[SCSI] ibmvfc: Fix soft lockup on resume
This fixes a softlockup seen on resume. During resume, the CRQ
must be reenabled. However, the H_ENABLE_CRQ hcall used to do
this may return H_BUSY or H_LONG_BUSY. When this happens, the
caller is expected to retry later. Normally the H_ENABLE_CRQ
succeeds relatively soon. However, we have seen cases where
this can take long enough to see softlockup warnings.
This patch changes a simple loop, which was causing the
softlockup, to a loop at task level which sleeps between
retries rather than simply spinning.
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
diff --git a/drivers/scsi/ibmvscsi/ibmvfc.c b/drivers/scsi/ibmvscsi/ibmvfc.c
index fef4952..d6fcb3f 100644
--- a/drivers/scsi/ibmvscsi/ibmvfc.c
+++ b/drivers/scsi/ibmvscsi/ibmvfc.c
@@ -504,12 +504,23 @@
if (vhost->action == IBMVFC_HOST_ACTION_ALLOC_TGTS)
vhost->action = action;
break;
- case IBMVFC_HOST_ACTION_LOGO:
case IBMVFC_HOST_ACTION_INIT:
case IBMVFC_HOST_ACTION_TGT_DEL:
+ switch (vhost->action) {
+ case IBMVFC_HOST_ACTION_RESET:
+ case IBMVFC_HOST_ACTION_REENABLE:
+ break;
+ default:
+ vhost->action = action;
+ break;
+ };
+ break;
+ case IBMVFC_HOST_ACTION_LOGO:
case IBMVFC_HOST_ACTION_QUERY_TGTS:
case IBMVFC_HOST_ACTION_TGT_DEL_FAILED:
case IBMVFC_HOST_ACTION_NONE:
+ case IBMVFC_HOST_ACTION_RESET:
+ case IBMVFC_HOST_ACTION_REENABLE:
default:
vhost->action = action;
break;
@@ -641,7 +652,7 @@
**/
static void ibmvfc_release_crq_queue(struct ibmvfc_host *vhost)
{
- long rc;
+ long rc = 0;
struct vio_dev *vdev = to_vio_dev(vhost->dev);
struct ibmvfc_crq_queue *crq = &vhost->crq;
@@ -649,6 +660,8 @@
free_irq(vdev->irq, vhost);
tasklet_kill(&vhost->tasklet);
do {
+ if (rc)
+ msleep(100);
rc = plpar_hcall_norets(H_FREE_CRQ, vdev->unit_address);
} while (rc == H_BUSY || H_IS_LONG_BUSY(rc));
@@ -667,11 +680,13 @@
**/
static int ibmvfc_reenable_crq_queue(struct ibmvfc_host *vhost)
{
- int rc;
+ int rc = 0;
struct vio_dev *vdev = to_vio_dev(vhost->dev);
/* Re-enable the CRQ */
do {
+ if (rc)
+ msleep(100);
rc = plpar_hcall_norets(H_ENABLE_CRQ, vdev->unit_address);
} while (rc == H_IN_PROGRESS || rc == H_BUSY || H_IS_LONG_BUSY(rc));
@@ -690,15 +705,19 @@
**/
static int ibmvfc_reset_crq(struct ibmvfc_host *vhost)
{
- int rc;
+ int rc = 0;
+ unsigned long flags;
struct vio_dev *vdev = to_vio_dev(vhost->dev);
struct ibmvfc_crq_queue *crq = &vhost->crq;
/* Close the CRQ */
do {
+ if (rc)
+ msleep(100);
rc = plpar_hcall_norets(H_FREE_CRQ, vdev->unit_address);
} while (rc == H_BUSY || H_IS_LONG_BUSY(rc));
+ spin_lock_irqsave(vhost->host->host_lock, flags);
vhost->state = IBMVFC_NO_CRQ;
vhost->logged_in = 0;
ibmvfc_set_host_action(vhost, IBMVFC_HOST_ACTION_NONE);
@@ -716,6 +735,7 @@
dev_warn(vhost->dev, "Partner adapter not ready\n");
else if (rc != 0)
dev_warn(vhost->dev, "Couldn't register crq (rc=%d)\n", rc);
+ spin_unlock_irqrestore(vhost->host->host_lock, flags);
return rc;
}
@@ -821,17 +841,9 @@
