scsi_add_host(sh, &h->pdev->dev); + if (error) + goto fail_host_put; + scsi_scan_host(sh); + return 0; + + fail_host_put: + dev_err(&h->pdev->dev, "hpsa_scsi_detect: scsi_add_host" + " failed for controller %d\n", h->ctlr); + scsi_host_put(sh); + return -1; + fail: + dev_err(&h->pdev->dev, "hpsa_scsi_detect: scsi_host_alloc" + " failed for controller %d\n", h->ctlr); + return -1; +} + +static void hpsa_pci_unmap(struct pci_dev *pdev, + struct CommandList *c, int sg_used, int data_direction) +{ + int i; + union u64bit addr64; + + for (i = 0; i < sg_used; i++) { + addr64.val32.lower = c->SG[i].Addr.lower; + addr64.val32.upper = c->SG[i].Addr.upper; + pci_unmap_single(pdev, (dma_addr_t) addr64.val, c->SG[i].Len, + data_direction); + } +} + +static void hpsa_map_one(struct pci_dev *pdev, + struct CommandList *cp, + unsigned char *buf, + size_t buflen, + int data_direction) +{ + __u64 addr64; + + if (buflen == 0 || data_direction == PCI_DMA_NONE) { + cp->Header.SGList = 0; + cp->Header.SGTotal = 0; + return; + } + + addr64 = (__u64) pci_map_single(pdev, buf, buflen, data_direction); + cp->SG[0].Addr.lower = + (__u32) (addr64 & (__u64) 0x00000000FFFFFFFF); + cp->SG[0].Addr.upper = + (__u32) ((addr64 >> 32) & (__u64) 0x00000000FFFFFFFF); + cp->SG[0].Len = buflen; + cp->Header.SGList = (__u8) 1; /* no. SGs contig in this cmd */ + cp->Header.SGTotal = (__u16) 1; /* total sgs in this cmd list */ +} + +static inline void hpsa_scsi_do_simple_cmd_core(struct ctlr_info *h, + struct CommandList *c) +{ + DECLARE_COMPLETION_ONSTACK(wait); + + c->waiting = &wait; + enqueue_cmd_and_start_io(h, c); + wait_for_completion(&wait); +} + +static void hpsa_scsi_do_simple_cmd_with_retry(struct ctlr_info *h, + struct CommandList *c, int data_direction) +{ + int retry_count = 0; + + do { + memset(c->err_info, 0, sizeof(c->err_info)); + hpsa_scsi_do_simple_cmd_core(h, c); + retry_count++; + } while (check_for_unit_attention(h, c) && retry_count <= 3); + hpsa_pci_unmap(h->pdev, c, 1, data_direction); +} + +static void hpsa_scsi_interpret_error(struct CommandList *cp) +{ + struct ErrorInfo *ei; + struct device *d = &cp->h->pdev->dev; + + ei = cp->err_info; + switch (ei->CommandStatus) { + case CMD_TARGET_STATUS: + dev_warn(d, "cmd %p has completed with errors\n", cp); + dev_warn(d, "cmd %p has SCSI Status = %x\n", cp, + ei->ScsiStatus); + if (ei->ScsiStatus == 0) + dev_warn(d, "SCSI status is abnormally zero. " + "(probably indicates selection timeout " + "reported incorrectly due to a known " + "firmware bug, circa July, 2001.)\n"); + break; + case CMD_DATA_UNDERRUN: /* let mid layer handle it. */ + dev_info(d, "UNDERRUN\n"); + break; + case CMD_DATA_OVERRUN: + dev_warn(d, "cp %p has completed with data overrun\n", cp); + break; + case CMD_INVALID: { + /* controller unfortunately reports SCSI passthru's + * to non-existent targets as invalid commands. + */ + dev_warn(d, "cp %p is reported invalid (probably means " + "target device no longer present)\n", cp); + /* print_bytes((unsigned char *) cp, sizeof(*cp), 1, 0); + print_cmd(cp); */ + } + break; + case CMD_PROTOCOL_ERR: + dev_warn(d, "cp %p has protocol error \n", cp); + break; + case CMD_HARDWARE_ERR: + /* cmd->result = DID_ERROR << 16; */ + dev_warn(d, "cp %p had hardware error\n", cp); + break; + case CMD_CONNECTION_LOST: + dev_warn(d, "cp %p had connection lost\n", cp); + break; + case CMD_ABORTED: + dev_warn(d, "cp %p was aborted\n", cp); + break; + case CMD_ABORT_FAILED: + dev_warn(d, "cp %p reports abort failed\n", cp); + break; + case CMD_UNSOLICITED_ABORT: + dev_warn(d, "cp %p aborted due to an unsolicited abort\n", cp); + break; + case CMD_TIMEOUT: + dev_warn(d, "cp %p timed out\n", cp); + break; + default: + dev_warn(d, "cp %p returned unknown status %x\n", cp, + ei->CommandStatus); + } +} + +static int hpsa_scsi_do_inquiry(struct ctlr_info *h, unsigned char *scsi3addr, + unsigned char page, unsigned char *buf, + unsigned char bufsize) +{ + int rc = IO_OK; + struct CommandList *c; + struct ErrorInfo *ei; + + c = cmd_special_alloc(h); + + if (c == NULL) { /* trouble... */ + dev_warn(&h->pdev->dev, "cmd_special_alloc returned NULL!