From f0c0a376d0fcd4c5579ecf5e95f88387cba85211 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Sun, 17 Aug 2008 15:24:38 -0500 Subject: [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 Signed-off-by: James Bottomley --- drivers/scsi/scsi.c | 10 ++++- drivers/scsi/scsi_lib.c | 104 +++++++++++++++++++++++++++++++++++++++-------- drivers/scsi/scsi_scan.c | 1 + 3 files changed, 97 insertions(+), 18 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/scsi.c b/drivers/scsi/scsi.c index 2ac3cb2b908..f8b79d401d5 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 98ee55ced59..91c74c55aa5 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 334862e26a1..b14dc02c3de 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; -- cgit v1.2.3-18-g5258 From c5e98e912c5423a0ec2eed7aa1064578d44f8a8e Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Sun, 17 Aug 2008 15:24:39 -0500 Subject: [SCSI] qla4xxx: return SCSI_MLQUEUE_TARGET_BUSY when driver has detected session error When qla4xxx begins recovery and the iscsi class is firing up to handle it, we need to retrn SCSI_MLQUEUE_TARGET_BUSY from the driver instead of host busy, because the session recovery only affects the one target. Signed-off-by: Mike Christie Acked-by: David C Somayajulu Signed-off-by: James Bottomley --- drivers/scsi/qla4xxx/ql4_os.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index de8279ad7d8..4255b36ff96 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -439,7 +439,7 @@ static int qla4xxx_queuecommand(struct scsi_cmnd *cmd, cmd->result = DID_NO_CONNECT << 16; goto qc_fail_command; } - goto qc_host_busy; + return SCSI_MLQUEUE_TARGET_BUSY; } if (test_bit(DPC_RESET_HA_INTR, &ha->dpc_flags)) -- cgit v1.2.3-18-g5258 From 7b594131c4f38edeb13d8c6c0147949173c47013 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Sun, 17 Aug 2008 15:24:40 -0500 Subject: [SCSI] qla2xxx: return SCSI_MLQUEUE_TARGET_BUSY when driver has detected rport error or race If the fcport is not online then we do not want to block IO to all ports on the host. We just want to stop IO on port not online, so we should be using the SCSI_MLQUEUE_TARGET_BUSY return value. For the case where we race with the rport memset initialization we do not want the queuecommand to be called again so we can just use SCSI_MLQUEUE_TARGET_BUSY for this. Signed-off-by: Mike Christie Acked-by: Andrew Vasquez Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_os.c | 26 ++++++++++++-------------- 1 file changed, 12 insertions(+), 14 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index 3433441b956..2aed4721c0d 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -394,10 +394,8 @@ qla2x00_queuecommand(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *)) } /* Close window on fcport/rport state-transitioning. */ - if (fcport->drport) { - cmd->result = DID_IMM_RETRY << 16; - goto qc_fail_command; - } + if (fcport->drport) + goto qc_target_busy; if (atomic_read(&fcport->state) != FCS_ONLINE) { if (atomic_read(&fcport->state) == FCS_DEVICE_DEAD || @@ -405,7 +403,7 @@ qla2x00_queuecommand(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *)) cmd->result = DID_NO_CONNECT << 16; goto qc_fail_command; } - goto qc_host_busy; + goto qc_target_busy; } spin_unlock_irq(ha->host->host_lock); @@ -428,10 +426,11 @@ qc_host_busy_free_sp: qc_host_busy_lock: spin_lock_irq(ha->host->host_lock); - -qc_host_busy: return SCSI_MLQUEUE_HOST_BUSY; +qc_target_busy: + return SCSI_MLQUEUE_TARGET_BUSY; + qc_fail_command: done(cmd); @@ -461,10 +460,8 @@ qla24xx_queuecommand(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *)) } /* Close window on fcport/rport state-transitioning. */ - if (fcport->drport) { - cmd->result = DID_IMM_RETRY << 16; - goto qc24_fail_command; - } + if (fcport->drport) + goto qc24_target_busy; if (atomic_read(&fcport->state) != FCS_ONLINE) { if (atomic_read(&fcport->state) == FCS_DEVICE_DEAD || @@ -472,7 +469,7 @@ qla24xx_queuecommand(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *)) cmd->result = DID_NO_CONNECT << 16; goto qc24_fail_command; } - goto qc24_host_busy; + goto qc24_target_busy; } spin_unlock_irq(ha->host->host_lock); @@ -495,10 +492,11 @@ qc24_host_busy_free_sp: qc24_host_busy_lock: spin_lock_irq(ha->host->host_lock); - -qc24_host_busy: return SCSI_MLQUEUE_HOST_BUSY; +qc24_target_busy: + return SCSI_MLQUEUE_TARGET_BUSY; + qc24_fail_command: done(cmd); -- cgit v1.2.3-18-g5258 From d6d13ee19da6d291c99f980dcb76f6b7dc676804 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Sun, 17 Aug 2008 15:24:43 -0500 Subject: [SCSI] libiscsi: Use SCSI_MLQUEUE_TARGET_BUSY For the conditions below we do not want the queuecommand function to call us right back, so return SCSI_MLQUEUE_TARGET_BUSY. Signed-off-by: Mike Christie Signed-off-by: James Bottomley --- drivers/scsi/libiscsi.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index da7b67d30d9..521dbf7e521 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -1194,12 +1194,10 @@ int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *)) switch (session->state) { case ISCSI_STATE_IN_RECOVERY: reason = FAILURE_SESSION_IN_RECOVERY; - sc->result = DID_IMM_RETRY << 16; - break; + goto reject; case ISCSI_STATE_LOGGING_OUT: reason = FAILURE_SESSION_LOGGING_OUT; - sc->result = DID_IMM_RETRY << 16; - break; + goto reject; case ISCSI_STATE_RECOVERY_FAILED: reason = FAILURE_SESSION_RECOVERY_TIMEOUT; sc->result = DID_NO_CONNECT << 16; @@ -1267,7 +1265,7 @@ reject: spin_unlock(&session->lock); debug_scsi("cmd 0x%x rejected (%d)\n", sc->cmnd[0], reason); spin_lock(host->host_lock); - return SCSI_MLQUEUE_HOST_BUSY; + return SCSI_MLQUEUE_TARGET_BUSY; fault: spin_unlock(&session->lock); -- cgit v1.2.3-18-g5258 From a93ce0244f2e94dd48e0b4a2742a4e3bf196ab53 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Sun, 17 Aug 2008 15:24:41 -0500 Subject: [SCSI] lpfc: use SCSI_MLQUEUE_TARGET_BUSY when catching the rport transition race We do want to call right back into the queuecommand during the race, so we can just use SCSI_MLQUEUE_TARGET_BUSY. Signed-off-by: Mike Christie Acked-by: James Smart Signed-off-by: James Bottomley --- drivers/scsi/lpfc/lpfc_scsi.