[SCSI] Add helper code so transport classes/driver can control queueing (v3)

SCSI-ml manages the queueing limits for the device and host, but
does not do so at the target level. However something something similar
can come in userful when a driver is transitioning a transport object to
the the blocked state, becuase at that time we do not want to queue
io and we do not want the queuecommand to be called again.

The patch adds code similar to the exisiting SCSI_ML_*BUSY handlers.
You can now return SCSI_MLQUEUE_TARGET_BUSY when we hit
a transport level queueing issue like the hw cannot allocate some
resource at the iscsi session/connection level, or the target has temporarily
closed or shrunk the queueing window, or if we are transitioning
to the blocked state.

bnx2i, when they rework their firmware according to netdev
developers requests, will also need to be able to limit queueing at this
level. bnx2i will hook into libiscsi, but will allocate a scsi host per
netdevice/hba, so unlike pure software iscsi/iser which is allocating
a host per session, it cannot set the scsi_host->can_queue and return
SCSI_MLQUEUE_HOST_BUSY to reflect queueing limits on the transport.

The iscsi class/driver can also set a scsi_target->can_queue value which
reflects the max commands the driver/class can support. For iscsi this
reflects the number of commands we can support for each session due to
session/connection hw limits, driver limits, and to also reflect the
session/targets's queueing window.

Changes:
v1 - initial patch.
v2 - Fix scsi_run_queue handling of multiple blocked targets.
Previously we would break from the main loop if a device was added back on
the starved list. We now run over the list and check if any target is
blocked.
v3 - Rediff for scsi-misc.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>

5 files changed, 108 insertions, 18 deletions

diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c
index 2ac3cb2b9081..f8b79d401d58 100644
--- a/drivers/scsi/scsi.c
+++ b/drivers/scsi/scsi.c
@@ -754,8 +754,12 @@ int scsi_dispatch_cmd(struct scsi_cmnd *cmd)
         }
         spin_unlock_irqrestore(host->host_lock, flags);
         if (rtn) {
-                scsi_queue_insert(cmd, (rtn == SCSI_MLQUEUE_DEVICE_BUSY) ?
-                                                rtn : SCSI_MLQUEUE_HOST_BUSY);
+                if (rtn != SCSI_MLQUEUE_DEVICE_BUSY &&
+                    rtn != SCSI_MLQUEUE_TARGET_BUSY)
+                        rtn = SCSI_MLQUEUE_HOST_BUSY;
+
+                scsi_queue_insert(cmd, rtn);
+
                 SCSI_LOG_MLQUEUE(3,
                     printk("queuecommand : request rejected\n"));
         }
@@ -800,6 +804,7 @@ static struct scsi_driver *scsi_cmd_to_driver(struct scsi_cmnd *cmd)
 void scsi_finish_command(struct scsi_cmnd *cmd)
 {
         struct scsi_device *sdev = cmd->device;
+        struct scsi_target *starget = scsi_target(sdev);
         struct Scsi_Host *shost = sdev->host;
         struct scsi_driver *drv;
         unsigned int good_bytes;
@@ -815,6 +820,7 @@ void scsi_finish_command(struct scsi_cmnd *cmd)
          * XXX(hch): What about locking?
          */
         shost->host_blocked = 0;
+        starget->target_blocked = 0;
         sdev->device_blocked = 0;
 
         /*
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 98ee55ced592..91c74c55aa5e 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -114,6 +114,7 @@ int scsi_queue_insert(struct scsi_cmnd *cmd, int reason)
 {
         struct Scsi_Host *host = cmd->device->host;
         struct scsi_device *device = cmd->device;
+        struct scsi_target *starget = scsi_target(device);
         struct request_queue *q = device->request_queue;
         unsigned long flags;
 
@@ -133,10 +134,17 @@ int scsi_queue_insert(struct scsi_cmnd *cmd, int reason)
          * if a command is requeued with no other commands outstanding
          * either for the device or for the host.
          */
-        if (reason == SCSI_MLQUEUE_HOST_BUSY)
+        switch (reason) {
+        case SCSI_MLQUEUE_HOST_BUSY:
                 host->host_blocked = host->max_host_blocked;
-        else if (reason == SCSI_MLQUEUE_DEVICE_BUSY)
+                break;
+        case SCSI_MLQUEUE_DEVICE_BUSY:
                 device->device_blocked = device->max_device_blocked;
+                break;
+        case SCSI_MLQUEUE_TARGET_BUSY:
+                starget->target_blocked = starget->max_target_blocked;
+                break;
+        }
 