\n"); + return -1; + } + + fill_cmd(c, HPSA_INQUIRY, h, buf, bufsize, page, scsi3addr, TYPE_CMD); + hpsa_scsi_do_simple_cmd_with_retry(h, c, PCI_DMA_FROMDEVICE); + ei = c->err_info; + if (ei->CommandStatus != 0 && ei->CommandStatus != CMD_DATA_UNDERRUN) { + hpsa_scsi_interpret_error(c); + rc = -1; + } + cmd_special_free(h, c); + return rc; +} + +static int hpsa_send_reset(struct ctlr_info *h, unsigned char *scsi3addr) +{ + int rc = IO_OK; + struct CommandList *c; + struct ErrorInfo *ei; + + c = cmd_special_alloc(h); + + if (c == NULL) { /* trouble... */ + dev_warn(&h->pdev->dev, "cmd_special_alloc returned NULL!\n"); + return -1; + } + + fill_cmd(c, HPSA_DEVICE_RESET_MSG, h, NULL, 0, 0, scsi3addr, TYPE_MSG); + hpsa_scsi_do_simple_cmd_core(h, c); + /* no unmap needed here because no data xfer. */ + + ei = c->err_info; + if (ei->CommandStatus != 0) { + hpsa_scsi_interpret_error(c); + rc = -1; + } + cmd_special_free(h, c); + return rc; +} + +static void hpsa_get_raid_level(struct ctlr_info *h, + unsigned char *scsi3addr, unsigned char *raid_level) +{ + int rc; + unsigned char *buf; + + *raid_level = RAID_UNKNOWN; + buf = kzalloc(64, GFP_KERNEL); + if (!buf) + return; + rc = hpsa_scsi_do_inquiry(h, scsi3addr, 0xC1, buf, 64); + if (rc == 0) + *raid_level = buf[8]; + if (*raid_level > RAID_UNKNOWN) + *raid_level = RAID_UNKNOWN; + kfree(buf); + return; +} + +/* Get the device id from inquiry page 0x83 */ +static int hpsa_get_device_id(struct ctlr_info *h, unsigned char *scsi3addr, + unsigned char *device_id, int buflen) +{ + int rc; + unsigned char *buf; + + if (buflen > 16) + buflen = 16; + buf = kzalloc(64, GFP_KERNEL); + if (!buf) + return -1; + rc = hpsa_scsi_do_inquiry(h, scsi3addr, 0x83, buf, 64); + if (rc == 0) + memcpy(device_id, &buf[8], buflen); + kfree(buf); + return rc != 0; +} + +static int hpsa_scsi_do_report_luns(struct ctlr_info *h, int logical, + struct ReportLUNdata *buf, int bufsize, + int extended_response) +{ + int rc = IO_OK; + struct CommandList *c; + unsigned char scsi3addr[8]; + struct ErrorInfo *ei; + + c = cmd_special_alloc(h); + if (c == NULL) { /* trouble... */ + dev_err(&h->pdev->dev, "cmd_special_alloc returned NULL!\n"); + return -1; + } + + memset(&scsi3addr[0], 0, 8); /* address the controller */ + + fill_cmd(c, logical ? HPSA_REPORT_LOG : HPSA_REPORT_PHYS, h, + buf, bufsize, 0, scsi3addr, TYPE_CMD); + if (extended_response) + c->Request.CDB[1] = extended_response; + hpsa_scsi_do_simple_cmd_with_retry(h, c, PCI_DMA_FROMDEVICE); + ei = c->err_info; + if (ei->CommandStatus != 0 && + ei->CommandStatus != CMD_DATA_UNDERRUN) { + hpsa_scsi_interpret_error(c); + rc = -1; + } + cmd_special_free(h, c); + return rc; +} + +static inline int hpsa_scsi_do_report_phys_luns(struct ctlr_info *h, + struct ReportLUNdata *buf, + int bufsize, int extended_response) +{ + return hpsa_scsi_do_report_luns(h, 0, buf, bufsize, extended_response); +} + +static inline int hpsa_scsi_do_report_log_luns(struct ctlr_info *h, + struct ReportLUNdata *buf, int bufsize) +{ + return hpsa_scsi_do_report_luns(h, 1, buf, bufsize, 0); +} + +static inline void hpsa_set_bus_target_lun(struct hpsa_scsi_dev_t *device, + int bus, int target, int lun) +{ + device->bus = bus; + device->target = target; + device->lun = lun; +} + +static int hpsa_update_device_info(struct ctlr_info *h, + unsigned char scsi3addr[], struct hpsa_scsi_dev_t *this_device) +{ +#define OBDR_TAPE_INQ_SIZE 49 + unsigned char *inq_buff = NULL; + + inq_buff = kmalloc(OBDR_TAPE_INQ_SIZE, GFP_KERNEL); + if (!inq_buff) + goto bail_out; + + memset(inq_buff, 0, OBDR_TAPE_INQ_SIZE); + /* Do an inquiry to the device to see what it is. */ + if (hpsa_scsi_do_inquiry(h, scsi3addr, 0, inq_buff, + (unsigned char) OBDR_TAPE_INQ_SIZE) != 0) { + /* Inquiry failed (msg printed already) */ + dev_err(&h->pdev->dev, + "hpsa_update_device_info: inquiry failed\n"); + goto bail_out; + } + + /* As a side effect, record the firmware version number + * if we happen to be talking to the RAID controller. + */ + if (is_hba_lunid(scsi3addr)) + memcpy(h->firm_ver, &inq_buff[32], 4); + + this_device->devtype = (inq_buff[0] & 0x1f); + memcpy(this_device->scsi3addr, scsi3addr, 