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c index 1bcebbd3dfa..a22bdf9548a 100644 --- a/drivers/scsi/lpfc/lpfc_scsi.c +++ b/drivers/scsi/lpfc/lpfc_scsi.c @@ -966,10 +966,9 @@ lpfc_queuecommand(struct scsi_cmnd *cmnd, void (*done) (struct scsi_cmnd *)) * Catch race where our node has transitioned, but the * transport is still transitioning. */ - if (!ndlp || !NLP_CHK_NODE_ACT(ndlp)) { - cmnd->result = ScsiResult(DID_BUS_BUSY, 0); - goto out_fail_command; - } + if (!ndlp || !NLP_CHK_NODE_ACT(ndlp)) + goto out_target_busy; + lpfc_cmd = lpfc_get_scsi_buf(phba); if (lpfc_cmd == NULL) { lpfc_adjust_queue_depth(phba); @@ -1014,6 +1013,8 @@ lpfc_queuecommand(struct scsi_cmnd *cmnd, void (*done) (struct scsi_cmnd *)) lpfc_release_scsi_buf(phba, lpfc_cmd); out_host_busy: return SCSI_MLQUEUE_HOST_BUSY; + out_target_busy: + return SCSI_MLQUEUE_TARGET_BUSY; out_fail_command: done(cmnd); -- cgit v1.2.3-18-g5258 From fff9d40ce0eb4b46f3e186823ceab6bc02c3e5d3 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Tue, 19 Aug 2008 18:45:23 -0500 Subject: [SCSI] fc class: unblock target after calling terminate callback (take 2) When we block a rport and the driver implements the terminate callback we will fail IO that was running quickly. However IO that was in the scsi_device/block queue sits there until the dev_loss_tmo fires, and this can make it look like IO is lost because new IO will get executed but that IO stuck in the blocked queue sits there for some time longer. With this patch when the fast io fail tmo fires, we will fail the blocked IO and any new IO. This patch also allows all drivers to partially support the fast io fail tmo. If the terminate io callback is not implemented, we will still fail blocked IO and any new IO, so multipath can handle that. This patch also allows the fc and iscsi classes to implement the same behavior. The timers are just unfornately named differently. This patch also fixes the problem where drivers were unblocking the target in their terminate callback, which was needed for rport removal, but for fast io fail timeout it would cause IO to bounce arround the scsi/block layer and the LLD queuecommand. And it for drivers that could have IO stuck but did not have a terminate callback the unblock calls in the class will fix them. v2. - fix up bit setting style to meet JamesS's pref. - Broke out new host byte error changes to make it easier to read. - added JamesS's ack from list. v1 - initial patch Signed-off-by: Mike Christie Acked-by: James Smart Signed-off-by: James Bottomley --- drivers/scsi/scsi_transport_fc.c | 47 ++++++++++++++++++++++++---------------- 1 file changed, 28 insertions(+), 19 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c index d5f7653bb94..1e71abf0607 100644 --- a/drivers/scsi/scsi_transport_fc.c +++ b/drivers/scsi/scsi_transport_fc.c @@ -2133,8 +2133,7 @@ fc_attach_transport(struct fc_function_template *ft) SETUP_PRIVATE_RPORT_ATTRIBUTE_RD(roles); SETUP_PRIVATE_RPORT_ATTRIBUTE_RD(port_state); SETUP_PRIVATE_RPORT_ATTRIBUTE_RD(scsi_target_id); - if (ft->terminate_rport_io) - SETUP_PRIVATE_RPORT_ATTRIBUTE_RW(fast_io_fail_tmo); + SETUP_PRIVATE_RPORT_ATTRIBUTE_RW(fast_io_fail_tmo); BUG_ON(count > FC_RPORT_NUM_ATTRS); @@ -2328,6 +2327,22 @@ fc_remove_host(struct Scsi_Host *shost) } EXPORT_SYMBOL(fc_remove_host); +static void fc_terminate_rport_io(struct fc_rport *rport) +{ + struct Scsi_Host *shost = rport_to_shost(rport); + struct fc_internal *i = to_fc_internal(shost->transportt); + + /* Involve the LLDD if possible to terminate all io on the rport. */ + if (i->f->terminate_rport_io) + i->f->terminate_rport_io(rport); + + /* + * must unblock to flush queued IO. The caller will have set + * the port_state or flags, so that fc_remote_port_chkready will + * fail IO. + */ + scsi_target_unblock(&rport->dev); +} /** * fc_starget_delete - called to delete the scsi decendents of an rport @@ -2340,13 +2355,8 @@ fc_starget_delete(struct work_struct *work) { struct fc_rport *rport = container_of(work, struct fc_rport, stgt_delete_work); - struct Scsi_Host *shost = rport_to_shost(rport); - struct fc_internal *i = to_fc_internal(shost->transportt); - - /* Involve the LLDD if possible to terminate all io on the rport. */ - if (i->f->terminate_rport_io) - i->f->terminate_rport_io(rport); + fc_terminate_rport_io(rport); scsi_remove_target(&rport->dev); } @@ -2372,10 +2382,7 @@ fc_rport_final_delete(struct work_struct *work) if (rport->flags & FC_RPORT_SCAN_PENDING) scsi_flush_work(shost); - /* involve the LLDD to terminate all pending i/o */ - if (i->f->terminate_rport_io) - i->f->terminate_rport_io(rport); - + fc_terminate_rport_io(rport); /* * Cancel any outstanding timers. These should really exist * only when rmmod'ing the LLDD and we're asking for @@ -2639,7 +2646,8 @@ fc_remote_port_add(struct Scsi_Host *shost, int channel, spin_lock_irqsave(shost->host_lock, flags); - rport->flags &= ~FC_RPORT_DEVLOSS_PENDING; + rport->flags &= ~(FC_RPORT_FAST_FAIL_TIMEDOUT | + FC_RPORT_DEVLOSS_PENDING); /* if target, initiate a scan */ if (rport->scsi_target_id != -1) { @@ -2702,6 +2710,7 @@ fc_remote_port_add(struct Scsi_Host *shost, int channel, rport->port_id = ids->port_id; rport->roles = ids->roles; rport->port_state = FC_PORTSTATE_ONLINE; + rport->flags &= ~FC_RPORT_FAST_FAIL_TIMEDOUT; if (fci->f->dd_fcrport_size) memset(rport->dd_data, 0, @@ -2784,7 +2793,6 @@ void fc_remote_port_delete(struct fc_rport *rport) { struct Scsi_Host *shost = rport_to_shost(rport); - struct fc_internal *i = to_fc_internal(shost->transportt); int timeout = rport->dev_loss_tmo; unsigned long flags; @@ -2830,7 +2838,7 @@ fc_remote_port_delete(struct fc_rport *rport) /* see if we need to kill io faster than waiting for device loss */ if ((rport->fast_io_fail_tmo != -1) && - (rport->fast_io_fail_tmo < timeout) && (i->f->terminate_rport_io)) + (rport->fast_io_fail_tmo < timeout)) fc_queue_devloss_work(shost, &rport->fail_io_work, rport->fast_io_fail_tmo * HZ); @@ -2906,7 +2914,8 @@ fc_remote_port_rolechg(struct fc_rport *rport, u32 roles) fc_flush_devloss(shost); spin_lock_irqsave(shost->host_lock, flags); - rport->flags &= ~FC_RPORT_DEVLOSS_PENDING; + rport->flags &= ~(FC_RPORT_FAST_FAIL_TIMEDOUT | + FC_RPORT_DEVLOSS_PENDING); spin_unlock_irqrestore(shost->host_lock, flags); /* ensure