**/
static void ibmvfc_hard_reset_host(struct ibmvfc_host *vhost)
{
- int rc;
-
- scsi_block_requests(vhost->host);
ibmvfc_purge_requests(vhost, DID_ERROR);
- if ((rc = ibmvfc_reset_crq(vhost)) ||
- (rc = ibmvfc_send_crq_init(vhost)) ||
- (rc = vio_enable_interrupts(to_vio_dev(vhost->dev)))) {
- dev_err(vhost->dev, "Error after reset rc=%d\n", rc);
- ibmvfc_link_down(vhost, IBMVFC_LINK_DEAD);
- } else
- ibmvfc_link_down(vhost, IBMVFC_LINK_DOWN);
+ ibmvfc_link_down(vhost, IBMVFC_LINK_DOWN);
+ ibmvfc_set_host_action(vhost, IBMVFC_HOST_ACTION_RESET);
}
/**
@@ -2606,22 +2618,13 @@
dev_info(vhost->dev, "Re-enabling adapter\n");
vhost->client_migrated = 1;
ibmvfc_purge_requests(vhost, DID_REQUEUE);
- if ((rc = ibmvfc_reenable_crq_queue(vhost)) ||
- (rc = ibmvfc_send_crq_init(vhost))) {
- ibmvfc_link_down(vhost, IBMVFC_LINK_DEAD);
- dev_err(vhost->dev, "Error after enable (rc=%ld)\n", rc);
- } else
- ibmvfc_link_down(vhost, IBMVFC_LINK_DOWN);
+ ibmvfc_link_down(vhost, IBMVFC_LINK_DOWN);
+ ibmvfc_set_host_action(vhost, IBMVFC_HOST_ACTION_REENABLE);
} else {
dev_err(vhost->dev, "Virtual adapter failed (rc=%d)\n", crq->format);
-
ibmvfc_purge_requests(vhost, DID_ERROR);
- if ((rc = ibmvfc_reset_crq(vhost)) ||
- (rc = ibmvfc_send_crq_init(vhost))) {
- ibmvfc_link_down(vhost, IBMVFC_LINK_DEAD);
- dev_err(vhost->dev, "Error after reset (rc=%ld)\n", rc);
- } else
- ibmvfc_link_down(vhost, IBMVFC_LINK_DOWN);
+ ibmvfc_link_down(vhost, IBMVFC_LINK_DOWN);
+ ibmvfc_set_host_action(vhost, IBMVFC_HOST_ACTION_RESET);
}
return;
case IBMVFC_CRQ_CMD_RSP:
@@ -4123,6 +4126,8 @@
case IBMVFC_HOST_ACTION_TGT_DEL:
case IBMVFC_HOST_ACTION_TGT_DEL_FAILED:
case IBMVFC_HOST_ACTION_QUERY:
+ case IBMVFC_HOST_ACTION_RESET:
+ case IBMVFC_HOST_ACTION_REENABLE:
default:
break;
};
@@ -4220,6 +4225,7 @@
struct ibmvfc_target *tgt;
unsigned long flags;
struct fc_rport *rport;
+ int rc;
ibmvfc_log_ae(vhost, vhost->events_to_log);
spin_lock_irqsave(vhost->host->host_lock, flags);
@@ -4229,6 +4235,27 @@
case IBMVFC_HOST_ACTION_LOGO_WAIT:
case IBMVFC_HOST_ACTION_INIT_WAIT:
break;
+ case IBMVFC_HOST_ACTION_RESET:
+ vhost->action = IBMVFC_HOST_ACTION_TGT_DEL;
+ spin_unlock_irqrestore(vhost->host->host_lock, flags);
+ rc = ibmvfc_reset_crq(vhost);
+ spin_lock_irqsave(vhost->host->host_lock, flags);
+ if (rc || (rc = ibmvfc_send_crq_init(vhost)) ||
+ (rc = vio_enable_interrupts(to_vio_dev(vhost->dev)))) {
+ ibmvfc_link_down(vhost, IBMVFC_LINK_DEAD);
+ dev_err(vhost->dev, "Error after reset (rc=%d)\n", rc);
+ }
+ break;
+ case IBMVFC_HOST_ACTION_REENABLE:
+ vhost->action = IBMVFC_HOST_ACTION_TGT_DEL;
+ spin_unlock_irqrestore(vhost->host->host_lock, flags);
+ rc = ibmvfc_reenable_crq_queue(vhost);
+ spin_lock_irqsave(vhost->host->host_lock, flags);
+ if (rc || (rc = ibmvfc_send_crq_init(vhost))) {
+ ibmvfc_link_down(vhost, IBMVFC_LINK_DEAD);
+ dev_err(vhost->dev, "Error after enable (rc=%d)\n", rc);
+ }
+ break;
case IBMVFC_HOST_ACTION_LOGO:
vhost->job_step(vhost);
break;