8); + memcpy(this_device->vendor, &inq_buff[8], + sizeof(this_device->vendor)); + memcpy(this_device->model, &inq_buff[16], + sizeof(this_device->model)); + memcpy(this_device->revision, &inq_buff[32], + sizeof(this_device->revision)); + memset(this_device->device_id, 0, + sizeof(this_device->device_id)); + hpsa_get_device_id(h, scsi3addr, this_device->device_id, + sizeof(this_device->device_id)); + + if (this_device->devtype == TYPE_DISK && + is_logical_dev_addr_mode(scsi3addr)) + hpsa_get_raid_level(h, scsi3addr, &this_device->raid_level); + else + this_device->raid_level = RAID_UNKNOWN; + + kfree(inq_buff); + return 0; + +bail_out: + kfree(inq_buff); + return 1; +} + +static unsigned char *msa2xxx_model[] = { + "MSA2012", + "MSA2024", + "MSA2312", + "MSA2324", + NULL, +}; + +static int is_msa2xxx(struct ctlr_info *h, struct hpsa_scsi_dev_t *device) +{ + int i; + + for (i = 0; msa2xxx_model[i]; i++) + if (strncmp(device->model, msa2xxx_model[i], + strlen(msa2xxx_model[i])) == 0) + return 1; + return 0; +} + +/* Helper function to assign bus, target, lun mapping of devices. + * Puts non-msa2xxx logical volumes on bus 0, msa2xxx logical + * volumes on bus 1, physical devices on bus 2. and the hba on bus 3. + * Logical drive target and lun are assigned at this time, but + * physical device lun and target assignment are deferred (assigned + * in hpsa_find_target_lun, called by hpsa_scsi_add_entry.) + */ +static void figure_bus_target_lun(struct ctlr_info *h, + __u8 *lunaddrbytes, int *bus, int *target, int *lun, + struct hpsa_scsi_dev_t *device) +{ + + __u32 lunid; + + if (is_logical_dev_addr_mode(lunaddrbytes)) { + /* logical device */ + memcpy(&lunid, lunaddrbytes, sizeof(lunid)); + lunid = le32_to_cpu(lunid); + + if (is_msa2xxx(h, device)) { + *bus = 1; + *target = (lunid >> 16) & 0x3fff; + *lun = lunid & 0x00ff; + } else { + *bus = 0; + *lun = 0; + *target = lunid & 0x3fff; + } + } else { + /* physical device */ + if (is_hba_lunid(lunaddrbytes)) + *bus = 3; + else + *bus = 2; + *target = -1; + *lun = -1; /* we will fill these in later. */ + } +} + +/* + * If there is no lun 0 on a target, linux won't find any devices. + * For the MSA2xxx boxes, we have to manually detect the enclosure + * which is at lun zero, as CCISS_REPORT_PHYSICAL_LUNS doesn't report + * it for some reason. *tmpdevice is the target we're adding, + * this_device is a pointer into the current element of currentsd[] + * that we're building up in update_scsi_devices(), below. + * lunzerobits is a bitmap that tracks which targets already have a + * lun 0 assigned. + * Returns 1 if an enclosure was added, 0 if not. + */ +static int add_msa2xxx_enclosure_device(struct ctlr_info *h, + struct hpsa_scsi_dev_t *tmpdevice, + struct hpsa_scsi_dev_t *this_device, __u8 *lunaddrbytes, + int bus, int target, int lun, unsigned long lunzerobits[], + int *nmsa2xxx_enclosures) +{ + unsigned char scsi3addr[8]; + + if (test_bit(target, lunzerobits)) + return 0; /* There is already a lun 0 on this target. */ + + if (!is_logical_dev_addr_mode(lunaddrbytes)) + return 0; /* It's the logical targets that may lack lun 0. */ + + if (!is_msa2xxx(h, tmpdevice)) + return 0; /* It's only the MSA2xxx that have this problem. */ + + if (lun == 0) /* if lun is 0, then obviously we have a lun 0. */ + return 0; + + if (is_hba_lunid(scsi3addr)) + return 0; /* Don't add the RAID controller here. */ + +#define MAX_MSA2XXX_ENCLOSURES 32 + if (*nmsa2xxx_enclosures >= MAX_MSA2XXX_ENCLOSURES) { + dev_warn(&h->pdev->dev, "Maximum number of MSA2XXX " + "enclosures exceeded. Check your hardware " + "configuration."); + return 0; + } + + memset(scsi3addr, 0, 8); + scsi3addr[3] = target; + if (hpsa_update_device_info(h, scsi3addr, this_device)) + return 0; + (*nmsa2xxx_enclosures)++; + hpsa_set_bus_target_lun(this_device, bus, target, 0); + set_bit(target, lunzerobits); + return 1; +} + +/* + * Do CISS_REPORT_PHYS and CISS_REPORT_LOG. Data is returned in physdev, + * logdev. The number of luns in physdev and logdev are returned in + * *nphysicals and *nlogicals, respectively. + * Returns 0 on success, -1 otherwise. + */ +static int hpsa_gather_lun_info(struct ctlr_info *h, + int reportlunsize, + struct ReportLUNdata *physdev, __u32 *nphysicals, + struct ReportLUNdata *logdev, __u32 *nlogicals) +{ + if (hpsa_scsi_do_report_phys_luns(h, physdev, reportlunsize, 0)) { + dev_err(&h->pdev->dev, "report physical LUNs failed.