any stgt delete functions are done */ @@ -3001,6 +3010,7 @@ fc_timeout_deleted_rport(struct work_struct *work) rport->supported_classes = FC_COS_UNSPECIFIED; rport->roles = FC_PORT_ROLE_UNKNOWN; rport->port_state = FC_PORTSTATE_NOTPRESENT; + rport->flags &= ~FC_RPORT_FAST_FAIL_TIMEDOUT; /* remove the identifiers that aren't used in the consisting binding */ switch (fc_host->tgtid_bind_type) { @@ -3043,13 +3053,12 @@ fc_timeout_fail_rport_io(struct work_struct *work) { struct fc_rport *rport = container_of(work, struct fc_rport, fail_io_work.work); - struct Scsi_Host *shost = rport_to_shost(rport); - struct fc_internal *i = to_fc_internal(shost->transportt); if (rport->port_state != FC_PORTSTATE_BLOCKED) return; - i->f->terminate_rport_io(rport); + rport->flags |= FC_RPORT_FAST_FAIL_TIMEDOUT; + fc_terminate_rport_io(rport); } /** -- cgit v1.2.3-18-g5258 From 9cc328f502eacfcc52ab1c1bf9a7729cf12f14be Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Tue, 19 Aug 2008 18:45:24 -0500 Subject: [SCSI] ibmvfc, qla2xxx, lpfc: remove scsi_target_unblock calls in terminate callbacks The fc class now calls scsi_target_unblock after calling the terminate callback, so this patch removes the calls from the drivers. Signed-off-by: Mike Christie Signed-off-by: James Bottomley --- drivers/scsi/ibmvscsi/ibmvfc.c | 2 -- drivers/scsi/lpfc/lpfc_hbadisc.c | 8 -------- drivers/scsi/qla2xxx/qla_attr.c | 1 - 3 files changed, 11 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/ibmvscsi/ibmvfc.c b/drivers/scsi/ibmvscsi/ibmvfc.c index 4e0b7c8eb32..7650707a40d 100644 --- a/drivers/scsi/ibmvscsi/ibmvfc.c +++ b/drivers/scsi/ibmvscsi/ibmvfc.c @@ -2031,8 +2031,6 @@ static void ibmvfc_terminate_rport_io(struct fc_rport *rport) spin_unlock_irqrestore(shost->host_lock, flags); } else ibmvfc_issue_fc_host_lip(shost); - - scsi_target_unblock(&rport->dev); LEAVE; } diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c index a98d11bf357..aaf398e5c93 100644 --- a/drivers/scsi/lpfc/lpfc_hbadisc.c +++ b/drivers/scsi/lpfc/lpfc_hbadisc.c @@ -88,14 +88,6 @@ lpfc_terminate_rport_io(struct fc_rport *rport) &phba->sli.ring[phba->sli.fcp_ring], ndlp->nlp_sid, 0, LPFC_CTX_TGT); } - - /* - * A device is normally blocked for rediscovery and unblocked when - * devloss timeout happens. In case a vport is removed or driver - * unloaded before devloss timeout happens, we need to unblock here. - */ - scsi_target_unblock(&rport->dev); - return; } /* diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c index 0ddfe7106b3..ed731968f15 100644 --- a/drivers/scsi/qla2xxx/qla_attr.c +++ b/drivers/scsi/qla2xxx/qla_attr.c @@ -1006,7 +1006,6 @@ qla2x00_terminate_rport_io(struct fc_rport *rport) } qla2x00_abort_fcport_cmds(fcport); - scsi_target_unblock(&rport->dev); } static int -- cgit v1.2.3-18-g5258 From a4dfaa6f2e55b736adf2719133996f7e7dc309bc Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Tue, 19 Aug 2008 18:45:25 -0500 Subject: [SCSI] scsi: add transport host byte errors (v3) Currently, if there is a transport problem the iscsi drivers will return outstanding commands (commands being exeucted by the driver/fw/hw) with DID_BUS_BUSY and block the session so no new commands can be queued. Commands that are caught between the failure handling and blocking are failed with DID_IMM_RETRY or one of the scsi ml queuecommand return values. When the recovery_timeout fires, the iscsi drivers then fail IO with DID_NO_CONNECT. For fcp, some drivers will fail some outstanding IO (disk but possibly not tape) with DID_BUS_BUSY or DID_ERROR or some other value that causes a retry and hits the scsi_error.c failfast check, block the rport, and commands caught in the race are failed with DID_IMM_RETRY. Other drivers, may hold onto all IO and wait for the terminate_rport_io or dev_loss_tmo_callbk to be called. The following patches attempt to unify what upper layers will see drivers like multipath can make a good guess. This relies on drivers being hooked into their transport class. This first patch just defines two new host byte errors so drivers can return the same value for when a rport/session is blocked and for when the fast_io_fail_tmo fires. The idea is that if the LLD/class detects a problem and is going to block a rport/session, then if the LLD wants or must return the command to scsi-ml, then it can return it with DID_TRANSPORT_DISRUPTED. This will requeue the IO into the same scsi queue it came from, until the fast io fail timer fires and the class decides what to do. When using multipath and the fast_io_fail_tmo fires then the class can fail commands with DID_TRANSPORT_FAILFAST or drivers can use DID_TRANSPORT_FAILFAST in their terminate_rport_io callbacks or the equivlent in iscsi if we ever implement more advanced recovery methods. A LLD, like lpfc, could continue to return DID_ERROR and then it will hit the normal failfast path, so drivers do not have fully be ported to work better. The point of the patches is that upper layers will not see a failure that could be recovered from while the rport/session is blocked until fast_io_fail_tmo/recovery_timeout fires. V3 Remove some comments. V2 Fixed patch/diff errors and renamed DID_TRANSPORT_BLOCKED to DID_TRANSPORT_DISRUPTED. V1 initial patch. Signed-off-by: Mike Christie Signed-off-by: James Bottomley --- drivers/scsi/constants.c | 3 ++- drivers/scsi/scsi_error.c | 15 ++++++++++++++- 2 files changed, 16 insertions(+), 2 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/constants.c b/drivers/scsi/constants.c index 9785d738419..4003deefb7d 100644 --- a/drivers/scsi/constants.c +++ b/drivers/scsi/constants.c @@ -1364,7 +1364,8 @@ EXPORT_SYMBOL(scsi_print_sense); static const char * const hostbyte_table[]={ "DID_OK", "DID_NO_CONNECT", "DID_BUS_BUSY", "DID_TIME_OUT", "DID_BAD_TARGET", "DID_ABORT", "DID_PARITY", "DID_ERROR", "DID_RESET", "DID_BAD_INTR", -"DID_PASSTHROUGH", "DID_SOFT_ERROR", "DID_IMM_RETRY", "DID_REQUEUE"}; +"DID_PASSTHROUGH", "DID_SOFT_ERROR", "DID_IMM_RETRY", "DID_REQUEUE", +"DID_TRANSPORT_DISRUPTED", "DID_TRANSPORT_FAILFAST" }; #define NUM_HOSTBYTE_STRS ARRAY_SIZE(hostbyte_table) static const char * const driverbyte_table[]={ diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index