         /*
          * Decrement the counters, since these commands are no longer
@@ -460,10 +468,12 @@ static void scsi_init_cmd_errh(struct scsi_cmnd *cmd)
 void scsi_device_unbusy(struct scsi_device *sdev)
 {
         struct Scsi_Host *shost = sdev->host;
+        struct scsi_target *starget = scsi_target(sdev);
         unsigned long flags;
 
         spin_lock_irqsave(shost->host_lock, flags);
         shost->host_busy--;
+        starget->target_busy--;
         if (unlikely(scsi_host_in_recovery(shost) &&
                      (shost->host_failed || shost->host_eh_scheduled)))
                 scsi_eh_wakeup(shost);
@@ -519,6 +529,13 @@ static void scsi_single_lun_run(struct scsi_device *current_sdev)
         spin_unlock_irqrestore(shost->host_lock, flags);
 }
 
+static inline int scsi_target_is_busy(struct scsi_target *starget)
+{
+        return ((starget->can_queue > 0 &&
+                 starget->target_busy >= starget->can_queue) ||
+                 starget->target_blocked);
+}
+
 /*
  * Function:    scsi_run_queue()
  *
@@ -533,7 +550,7 @@ static void scsi_single_lun_run(struct scsi_device *current_sdev)
  */
 static void scsi_run_queue(struct request_queue *q)
 {
-        struct scsi_device *sdev = q->queuedata;
+        struct scsi_device *starved_head = NULL, *sdev = q->queuedata;
         struct Scsi_Host *shost = sdev->host;
         unsigned long flags;
 
@@ -560,6 +577,21 @@ static void scsi_run_queue(struct request_queue *q)
                  */
                 sdev = list_entry(shost->starved_list.next,
                                           struct scsi_device, starved_entry);
+                /*
+                 * The *queue_ready functions can add a device back onto the
+                 * starved list's tail, so we must check for a infinite loop.
+                 */
+                if (sdev == starved_head)
+                        break;
+                if (!starved_head)
+                        starved_head = sdev;
+
+                if (scsi_target_is_busy(scsi_target(sdev))) {
+                        list_move_tail(&sdev->starved_entry,
+                                       &shost->starved_list);
+                        continue;
+                }
+
                 list_del_init(&sdev->starved_entry);
                 spin_unlock(shost->host_lock);
 
@@ -575,13 +607,6 @@ static void scsi_run_queue(struct request_queue *q)
                 spin_unlock(sdev->request_queue->queue_lock);
 
                 spin_lock(shost->host_lock);
-                if (unlikely(!list_empty(&sdev->starved_entry)))
-                        /*
-                         * sdev lost a race, and was put back on the
-                         * starved list. This is unlikely but without this
-                         * in theory we could loop forever.
-                         */
-                        break;
         }
         spin_unlock_irqrestore(shost->host_lock, flags);
 
@@ -1344,6 +1369,52 @@ static inline int scsi_dev_queue_ready(struct request_queue *q,
         return 1;
 }
 