\n"); + return -1; + } + memcpy(nphysicals, &physdev->LUNListLength[0], sizeof(*nphysicals)); + *nphysicals = be32_to_cpu(*nphysicals) / 8; +#ifdef DEBUG + dev_info(&h->pdev->dev, "number of physical luns is %d\n", *nphysicals); +#endif + if (*nphysicals > HPSA_MAX_PHYS_LUN) { + dev_warn(&h->pdev->dev, "maximum physical LUNs (%d) exceeded." + " %d LUNs ignored.\n", HPSA_MAX_PHYS_LUN, + *nphysicals - HPSA_MAX_PHYS_LUN); + *nphysicals = HPSA_MAX_PHYS_LUN; + } + if (hpsa_scsi_do_report_log_luns(h, logdev, reportlunsize)) { + dev_err(&h->pdev->dev, "report logical LUNs failed.\n"); + return -1; + } + memcpy(nlogicals, &logdev->LUNListLength[0], sizeof(*nlogicals)); + *nlogicals = be32_to_cpu(*nlogicals) / 8; +#ifdef DEBUG + dev_info(&h->pdev->dev, "number of logical luns is %d\n", *nlogicals); +#endif + /* Reject Logicals in excess of our max capability. */ + if (*nlogicals > HPSA_MAX_LUN) { + dev_warn(&h->pdev->dev, + "maximum logical LUNs (%d) exceeded. " + "%d LUNs ignored.\n", HPSA_MAX_LUN, + *nlogicals - HPSA_MAX_LUN); + *nlogicals = HPSA_MAX_LUN; + } + if (*nlogicals + *nphysicals > HPSA_MAX_PHYS_LUN) { + dev_warn(&h->pdev->dev, + "maximum logical + physical LUNs (%d) exceeded. " + "%d LUNs ignored.\n", HPSA_MAX_PHYS_LUN, + *nphysicals + *nlogicals - HPSA_MAX_PHYS_LUN); + *nlogicals = HPSA_MAX_PHYS_LUN - *nphysicals; + } + return 0; +} + +static void hpsa_update_scsi_devices(struct ctlr_info *h, int hostno) +{ + /* the idea here is we could get notified + * that some devices have changed, so we do a report + * physical luns and report logical luns cmd, and adjust + * our list of devices accordingly. + * + * The scsi3addr's of devices won't change so long as the + * adapter is not reset. That means we can rescan and + * tell which devices we already know about, vs. new + * devices, vs. disappearing devices. + */ + struct ReportLUNdata *physdev_list = NULL; + struct ReportLUNdata *logdev_list = NULL; + unsigned char *inq_buff = NULL; + __u32 nphysicals = 0; + __u32 nlogicals = 0; + __u32 ndev_allocated = 0; + struct hpsa_scsi_dev_t **currentsd, *this_device, *tmpdevice; + int ncurrent = 0; + int reportlunsize = sizeof(*physdev_list) + HPSA_MAX_PHYS_LUN * 8; + int i, nmsa2xxx_enclosures, ndevs_to_allocate; + int bus, target, lun; + DECLARE_BITMAP(lunzerobits, HPSA_MAX_TARGETS_PER_CTLR); + + currentsd = kzalloc(sizeof(*currentsd) * HPSA_MAX_SCSI_DEVS_PER_HBA, + GFP_KERNEL); + physdev_list = kzalloc(reportlunsize, GFP_KERNEL); + logdev_list = kzalloc(reportlunsize, GFP_KERNEL); + inq_buff = kmalloc(OBDR_TAPE_INQ_SIZE, GFP_KERNEL); + tmpdevice = kzalloc(sizeof(*tmpdevice), GFP_KERNEL); + + if (!currentsd || !physdev_list || !logdev_list || + !inq_buff || !tmpdevice) { + dev_err(&h->pdev->dev, "out of memory\n"); + goto out; + } + memset(lunzerobits, 0, sizeof(lunzerobits)); + + if (hpsa_gather_lun_info(h, reportlunsize, physdev_list, &nphysicals, + logdev_list, &nlogicals)) + goto out; + + /* We might see up to 32 MSA2xxx enclosures, actually 8 of them + * but each of them 4 times through different paths. The plus 1 + * is for the RAID controller. + */ + ndevs_to_allocate = nphysicals + nlogicals + MAX_MSA2XXX_ENCLOSURES + 1; + + /* Allocate the per device structures */ + for (i = 0; i < ndevs_to_allocate; i++) { + currentsd[i] = kzalloc(sizeof(*currentsd[i]), GFP_KERNEL); + if (!currentsd[i]) { + dev_warn(&h->pdev->dev, "out of memory at %s:%d\n", + __FILE__, __LINE__); + goto out; + } + ndev_allocated++; + } + + /* adjust our table of devices */ + nmsa2xxx_enclosures = 0; + for (i = 0; i < nphysicals + nlogicals + 1; i++) { + __u8 *lunaddrbytes; + + /* Figure out where the LUN ID info is coming from */ + if (i < nphysicals) + lunaddrbytes = &physdev_list->LUN[i][0]; + else + if (i < nphysicals + nlogicals) + lunaddrbytes = + &logdev_list->LUN[i-nphysicals][0]; + else /* jam in the RAID controller at the end */ + lunaddrbytes = RAID_CTLR_LUNID; + + /* skip masked physical devices. */ + if (lunaddrbytes[3] & 0xC0 && i < nphysicals) + continue; + + /* Get device type, vendor, model, device id */ + if (hpsa_update_device_info(h, lunaddrbytes, tmpdevice)) + continue; /* skip it if we can't talk to it. */ + figure_bus_target_lun(h, lunaddrbytes, &bus, &target, &lun, + tmpdevice); + this_device = currentsd[ncurrent]; + + /* + * For the msa2xxx boxes, we have to insert a LUN 0 which + * doesn't show up in CCISS_REPORT_PHYSICAL data, but there + * is nonetheless an enclosure device there. We have to + * present that otherwise linux won't find anything if + * there is no lun 0. + */ + if (add_msa2xxx_enclosure_device(h, tmpdevice, this_device, + lunaddrbytes, bus, target, lun, lunzerobits, + &nmsa2xxx_enclosures)) { + ncurrent++; + this_device = currentsd[ncurrent]; + } + + *this_device = *tmpdevice; + hpsa_set_bus_target_lun(this_device, bus, target, lun); + + switch (this_device->devtype) { + case TYPE_ROM: { + /* We don't *really* support actual CD-ROM devices, + * just "One Button Disaster Recovery" tape drive + * which temporarily pretends to be a CD-ROM drive. + * So we check that the device is really an OBDR tape + * device by checking for "$DR-10" in bytes 43-48 of + * the inquiry data. + */ + char obdr_sig[7]; +#define OBDR_TAPE_SIG "$DR-10" + strncpy(obdr_sig, &inq_buff[43], 6); + obdr_sig[6] = '\0'; + if (strncmp(obdr_sig, OBDR_TAPE_SIG, 6) != 0) + /* Not OBDR device, ignore it. */ + break; + } + ncurrent++; + break; + case TYPE_DISK: + if (i < nphysicals) + break; + ncurrent++; + break; + case TYPE_TAPE: + case TYPE_MEDIUM_CHANGER: + ncurrent++; + break; + case TYPE_RAID: + /* Only present the Smartarray HBA as a RAID controller. + * If it's a RAID controller other than the HBA itself + * (an external RAID controller, MSA500 or similar) + * don't present it. + */ + if (!is_hba_lunid(lunaddrbytes)) + break; + ncurrent++; + break; + default: + break; + } + if (ncurrent >= HPSA_MAX_SCSI_DEVS_PER_HBA) + break; + } + adjust_hpsa_scsi_table(h, hostno, currentsd, ncurrent); +out: + kfree(tmpdevice); + for (i = 0; i < ndev_allocated; i++) + kfree(currentsd[i]); + kfree(currentsd); + kfree(inq_buff); + kfree(physdev_list); + kfree(logdev_list); + return; +} + +/* hpsa_scatter_gather takes a struct scsi_cmnd, (cmd), and does the pci + * dma mapping and fills in the scatter gather entries of the + * hpsa command, cp. + */ +static int hpsa_scatter_gather(struct pci_dev *pdev, + struct CommandList *cp, + struct scsi_cmnd *cmd) +{ + unsigned int len; + struct scatterlist *sg; + __u64 addr64; + int use_sg, i; + + BUG_ON(scsi_sg_count(cmd) > MAXSGENTRIES); + + use_sg = scsi_dma_map(cmd); + if (use_sg < 0) + return use_sg; + + if (!use_sg) + goto sglist_finished; + + scsi_for_each_sg(cmd, sg, use_sg, i) { + addr64 = (__u64) sg_dma_address(sg); + len = sg_dma_len(sg); + cp->SG[i].Addr.lower = + (__u32) (addr64 & (__u64) 0x00000000FFFFFFFF); + cp->SG[i].Addr.upper = + (__u32) ((addr64 >> 32) & (__u64) 0x00000000FFFFFFFF); + cp->SG[i].Len = len; + cp->SG[i].Ext = 0; /* we are not chaining */ + } + +sglist_finished: + + cp->Header.SGList = (__u8) use_sg; /* no. SGs contig in this cmd */ + cp->Header.SGTotal = (__u16) use_sg; /* total sgs in this cmd list */ + return 0; +} + + +static int hpsa_scsi_queue_command(struct scsi_cmnd *cmd, + void (*done)(struct scsi_cmnd *)) +{ + struct ctlr_info *h; + struct hpsa_scsi_dev_t *dev; + unsigned char scsi3addr[8]; + struct CommandList *c; + unsigned long flags; + + /* Get the ptr to our adapter structure out of cmd->host. */ + h = sdev_to_hba(cmd->device); + dev = cmd->device->hostdata; + if (!dev) { + cmd->result = DID_NO_CONNECT << 16; + done(cmd); + return 0; + } + memcpy(scsi3addr, dev->scsi3addr, sizeof(scsi3addr)); + + /* Need a lock as this is being allocated from the pool */ + spin_lock_irqsave(&h->lock, flags); + c = cmd_alloc(h); + spin_unlock_irqrestore(&h->lock, flags); + if (c == NULL) { /* trouble... */ + dev_err(&h->pdev->dev, "cmd_alloc returned NULL!