fecefa05cb6..5bf8be21a16 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -1290,7 +1290,20 @@ int scsi_decide_disposition(struct scsi_cmnd *scmd) case DID_REQUEUE: return ADD_TO_MLQUEUE; - + case DID_TRANSPORT_DISRUPTED: + /* + * LLD/transport was disrupted during processing of the IO. + * The transport class is now blocked/blocking, + * and the transport will decide what to do with the IO + * based on its timers and recovery capablilities. + */ + return ADD_TO_MLQUEUE; + case DID_TRANSPORT_FAILFAST: + /* + * The transport decided to failfast the IO (most likely + * the fast io fail tmo fired), so send IO directly upwards. + */ + return SUCCESS; case DID_ERROR: if (msg_byte(scmd->result) == COMMAND_COMPLETE && status_byte(scmd->result) == RESERVATION_CONFLICT) -- cgit v1.2.3-18-g5258 From 56d7fcfa815564b40a1b0ec7a30ea8cb3bc0713e Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Tue, 19 Aug 2008 18:45:26 -0500 Subject: [SCSI] iscsi class, libiscsi and qla4xxx: convert to new transport host byte values This patch converts the iscsi drivers to the new host byte values. v2 Drop some conversions. Want to avoid conflicts with other patches. v1 initial patch. Signed-off-by: Mike Christie Signed-off-by: James Bottomley --- drivers/scsi/libiscsi.c | 8 +++++--- drivers/scsi/qla4xxx/ql4_isr.c | 4 ++-- drivers/scsi/scsi_transport_iscsi.c | 4 ++-- 3 files changed, 9 insertions(+), 7 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index 521dbf7e521..0e8f26baca6 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -1200,7 +1200,7 @@ int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *)) goto reject; case ISCSI_STATE_RECOVERY_FAILED: reason = FAILURE_SESSION_RECOVERY_TIMEOUT; - sc->result = DID_NO_CONNECT << 16; + sc->result = DID_TRANSPORT_FAILFAST << 16; break; case ISCSI_STATE_TERMINATE: reason = FAILURE_SESSION_TERMINATE; @@ -2333,8 +2333,10 @@ static void iscsi_start_session_recovery(struct iscsi_session *session, * flush queues. */ spin_lock_bh(&session->lock); - fail_all_commands(conn, -1, - STOP_CONN_RECOVER ? DID_BUS_BUSY : DID_ERROR); + if (STOP_CONN_RECOVER) + fail_all_commands(conn, -1, DID_TRANSPORT_DISRUPTED); + else + fail_all_commands(conn, -1, DID_ERROR); flush_control_queues(session, conn); spin_unlock_bh(&session->lock); mutex_unlock(&session->eh_mutex); diff --git a/drivers/scsi/qla4xxx/ql4_isr.c b/drivers/scsi/qla4xxx/ql4_isr.c index a91a57c57bf..799120fcb9b 100644 --- a/drivers/scsi/qla4xxx/ql4_isr.c +++ b/drivers/scsi/qla4xxx/ql4_isr.c @@ -139,7 +139,7 @@ static void qla4xxx_status_entry(struct scsi_qla_host *ha, ha->host_no, cmd->device->channel, cmd->device->id, cmd->device->lun)); - cmd->result = DID_BUS_BUSY << 16; + cmd->result = DID_TRANSPORT_DISRUPTED << 16; /* * Mark device missing so that we won't continue to send @@ -243,7 +243,7 @@ static void qla4xxx_status_entry(struct scsi_qla_host *ha, if (atomic_read(&ddb_entry->state) == DDB_STATE_ONLINE) qla4xxx_mark_device_missing(ha, ddb_entry); - cmd->result = DID_BUS_BUSY << 16; + cmd->result = DID_TRANSPORT_DISRUPTED << 16; break; case SCS_QUEUE_FULL: diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 0ce5f7cdfe2..cbaae48f47e 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -374,10 +374,10 @@ int iscsi_session_chkready(struct iscsi_cls_session *session) err = 0; break; case ISCSI_SESSION_FAILED: - err = DID_IMM_RETRY << 16; + err = DID_TRANSPORT_DISRUPTED << 16; break; case ISCSI_SESSION_FREE: - err = DID_NO_CONNECT << 16; + err = DID_TRANSPORT_FAILFAST << 16; break; default: err = DID_NO_CONNECT << 16; -- cgit v1.2.3-18-g5258 From 056a44834950ffa51fafa6c76a720fa32e86851a Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Tue, 19 Aug 2008 18:45:29 -0500 Subject: [SCSI] qla2xxx: use new host byte transport errors. This has qla2xxx use the new transport error values instead of DID_BUS_BUSY. I am not sure if all the errors in qla_isr.c I changed are transport related. We end up blocking/deleting the rport for all of them so it is better to use the new transport error since the fc classs will decide when to fail the IO. With this patch if I pull a cable then IO that had reached the driver, will be failed with DID_TRANSPORT_DISRUPTED (not including tape). The fc class will then fail the IO when the fast io fail tmo has fired, and the driver will flush any other commands running. Signed-off-by: Mike Christie Cc: Andrew Vasquez Signed-off-by: James Bottomley --- drivers/scsi/qla2xxx/qla_isr.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index fc4bfa7f839..a76efd99d00 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -1187,7 +1187,12 @@ qla2x00_status_entry(scsi_qla_host_t *ha, void *pkt) cp->serial_number, comp_status, atomic_read(&fcport->state))); - cp->result = DID_BUS_BUSY << 16; + /* + * We are going to have the fc class block the rport + * while we try to recover so instruct the mid layer + * to requeue until the class decides how to handle this. + */ + cp->result = DID_TRANSPORT_DISRUPTED << 16; if (atomic_read(&fcport->state) == FCS_ONLINE) qla2x00_mark_device_lost(fcport->ha, fcport, 1, 1); break; @@ -1214,7 +1219,12 @@ qla2x00_status_entry(scsi_qla_host_t *ha, void *pkt) break; case CS_TIMEOUT: - cp->result = DID_BUS_BUSY << 16; + /* + * We are going to have the fc class block the rport + * while we try to recover so instruct the mid layer + * to requeue until the class decides how to handle this. + */ + cp->result = DID_TRANSPORT_DISRUPTED << 16; if (IS_FWI2_CAPABLE(ha)) { DEBUG2(printk(KERN_INFO -- cgit v1.2.3-18-g5258 From 6000a368cd8e6da1caf101411bdb494cd6fb8b09 Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Tue, 19 Aug 2008 18:45:30 -0500 Subject: [SCSI] block: separate failfast into multiple bits. Multipath is best at handling transport errors. If it gets a device error then there is not much the multipath layer can do. It will just access the same device but from a different path. This patch breaks up failfast into device, transport and driver errors. The multipath layers (md and dm mutlipath) only ask the lower levels to fast fail transport errors. The user of failfast, read ahead, will ask to fast fail on all errors. Note that blk_noretry_request will