+
+/*
+ * scsi_target_queue_ready: checks if there we can send commands to target
+ * @sdev: scsi device on starget to check.
+ *
+ * Called with the host lock held.
+ */
+static inline int scsi_target_queue_ready(struct Scsi_Host *shost,
+                                           struct scsi_device *sdev)
+{
+        struct scsi_target *starget = scsi_target(sdev);
+
+        if (starget->single_lun) {
+                if (starget->starget_sdev_user &&
+                    starget->starget_sdev_user != sdev)
+                        return 0;
+                starget->starget_sdev_user = sdev;
+        }
+
+        if (starget->target_busy == 0 && starget->target_blocked) {
+                /*
+                 * unblock after target_blocked iterates to zero
+                 */
+                if (--starget->target_blocked == 0) {
+                        SCSI_LOG_MLQUEUE(3, starget_printk(KERN_INFO, starget,
+                                         "unblocking target at zero depth\n"));
+                } else {
+                        blk_plug_device(sdev->request_queue);
+                        return 0;
+                }
+        }
+
+        if (scsi_target_is_busy(starget)) {
+                if (list_empty(&sdev->starved_entry)) {
+                        list_add_tail(&sdev->starved_entry,
+                                      &shost->starved_list);
+                        return 0;
+                }
+        }
+
+        /* We're OK to process the command, so we can't be starved */
+        if (!list_empty(&sdev->starved_entry))
+                list_del_init(&sdev->starved_entry);
+        return 1;
+}
+
 /*
  * scsi_host_queue_ready: if we can send requests to shost, return 1 else
  * return 0. We must end up running the queue again whenever 0 is
@@ -1390,6 +1461,7 @@ static void scsi_kill_request(struct request *req, struct request_queue *q)
 {
         struct scsi_cmnd *cmd = req->special;
         struct scsi_device *sdev = cmd->device;
+        struct scsi_target *starget = scsi_target(sdev);
         struct Scsi_Host *shost = sdev->host;
 
         blkdev_dequeue_request(req);
@@ -1413,6 +1485,7 @@ static void scsi_kill_request(struct request *req, struct request_queue *q)
         spin_unlock(sdev->request_queue->queue_lock);
         spin_lock(shost->host_lock);
         shost->host_busy++;
+        starget->target_busy++;
         spin_unlock(shost->host_lock);
         spin_lock(sdev->request_queue->queue_lock);
 
@@ -1550,14 +1623,13 @@ static void scsi_request_fn(struct request_queue *q)
                         goto not_ready;
                 }
 
+                if (!scsi_target_queue_ready(shost, sdev))
+                        goto not_ready;
+
                 if (!scsi_host_queue_ready(q, shost, sdev))
                         goto not_ready;
-                if (scsi_target(sdev)->single_lun) {
-                        if (scsi_target(sdev)->starget_sdev_user &&
-                            scsi_target(sdev)->starget_sdev_user != sdev)
-                                goto not_ready;
-                        scsi_target(sdev)->starget_sdev_user = sdev;
-                }
+
+                scsi_target(sdev)->target_busy++;
                 shost->host_busy++;
 
                 /*
diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index 334862e26a1b..b14dc02c3ded 100644
--- a/drivers/scsi/scsi_scan.c
+++ b/drivers/scsi/scsi_scan.c
@@ -419,6 +419,7 @@ static struct scsi_target *scsi_alloc_target(struct device *parent,
         dev->type = &scsi_target_type;
         starget->id = id;
         starget->channel = channel;
+        starget->can_queue = 0;
         INIT_LIST_HEAD(&starget->siblings);
         INIT_LIST_HEAD(&starget->devices);
         starget->state = STARGET_CREATED;
diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h
index 192f8716aa9e..3a5662b2817e 100644
--- a/include/scsi/scsi.h
+++ b/include/scsi/scsi.h
@@ -426,6 +426,7 @@ static inline int scsi_is_wlun(unsigned int lun)
 #define SCSI_MLQUEUE_HOST_BUSY   0x1055
 #define SCSI_MLQUEUE_DEVICE_BUSY 0x1056
 #define SCSI_MLQUEUE_EH_RETRY    0x1057
+#define SCSI_MLQUEUE_TARGET_BUSY 0x1058
 
 /*
  *  Use these to separate status msg and our bytes
diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h
index b49e725be039..a37a8148a310 100644
--- a/include/scsi/scsi_device.h
+++ b/include/scsi/scsi_device.h
@@ -238,6 +238,16 @@ struct scsi_target {
                                                  * for the device at a time. */
         unsigned int            pdt_1f_for_no_lun;      /* PDT = 0x1f */
                                                 /* means no lun present */
+        /* commands actually active on LLD. protected by host lock. */
+        unsigned int            target_busy;
+        /*
+         * LLDs should set this in the slave_alloc host template callout.
+         * If set to zero then there is not limit.
+         */
+        unsigned int            can_queue;
+        unsigned int            target_blocked;
+        unsigned int            max_target_blocked;
+#define SCSI_DEFAULT_TARGET_BLOCKED     3
 
         char                    scsi_level;
         struct execute_work     ew;