\n"); + cmd->result = DID_NO_CONNECT << 16; + done(cmd); + return 0; + } + + /* Fill in the command list header */ + + cmd->scsi_done = done; /* save this for use by completion code */ + + /* save c in case we have to abort it */ + cmd->host_scribble = (unsigned char *) c; + + c->cmd_type = CMD_SCSI; + c->scsi_cmd = cmd; + c->Header.ReplyQueue = 0; /* unused in simple mode */ + memcpy(&c->Header.LUN.LunAddrBytes[0], &scsi3addr[0], 8); + c->Header.Tag.lower = c->busaddr; /* Use k. address of cmd as tag */ + + /* Fill in the request block... */ + + c->Request.Timeout = 0; + memset(c->Request.CDB, 0, sizeof(c->Request.CDB)); + BUG_ON(cmd->cmd_len > sizeof(c->Request.CDB)); + c->Request.CDBLen = cmd->cmd_len; + memcpy(c->Request.CDB, cmd->cmnd, cmd->cmd_len); + c->Request.Type.Type = TYPE_CMD; + c->Request.Type.Attribute = ATTR_SIMPLE; + switch (cmd->sc_data_direction) { + case DMA_TO_DEVICE: + c->Request.Type.Direction = XFER_WRITE; + break; + case DMA_FROM_DEVICE: + c->Request.Type.Direction = XFER_READ; + break; + case DMA_NONE: + c->Request.Type.Direction = XFER_NONE; + break; + case DMA_BIDIRECTIONAL: + /* This can happen if a buggy application does a scsi passthru + * and sets both inlen and outlen to non-zero. ( see + * ../scsi/scsi_ioctl.c:scsi_ioctl_send_command() ) + */ + + c->Request.Type.Direction = XFER_RSVD; + /* This is technically wrong, and hpsa controllers should + * reject it with CMD_INVALID, which is the most correct + * response, but non-fibre backends appear to let it + * slide by, and give the same results as if this field + * were set correctly. Either way is acceptable for + * our purposes here. + */ + + break; + + default: + dev_err(&h->pdev->dev, "unknown data direction: %d\n", + cmd->sc_data_direction); + BUG(); + break; + } + + if (hpsa_scatter_gather(h->pdev, c, cmd) < 0) { /* Fill SG list */ + cmd_free(h, c); + return SCSI_MLQUEUE_HOST_BUSY; + } + enqueue_cmd_and_start_io(h, c); + /* the cmd'll come back via intr handler in complete_scsi_command() */ + return 0; +} + +static void hpsa_unregister_scsi(struct ctlr_info *h) +{ + /* we are being forcibly unloaded, and may not refuse. */ + scsi_remove_host(h->scsi_host); + scsi_host_put(h->scsi_host); + h->scsi_host = NULL; +} + +static int hpsa_register_scsi(struct ctlr_info *h) +{ + int rc; + + hpsa_update_scsi_devices(h, -1); + rc = hpsa_scsi_detect(h); + if (rc != 0) + dev_err(&h->pdev->dev, "hpsa_register_scsi: failed" + " hpsa_scsi_detect(), rc is %d\n", rc); + return rc; +} + +static int wait_for_device_to_become_ready(struct ctlr_info *h, + unsigned char lunaddr[]) +{ + int rc = 0; + int count = 0; + int waittime = HZ; + struct CommandList *c; + + c = cmd_special_alloc(h); + if (!c) { + dev_warn(&h->pdev->dev, "out of memory in " + "wait_for_device_to_become_ready.\n"); + return IO_ERROR; + } + + /* Send test unit ready until device ready, or give up. */ + while (count < HPSA_TUR_RETRY_LIMIT) { + + /* Wait for a bit. do this first, because if we send + * the TUR right away, the reset will just abort it. + */ + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_timeout(waittime); + count++; + + /* Increase wait time with each try, up to a point. */ + if (waittime < (HZ * HPSA_MAX_WAIT_INTERVAL_SECS)) + waittime = waittime * 2; + + /* Send the Test Unit Ready */ + fill_cmd(c, TEST_UNIT_READY, h, NULL, 0, 0, lunaddr, TYPE_CMD); + hpsa_scsi_do_simple_cmd_core(h, c); + /* no unmap needed here because no data xfer. */ + + if (c->err_info->CommandStatus == CMD_SUCCESS) + break; + + if (c->err_info->CommandStatus == CMD_TARGET_STATUS && + c->err_info->ScsiStatus == SAM_STAT_CHECK_CONDITION && + (c->err_info->SenseInfo[2] == NO_SENSE || + c->err_info->SenseInfo[2] == UNIT_ATTENTION)) + break; + + dev_warn(&h->pdev->dev, "waiting %d secs " + "for device to become ready.\n", waittime / HZ); + rc = 1; /* device not ready. */ + } + + if (rc) + dev_warn(&h->pdev->dev, "giving up on device.\n"); + else + dev_warn(&h->pdev->dev, "device is ready.\n"); + + cmd_special_free(h, c); + return rc; +} + +/* Need at least one of these error handlers to keep ../scsi/hosts.c from + * complaining. Doing a host- or bus-reset can't do anything good here. + */ +static int hpsa_eh_device_reset_handler(struct scsi_cmnd *scsicmd) +{ + int rc; + struct ctlr_info *h; + struct hpsa_scsi_dev_t *dev; + + /* find the controller to which the command to be aborted was sent */ + h = sdev_to_hba(scsicmd->device); + if (h == NULL) /* paranoia */ + return FAILED; + dev_warn(&h->pdev->dev, "resetting drive\n"); + + dev = scsicmd->device->hostdata; + if (!dev) { + dev_err(&h->pdev->dev, "hpsa_eh_device_reset_handler: " + "device lookup failed.