return true if any failfast bit is set. This allows drivers that do not support the multipath failfast bits to continue to fail on any failfast error like before. Drivers like scsi that are able to fail fast specific errors can check for the specific fail fast type. In the next patch I will convert scsi. Signed-off-by: Mike Christie Cc: Jens Axboe Signed-off-by: James Bottomley --- drivers/scsi/device_handler/scsi_dh_alua.c | 3 ++- drivers/scsi/device_handler/scsi_dh_emc.c | 3 ++- drivers/scsi/device_handler/scsi_dh_hp_sw.c | 6 ++++-- drivers/scsi/device_handler/scsi_dh_rdac.c | 3 ++- drivers/scsi/scsi_transport_spi.c | 4 +++- 5 files changed, 13 insertions(+), 6 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/device_handler/scsi_dh_alua.c b/drivers/scsi/device_handler/scsi_dh_alua.c index 708e475896b..cb8aa3b58c2 100644 --- a/drivers/scsi/device_handler/scsi_dh_alua.c +++ b/drivers/scsi/device_handler/scsi_dh_alua.c @@ -109,7 +109,8 @@ static struct request *get_alua_req(struct scsi_device *sdev, } rq->cmd_type = REQ_TYPE_BLOCK_PC; - rq->cmd_flags |= REQ_FAILFAST | REQ_NOMERGE; + rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | + REQ_FAILFAST_DRIVER | REQ_NOMERGE; rq->retries = ALUA_FAILOVER_RETRIES; rq->timeout = ALUA_FAILOVER_TIMEOUT; diff --git a/drivers/scsi/device_handler/scsi_dh_emc.c b/drivers/scsi/device_handler/scsi_dh_emc.c index 8f45570a8a0..0e572d2c5b0 100644 --- a/drivers/scsi/device_handler/scsi_dh_emc.c +++ b/drivers/scsi/device_handler/scsi_dh_emc.c @@ -303,7 +303,8 @@ static struct request *get_req(struct scsi_device *sdev, int cmd, rq->cmd[4] = len; rq->cmd_type = REQ_TYPE_BLOCK_PC; - rq->cmd_flags |= REQ_FAILFAST; + rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | + REQ_FAILFAST_DRIVER; rq->timeout = CLARIION_TIMEOUT; rq->retries = CLARIION_RETRIES; diff --git a/drivers/scsi/device_handler/scsi_dh_hp_sw.c b/drivers/scsi/device_handler/scsi_dh_hp_sw.c index 5e93c88ad66..9aec4ca64e5 100644 --- a/drivers/scsi/device_handler/scsi_dh_hp_sw.c +++ b/drivers/scsi/device_handler/scsi_dh_hp_sw.c @@ -112,7 +112,8 @@ static int hp_sw_tur(struct scsi_device *sdev, struct hp_sw_dh_data *h) return SCSI_DH_RES_TEMP_UNAVAIL; req->cmd_type = REQ_TYPE_BLOCK_PC; - req->cmd_flags |= REQ_FAILFAST; + req->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | + REQ_FAILFAST_DRIVER; req->cmd_len = COMMAND_SIZE(TEST_UNIT_READY); req->cmd[0] = TEST_UNIT_READY; req->timeout = HP_SW_TIMEOUT; @@ -204,7 +205,8 @@ static int hp_sw_start_stop(struct scsi_device *sdev, struct hp_sw_dh_data *h) return SCSI_DH_RES_TEMP_UNAVAIL; req->cmd_type = REQ_TYPE_BLOCK_PC; - req->cmd_flags |= REQ_FAILFAST; + req->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | + REQ_FAILFAST_DRIVER; req->cmd_len = COMMAND_SIZE(START_STOP); req->cmd[0] = START_STOP; req->cmd[4] = 1; /* Start spin cycle */ diff --git a/drivers/scsi/device_handler/scsi_dh_rdac.c b/drivers/scsi/device_handler/scsi_dh_rdac.c index 50bf95f3b5c..a43c3ed4df2 100644 --- a/drivers/scsi/device_handler/scsi_dh_rdac.c +++ b/drivers/scsi/device_handler/scsi_dh_rdac.c @@ -226,7 +226,8 @@ static struct request *get_rdac_req(struct scsi_device *sdev, } rq->cmd_type = REQ_TYPE_BLOCK_PC; - rq->cmd_flags |= REQ_FAILFAST | REQ_NOMERGE; + rq->cmd_flags |= REQ_FAILFAST_DEV | REQ_FAILFAST_TRANSPORT | + REQ_FAILFAST_DRIVER; rq->retries = RDAC_RETRIES; rq->timeout = RDAC_TIMEOUT; diff --git a/drivers/scsi/scsi_transport_spi.c b/drivers/scsi/scsi_transport_spi.c index b29360ed0bd..7c2d28924d2 100644 --- a/drivers/scsi/scsi_transport_spi.c +++ b/drivers/scsi/scsi_transport_spi.c @@ -109,7 +109,9 @@ static int spi_execute(struct scsi_device *sdev, const void *cmd, for(i = 0; i < DV_RETRIES; i++) { result = scsi_execute(sdev, cmd, dir, buffer, bufflen, sense, DV_TIMEOUT, /* retries */ 1, - REQ_FAILFAST); + REQ_FAILFAST_DEV | + REQ_FAILFAST_TRANSPORT | + REQ_FAILFAST_DRIVER); if (result & DRIVER_SENSE) { struct scsi_sense_hdr sshdr_tmp; if (!sshdr) -- cgit v1.2.3-18-g5258 From 4a27446f3e39b06c28d1c8e31d33a5340826ed5c Mon Sep 17 00:00:00 2001 From: Mike Christie Date: Tue, 19 Aug 2008 18:45:31 -0500 Subject: [SCSI] modify scsi to handle new fail fast flags. This checks the errors the scsi-ml determined were retryable and returns if we should fast fail it based on the request fail fast flags. Without the patch, drivers like lpfc, qla2xxx and fcoe would return DID_ERROR for what it determines is a temporary communication problem. There is no loss of connectivity at that time and the driver thinks that it would be fast to retry at the driver level. SCSI-ml will however sees fast fail on the request and DID_ERROR and will fast fail the io. This will then cause dm-multipath to fail the path and possibley switch target controllers when we should be retrying at the scsi layer. We also were fast failing device errors to dm multiapth when unless the scsi_dh modules think otherwis we want to retry at the scsi layer because multipath can only retry the IO like scsi should have done. multipath is a little dumber though because it does not what the error was for and assumes that it should fail the paths. Signed-off-by: Mike Christie Signed-off-by: James Bottomley --- drivers/scsi/scsi_error.c | 38 ++++++++++++++++++++++++++++++++++++-- drivers/scsi/scsi_lib.c | 2 +- drivers/scsi/scsi_priv.h | 1 + 3 files changed, 38 insertions(+), 3 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 5bf8be21a16..ad019ece213 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -1218,6 +1218,40 @@ static void scsi_eh_offline_sdevs(struct list_head *work_q, return; } +/** + * scsi_noretry_cmd - determinte if command should be failed fast + * @scmd: SCSI cmd to examine. + */ +int scsi_noretry_cmd(struct scsi_cmnd *scmd) +{ + switch (host_byte(scmd->result)) { + case DID_OK: + break; + case DID_BUS_BUSY: + return blk_failfast_transport(scmd->request); + case DID_PARITY: + return blk_failfast_dev(scmd->request); + case DID_ERROR: + if (msg_byte(scmd->result) == COMMAND_COMPLETE && + status_byte(scmd->result) == RESERVATION_CONFLICT) + return 0; + /* fall through */ + case DID_SOFT_ERROR: + return blk_failfast_driver(scmd->request); + } + + switch (status_byte(scmd->result)) { + case CHECK_CONDITION: + /* + * assume caller has checked sense and determinted + * the check condition was retryable. + */ + return blk_failfast_dev(scmd->request); + } + + return 0; +} + /** * scsi_decide_disposition - Disposition a cmd on return from LLD. * @scmd: SCSI cmd to examine. @@ -1396,7 +1430,7 @@ int scsi_decide_disposition(struct scsi_cmnd *scmd) * even if the request is marked fast fail, we still requeue * for queue congestion conditions (QUEUE_FULL or BUSY) */ if ((++scmd->retries) <= scmd->allowed - && !blk_noretry_request(scmd->request)) { + && !scsi_noretry_cmd(scmd)) { return NEEDS_RETRY; } else { /* @@ -1521,7 +1555,7 @@ void scsi_eh_flush_done_q(struct list_head *done_q) list_for_each_entry_safe(scmd, next, done_q, eh_entry) { list_del_init(&scmd->eh_entry); if (scsi_device_online(scmd->device) && - !blk_noretry_request(scmd->request) && + !scsi_noretry_cmd(scmd) && (++scmd->retries <= scmd->allowed)) { SCSI_LOG_ERROR_RECOVERY(3, printk("%s: flush" " retry cmd: %p\n", diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 91c74c55aa5..e5a9526d203 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -706,7 +706,7 @@ static struct scsi_cmnd *scsi_end_request(struct scsi_cmnd *cmd, int error, leftover = req->data_len; /* kill remainder if no retrys */ - if (error && blk_noretry_request(req)) + if (error && scsi_noretry_cmd(cmd)) blk_end_request(req, error, leftover); else { if (requeue) { diff --git a/drivers/scsi/scsi_priv.h b/drivers/scsi/scsi_priv.h index 6cddd5dd323..e1850904ff7 100644 --- a/drivers/scsi/scsi_priv.h +++ b/drivers/scsi/scsi_priv.h @@ -59,6 +59,7 @@ void scsi_eh_ready_devs(struct Scsi_Host *shost, struct list_head *done_q); int scsi_eh_get_sense(struct list_head *work_q, struct list_head *done_q); +int scsi_noretry_cmd(struct scsi_cmnd *scmd); /* scsi_lib.c */ extern int scsi_maybe_unblock_host(struct scsi_device *sdev); -- cgit v1.2.3-18-g5258 From e59058c44025d71c9b7f260076a932935d3bba95 Mon Sep 17 00:00:00 2001 From: James Smart Date: Sun, 24 Aug 2008 21:49:00 -0400 Subject: [SCSI] lpfc 8.2.8 : Add kernel-doc function headers Signed-off-by: James Smart Signed-off-by: James Bottomley --- drivers/scsi/lpfc/lpfc_attr.c | 919 +++++++++++++++++++++++++- drivers/scsi/lpfc/lpfc_debugfs.c | 327 ++++++++- drivers/scsi/lpfc/lpfc_els.c | 1345 +++++++++++++++++++++++++++++++++++++- drivers/scsi/lpfc/lpfc_init.c | 575 ++++++++++++---- drivers/scsi/lpfc/lpfc_mbox.c | 468 +++++++++++-- drivers/scsi/lpfc/lpfc_mem.c | 115 +++- drivers/scsi/lpfc/lpfc_sli.c | 1063 +++++++++++++++++++++++++++++- 7 files changed, 4556 insertions(+), 256 deletions(-) (limited to 'drivers/scsi') diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c index 37bfa0bd1da..2926a2a7ee7 100644 --- a/drivers/scsi/lpfc/lpfc_attr.c +++ b/drivers/scsi/lpfc/lpfc_attr.c @@ -49,6 +49,21 @@ #define LPFC_LINK_SPEED_BITMAP 0x00000117 #define LPFC_LINK_SPEED_STRING "0, 1, 2, 4, 8" +/** + * lpfc_jedec_to_ascii: Hex to ascii convertor according to JEDEC rules. + * @incr: integer to convert. + * @hdw: ascii string holding converted integer plus a string terminator. + * + * Description: + * JEDEC Joint Electron Device Engineering Council. + * Convert a 32 bit integer composed of 8 nibbles into an 8 byte ascii + * character string. The string is then terminated with a NULL in byte 9. + * Hex 0-9 becomes ascii '0' to '9'. + * Hex a-f becomes ascii '=' to 'B' capital B. + * + * Notes: + * Coded for 32 bit integers only. + **/ static void lpfc_jedec_to_ascii(int incr, char hdw[]) { @@ -65,6 +80,14 @@ lpfc_jedec_to_ascii(int incr, char hdw[]) return; } +/** + * lpfc_drvr_version_show: Return the Emulex driver string with version number. + * @dev: class unused variable. + * @attr: device attribute, not used. + * @buf: on return contains the module description text. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_drvr_version_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -72,6 +95,14 @@ lpfc_drvr_version_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, LPFC_MODULE_DESC "\n"); } +/** + * lpfc_info_show: Return some pci info about the host in ascii. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains the formatted text from lpfc_info(). + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_info_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -81,6 +112,14 @@ lpfc_info_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%s\n",lpfc_info(host)); } +/** + * lpfc_serialnum_show: Return the hba serial number in ascii. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains the formatted text serial number. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_serialnum_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -92,6 +131,18 @@ lpfc_serialnum_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%s\n",phba->SerialNumber); } +/** + * lpfc_temp_sensor_show: Return the temperature sensor level. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains the formatted support level. + * + * Description: + * Returns a number indicating the temperature sensor level currently + * supported, zero or one in ascii. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_temp_sensor_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -102,6 +153,14 @@ lpfc_temp_sensor_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%d\n",phba->temp_sensor_support); } +/** + * lpfc_modeldesc_show: Return the model description of the hba. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains the scsi vpd model description. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_modeldesc_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -113,6 +172,14 @@ lpfc_modeldesc_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%s\n",phba->ModelDesc); } +/** + * lpfc_modelname_show: Return the model name of the hba. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains the scsi vpd model name. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_modelname_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -124,6 +191,14 @@ lpfc_modelname_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%s\n",phba->ModelName); } +/** + * lpfc_programtype_show: Return the program type of the hba. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains the scsi vpd program type. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_programtype_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -135,6 +210,14 @@ lpfc_programtype_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%s\n",phba->ProgramType); } +/** + * lpfc_vportnum_show: Return the port number in ascii of the hba. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains scsi vpd program type. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_vportnum_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -146,6 +229,14 @@ lpfc_vportnum_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%s\n",phba->Port); } +/** + * lpfc_fwrev_show: Return the firmware rev running in the hba. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains the scsi vpd program type. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_fwrev_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -159,6 +250,14 @@ lpfc_fwrev_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%s, sli-%d\n", fwrev, phba->sli_rev); } +/** + * lpfc_hdw_show: Return the jedec information about the hba. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains the scsi vpd program type. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_hdw_show(struct device *dev, struct device_attribute *attr, char *buf) { @@ -171,6 +270,15 @@ lpfc_hdw_show(struct device *dev, struct device_attribute *attr, char *buf) lpfc_jedec_to_ascii(vp->rev.biuRev, hdw); return snprintf(buf, PAGE_SIZE, "%s\n", hdw); } + +/** + * lpfc_option_rom_version_show: Return the adapter ROM FCode version. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains the ROM and FCode ascii strings. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_option_rom_version_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -181,6 +289,18 @@ lpfc_option_rom_version_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%s\n", phba->OptionROMVersion); } + +/** + * lpfc_state_show: Return the link state of the port. + * @dev: class converted to a Scsi_host structure. + * @attr: device attribute, not used. + * @buf: on return contains text describing the state of the link. + * + * Notes: + * The switch statement has no default so zero will be returned. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_link_state_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -253,6 +373,18 @@ lpfc_link_state_show(struct device *dev, struct device_attribute *attr, return len; } +/** + * lpfc_num_discovered_ports_show: Return sum of mapped and unmapped vports. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: on return contains the sum of fc mapped and unmapped. + * + * Description: + * Returns the ascii text number of the sum of the fc mapped and unmapped + * vport counts. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_num_discovered_ports_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -264,7 +396,20 @@ lpfc_num_discovered_ports_show(struct device *dev, vport->fc_map_cnt + vport->fc_unmap_cnt); } - +/** + * lpfc_issue_lip: Misnomer, name carried over from long ago. + * @shost: Scsi_Host pointer. + * + * Description: + * Bring the link down gracefully then re-init the link. The firmware will + * re-init the fiber channel interface as required. Does not issue a LIP. + * + * Returns: + * -EPERM port offline or management commands are being blocked + * -ENOMEM cannot allocate memory for the mailbox command + * -EIO error sending the mailbox command + * zero for success + **/ static int lpfc_issue_lip(struct Scsi_Host *shost) { @@ -306,6 +451,21 @@ lpfc_issue_lip(struct Scsi_Host *shost) return 0; } +/** + * lpfc_do_offline: Issues a mailbox command to bring the link down. + * @phba: lpfc_hba pointer. + * @type: LPFC_EVT_OFFLINE, LPFC_EVT_WARM_START, LPFC_EVT_KILL. + * + * Notes: + * Assumes any error from lpfc_do_offline() will be negative. + * Can wait up to 5 seconds for the port ring buffers count + * to reach zero, prints a warning if it is not zero and continues. + * lpfc_workq_post_event() returns a non-zero return coce if call fails. + * + * Returns: + * -EIO error posting the event + * zero for success + **/ static int lpfc_do_offline(struct lpfc_hba *phba, uint32_t type) { @@ -353,6 +513,22 @@ lpfc_do_offline(struct lpfc_hba *phba, uint32_t type) return 0; } +/** + * lpfc_selective_reset: Offline then onlines the port. + * @phba: lpfc_hba pointer. + * + * Description: + * If the port is configured to allow a reset then the hba is brought + * offline then online. + * + * Notes: + * Assumes any error from lpfc_do_offline() will be negative. + * + * Returns: + * lpfc_do_offline() return code if not zero + * -EIO reset not configured or error posting the event + * zero for success + **/ static int lpfc_selective_reset(struct lpfc_hba *phba) { @@ -378,6 +554,27 @@ lpfc_selective_reset(struct lpfc_hba *phba) return 0; } +/** + * lpfc_issue_reset: Selectively resets an adapter. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: containing the string "selective". + * @count: unused variable. + * + * Description: + * If the buf contains the string "selective" then lpfc_selective_reset() + * is called to perform the reset. + * + * Notes: + * Assumes any error from lpfc_selective_reset() will be negative. + * If lpfc_selective_reset() returns zero then the length of the buffer + * is returned which indicates succcess + * + * Returns: + * -EINVAL if the buffer does not contain the string "selective" + * length of buf if lpfc-selective_reset() if the call succeeds + * return value of lpfc_selective_reset() if the call fails +**/ static ssize_t lpfc_issue_reset(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) @@ -397,6 +594,14 @@ lpfc_issue_reset(struct device *dev, struct device_attribute *attr, return status; } +/** + * lpfc_nport_evt_cnt_show: Return the number of nport events. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: on return contains the ascii number of nport events. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_nport_evt_cnt_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -408,6 +613,14 @@ lpfc_nport_evt_cnt_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%d\n", phba->nport_event_cnt); } +/** + * lpfc_board_mode_show: Return the state of the board. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: on return contains the state of the adapter. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_board_mode_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -429,6 +642,19 @@ lpfc_board_mode_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "%s\n", state); } +/** + * lpfc_board_mode_store: Puts the hba in online, offline, warm or error state. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: containing one of the strings "online", "offline", "warm" or "error". + * @count: unused variable. + * + * Returns: + * -EACCES if enable hba reset not enabled + * -EINVAL if the buffer does not contain a valid string (see above) + * -EIO if lpfc_workq_post_event() or lpfc_do_offline() fails + * buf length greater than zero indicates success + **/ static ssize_t lpfc_board_mode_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) @@ -462,6 +688,24 @@ lpfc_board_mode_store(struct device *dev, struct device_attribute *attr, return -EIO; } +/** + * lpfc_get_hba_info: Return various bits of informaton about the adapter. + * @phba: pointer to the adapter structure. + * @mxri max xri count. + * @axri available xri count. + * @mrpi max rpi count. + * @arpi available rpi count. + * @mvpi max vpi count. + * @avpi available vpi count. + * + * Description: + * If an integer pointer for an count is not null then the value for the + * count is returned. + * + * Returns: + * zero on error + * one for success + **/ static int lpfc_get_hba_info(struct lpfc_hba *phba, uint32_t *mxri, uint32_t *axri, @@ -524,6 +768,20 @@ lpfc_get_hba_info(struct lpfc_hba *phba, return 1; } +/** + * lpfc_max_rpi_show: Return maximum rpi. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: on return contains the maximum rpi count in decimal or "Unknown". + * + * Description: + * Calls lpfc_get_hba_info() asking for just the mrpi count. + * If lpfc_get_hba_info() returns zero (failure) the buffer text is set + * to "Unknown" and the buffer length is returned, therefore the caller + * must check for "Unknown" in the buffer to detect a failure. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_max_rpi_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -538,6 +796,20 @@ lpfc_max_rpi_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "Unknown\n"); } +/** + * lpfc_used_rpi_show: Return maximum rpi minus available rpi. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: containing the used rpi count in decimal or "Unknown". + * + * Description: + * Calls lpfc_get_hba_info() asking for just the mrpi and arpi counts. + * If lpfc_get_hba_info() returns zero (failure) the buffer text is set + * to "Unknown" and the buffer length is returned, therefore the caller + * must check for "Unknown" in the buffer to detect a failure. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_used_rpi_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -552,6 +824,20 @@ lpfc_used_rpi_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "Unknown\n"); } +/** + * lpfc_max_xri_show: Return maximum xri. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: on return contains the maximum xri count in decimal or "Unknown". + * + * Description: + * Calls lpfc_get_hba_info() asking for just the mrpi count. + * If lpfc_get_hba_info() returns zero (failure) the buffer text is set + * to "Unknown" and the buffer length is returned, therefore the caller + * must check for "Unknown" in the buffer to detect a failure. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_max_xri_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -566,6 +852,20 @@ lpfc_max_xri_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "Unknown\n"); } +/** + * lpfc_used_xri_show: Return maximum xpi minus the available xpi. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: on return contains the used xri count in decimal or "Unknown". + * + * Description: + * Calls lpfc_get_hba_info() asking for just the mxri and axri counts. + * If lpfc_get_hba_info() returns zero (failure) the buffer text is set + * to "Unknown" and the buffer length is returned, therefore the caller + * must check for "Unknown" in the buffer to detect a failure. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_used_xri_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -580,6 +880,20 @@ lpfc_used_xri_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "Unknown\n"); } +/** + * lpfc_max_vpi_show: Return maximum vpi. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: on return contains the maximum vpi count in decimal or "Unknown". + * + * Description: + * Calls lpfc_get_hba_info() asking for just the mvpi count. + * If lpfc_get_hba_info() returns zero (failure) the buffer text is set + * to "Unknown" and the buffer length is returned, therefore the caller + * must check for "Unknown" in the buffer to detect a failure. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_max_vpi_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -594,6 +908,20 @@ lpfc_max_vpi_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "Unknown\n"); } +/** + * lpfc_used_vpi_show: Return maximum vpi minus the available vpi. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: on return contains the used vpi count in decimal or "Unknown". + * + * Description: + * Calls lpfc_get_hba_info() asking for just the mvpi and avpi counts. + * If lpfc_get_hba_info() returns zero (failure) the buffer text is set + * to "Unknown" and the buffer length is returned, therefore the caller + * must check for "Unknown" in the buffer to detect a failure. + * + * Returns: size of formatted string. + **/ static ssize_t lpfc_used_vpi_show(struct device *dev, struct device_attribute *attr, char *buf) @@ -608,6 +936,19 @@ lpfc_used_vpi_show(struct device *dev, struct device_attribute *attr, return snprintf(buf, PAGE_SIZE, "Unknown\n"); } +/** + * lpfc_npiv_info_show: Return text about NPIV support for the adapter. + * @dev: class device that is converted into a Scsi_host. + * @attr: device attribute, not used. + * @buf: text that must be interpreted to determine if npiv is supported. + * + * Description: + * Buffer will contain text indicating npiv is not suppoerted on the po