\n"); + return FAILED; + } + /* send a reset to the SCSI LUN which the command was sent to */ + rc = hpsa_send_reset(h, dev->scsi3addr); + if (rc == 0 && wait_for_device_to_become_ready(h, dev->scsi3addr) == 0) + return SUCCESS; + + dev_warn(&h->pdev->dev, "resetting device failed.\n"); + return FAILED; +} + +/* + * For operations that cannot sleep, a command block is allocated at init, + * and managed by cmd_alloc() and cmd_free() using a simple bitmap to track + * which ones are free or in use. Lock must be held when calling this. + * cmd_free() is the complement. + */ +static struct CommandList *cmd_alloc(struct ctlr_info *h) +{ + struct CommandList *c; + int i; + union u64bit temp64; + dma_addr_t cmd_dma_handle, err_dma_handle; + + do { + i = find_first_zero_bit(h->cmd_pool_bits, h->nr_cmds); + if (i == h->nr_cmds) + return NULL; + } while (test_and_set_bit + (i & (BITS_PER_LONG - 1),
From: scameron on 16 Nov 2009 17:00
On Mon, Nov 16, 2009 at 03:06:49PM -0600, James Bottomley wrote: > On Wed, 2009-11-11 at 10:50 -0600, Stephen M. Cameron wrote: > > The following series implements hpsa scsi driver for HP Smart Arrays, > > and some updates since the last time. > > The first 5 patches in the series are already in Andrew Morton's tree. > > > > --- > > > > Andrew Morton (1): > > avoid helpful cleanup patches. > > > > Stephen M. Cameron (16): > > hpsa: fix typo that causes scsi status to be lost > > hpsa: Make fill_cmd() return void > > hpsa: Remove sendcmd, in no case are we required to poll for completions. > > hpsa: Flush cache with interrupts still enabled. > > hpsa: Retry driver initiated commands on unit attention > > hpsa: decode unit attention condition and retry commands. > > hpsa: Make hpsa_sdev_attrs static > > hpsa: Allow device rescan to be triggered via sysfs. > > Add thread to allow controllers to register for rescan for new devices > > hpsa: Factor out some pci_unmap code > > hpsa: Factor out command submission sequence > > hpsa: Use shost_priv instead of accessing host->hostdata[0] directly. > > hpsa: Allocate the correct amount of extra space for the scsi host > > Fix use of unallocated memory for MSA2xxx enclosure device data. > > hpsa: Fix vendor id check > > Add hpsa driver for HP Smart Array controllers. > > Actually, it's pretty difficult to review a 17 patch series like this > because the human mind (or at least mine) doesn't retain sufficient > context from patch to patch. I ended up just pulling all 17 into a tree > and reviewing the finished driver. > I was trying to err on the safe side, in case anyone wanted the separate patches, on the grounds that it's easier to bake a cake than to unbake one. If it's preferred, I can pre-bake them into one big patch before sending next time. Also, not all the patches are authored by me, so I figured I ought to preserve that. > That said: > > in hpsa.c: > > > static struct device_attribute *hpsa_shost_attrs[] = { > > &dev_attr_rescan, > > NULL, > > }; > > We already have a host scan attribute which (admittedly using the > transport class logic) you can plug into ... can't you just use it? Maybe. I'll take a look at it. > It > supplies user context, so you could dispense with all that scan thread > stuff as well, I think. > > > static DEFINE_MUTEX(scan_mutex); > > static LIST_HEAD(scan_q); > > static int scan_thread(void *data); > > These names are too generic. We already have a scan_mutex at least > defined at the top level. I know they're protected by static, but that > doesn't necessarily help if they show up in a debug stack trace. Ok. > > All of this report luns stuff looks fairly identical to the report luns > we do in scsi_scan.c ... barring the initial command, which could be > translated. Wouldn't it be easier to have the generic code parse and do > all of this? > Maybe. There are complications. We present some physical devices (tape drives, changers, the RAID controller) some logical (logical drives), the RAID controller is scsi revision 0, not 5, if we present it first, the REPORT_LUNS is never sent by the upper layers because of this, and only 7 devices get discovered, etc. If we present the first logical drive first, it sends the REPORT LUNS to the the drive, etc. It gets a little weird, as our hardware is a little weird. If it is possible, I doubt I would be able to have such a change ready by 2.6.33 timeframe. > > static int hpsa_scsi_queue_command(struct scsi_cmnd *cmd, > > void (*done)(struct scsi_cmnd *)) > > { > [...] > > c = cmd_alloc(h); > > spin_unlock_irqrestore(&h->lock, flags); > > if (c == NULL) { /* trouble... */ > > dev_err(&h->pdev->dev, "cmd_alloc returned NULL!\n"); > > cmd->result = DID_NO_CONNECT << 16; > > done(cmd); > > return 0; > > } > > I think you want to return SCSI_MLQUEUE_HOST_BUSY here, which will > trigger a throttle and retry after either something frees or I/O > pressure builds more. Ok. > > > static int hpsa_eh_device_reset_handler(struct scsi_cmnd *scsicmd) > > { > [...] > > rc = hpsa_send_reset(h, dev->scsi3addr); > > if (rc == 0 && wait_for_device_to_become_ready(h, dev->scsi3addr) == 0) > > return SUCCESS; > > So the first thing we do after a device reset successful return is send > a test unit ready to the failing device ... there's no real need for you > to duplicate that, is there? It didn't seem to work until I did that. this was awhile ago that I wrote this so my memory of it isn't too clear, but as I recall, I tested this (in cciss, which is what this code evolved from) with a tape drive with the long timeout set rather short, and "mt -f /dev/st0 erase" The artificially shortened timeout would mean that the erase would time out, and this reset code would get triggered. I found that without my change, the device didn't become ready in time, even though the upper layers sent TURs, and the device would get kicked off. So I put in that code to make it work. Maybe it was the mid layer that needed tweaking? maybe the slow to recover tape drives deserved to get kicked off? > > The ioctl stuff looks like you could do it all with SG_IO now rather > than rolling your own versions ... or is there some backward > compatibility problem here? > Yeah, mainly backward compatibility. HP's array config utility (ACU), snmp storage agents, firmware flashing utilities, cciss_vol_status, arrayprobe, to name a few things, depend on the passthrough ioctl and there is more complete error information returned by the passthrough. I also have a vague recollection of some buffer size limitation with SG_IO I'm not sure, might be wrong about that, but I know flashing firmware takes some large buffers. > > > static __devinit int hpsa_hard_reset_controller(struct pci_dev *pdev) > > { > [...] > > set_current_state(TASK_UNINTERRUPTIBLE); > > schedule_timeout(HZ >> 1); > > msleep(500) please .. This isn't the only place this occurs, could you > replace all of them? Ok. > > in hpsa_cmd.h: > > > /* Unit Attentions ASC's as defined for the MSA2012sa */ > > #define POWER_OR_RESET 0x29 > > #define STATE_CHANGED 0x2a > > #define UNIT_ATTENTION_CLEARED 0x2f > > #define LUN_FAILED 0x3e > > #define REPORT_LUNS_CHANGED 0x3f > > > > /* Unit Attentions ASCQ's as defined for the MSA2012sa */ > > > > /* These ASCQ's defined for ASC = POWER_OR_RESET */ > > #define POWER_ON_RESET 0x00 > > #define POWER_ON_REBOOT 0x01 > > #define SCSI_BUS_RESET 0x02 > > #define MSA_TARGET_RESET 0x03 > > #define CONTROLLER_FAILOVER 0x04 > > #define TRANSCEIVER_SE 0x05 > > #define TRANSCEIVER_LVD 0x06 > > > > /* These ASCQ's defined for ASC = STATE_CHANGED */ > > #define RESERVATION_PREEMPTED 0x03 > > #define ASYM_ACCESS_CHANGED 0x06 > > #define LUN_CAPACITY_CHANGED 0x09 > > Traditionally we've shied away from putting ASC/ASCQ values into > defines ... but these all look to be global not hpsa local, so they > should be in a common central file. > Ok. I think there are some in there we didn't even use. We should probably omit them in those cases. > Otherwise looks OK to a cursory glance. > > James > -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |