From: Andrew Vasquez <andrew.vasquez@qlogic.com>
To: SCSI Mailing List <linux-scsi@vger.kernel.org>,
James Bottomley <James.Bottomley@steeleye.com>
Subject: PATCH [9/18] qla2xxx: Tape command handling fixes
Date: Mon, 21 Jun 2004 22:55:02 -0700 [thread overview]
Message-ID: <20040622055502.GA8446@linux.local.home> (raw)
ChangeSet
1.1845 04/06/04 08:58:01 andrew.vasquez@apc.qlogic.com +2 -0
Address several outstanding problem within the various
eh_*() functions:
o Fixup nested spinlock usage in error-hanlding
functions.
o Set the Loop down timer during abort isp if loop
was already down for the driver to detect cable
is unplugged at min time.
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
drivers/scsi/qla2xxx/qla_init.c | 4 ++
drivers/scsi/qla2xxx/qla_os.c | 72 ++++++++++++++++------------------------
2 files changed, 34 insertions(+), 42 deletions(-)
diff -Nru a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
--- a/drivers/scsi/qla2xxx/qla_init.c 2004-06-21 15:38:28 -07:00
+++ b/drivers/scsi/qla2xxx/qla_init.c 2004-06-21 15:38:28 -07:00
@@ -4116,6 +4116,10 @@
atomic_set(&ha->loop_state, LOOP_DOWN);
atomic_set(&ha->loop_down_timer, LOOP_DOWN_TIME);
qla2x00_mark_all_devices_lost(ha);
+ } else {
+ if (!atomic_read(&ha->loop_down_timer))
+ atomic_set(&ha->loop_down_timer,
+ LOOP_DOWN_TIME);
}
spin_lock_irqsave(&ha->hardware_lock, flags);
diff -Nru a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
--- a/drivers/scsi/qla2xxx/qla_os.c 2004-06-21 15:38:28 -07:00
+++ b/drivers/scsi/qla2xxx/qla_os.c 2004-06-21 15:38:28 -07:00
@@ -882,13 +882,12 @@
int done = 0;
srb_t *rp;
struct list_head *list, *temp;
- u_long cpu_flags = 0;
u_long max_wait_time = ABORT_WAIT_TIME;
do {
/* Check on done queue */
if (!found) {
- spin_lock_irqsave(&ha->list_lock, cpu_flags);
+ spin_lock(&ha->list_lock);
list_for_each_safe(list, temp, &ha->done_queue) {
rp = list_entry(list, srb_t, list);
@@ -903,7 +902,7 @@
break;
}
}
- spin_unlock_irqrestore(&ha->list_lock, cpu_flags);
+ spin_unlock(&ha->list_lock);
}
/* Checking to see if its returned to OS */
@@ -951,7 +950,7 @@
* Success (Adapter is online) : 0
* Failed (Adapter is offline/disabled) : 1
*/
-static inline int
+int
qla2x00_wait_for_hba_online(scsi_qla_host_t *ha)
{
int return_status;
@@ -960,8 +959,8 @@
wait_online = jiffies + (MAX_LOOP_TIMEOUT * HZ);
while (((test_bit(ISP_ABORT_NEEDED, &ha->dpc_flags)) ||
test_bit(ABORT_ISP_ACTIVE, &ha->dpc_flags) ||
- test_bit(ISP_ABORT_RETRY, &ha->dpc_flags)) &&
- time_before(jiffies, wait_online)) {
+ test_bit(ISP_ABORT_RETRY, &ha->dpc_flags) ||
+ ha->dpc_active) && time_before(jiffies, wait_online)) {
set_current_state(TASK_UNINTERRUPTIBLE);
schedule_timeout(HZ);
@@ -969,7 +968,6 @@
if (ha->flags.online == TRUE)
return_status = QLA_SUCCESS;
else
- /* Adapter is disabled/offline */
return_status = QLA_FUNCTION_FAILED;
DEBUG2(printk("%s return_status=%d\n",__func__,return_status));
@@ -1044,7 +1042,6 @@
struct Scsi_Host *host;
uint8_t found = 0;
unsigned int b, t, l;
- unsigned long flags;
/* Get the SCSI request ptr */
sp = (srb_t *) CMD_SP(cmd);
@@ -1059,7 +1056,7 @@
qla_printk(KERN_INFO, to_qla_host(cmd->device->host),
"qla2xxx_eh_abort: cmd already done sp=%p\n", sp);
DEBUG(printk("qla2xxx_eh_abort: cmd already done sp=%p\n", sp);)
- return(SUCCESS);
+ return SUCCESS;
}
if (sp) {
DEBUG(printk("qla2xxx_eh_abort: refcount %i \n",
@@ -1090,7 +1087,7 @@
"qla2x00: (%x:%x:%x) No LUN queue.\n", b, t, l);
/* no action - we don't have command */
- return(FAILED);
+ return FAILED;
}
DEBUG2(printk("scsi(%ld): ABORTing cmd=%p sp=%p jiffies = 0x%lx, "
@@ -1100,16 +1097,15 @@
DEBUG2(qla2x00_print_scsi_cmd(cmd));
spin_unlock_irq(ha->host->host_lock);
- /* Blocking call-Does context switching if abort isp is active etc */
if (qla2x00_wait_for_hba_online(ha) != QLA_SUCCESS) {
DEBUG2(printk("%s failed:board disabled\n", __func__);)
spin_lock_irq(ha->host->host_lock);
- return (FAILED);
+ return FAILED;
}
spin_lock_irq(ha->host->host_lock);
/* Search done queue */
- spin_lock_irqsave(&ha->list_lock,flags);
+ spin_lock(&ha->list_lock);
list_for_each_safe(list, temp, &ha->done_queue) {
rp = list_entry(list, srb_t, list);
@@ -1126,7 +1122,7 @@
break;
} /* list_for_each_safe() */
- spin_unlock_irqrestore(&ha->list_lock, flags);
+ spin_unlock(&ha->list_lock);
/*
* Return immediately if the aborted command was already in the done
@@ -1147,7 +1143,7 @@
DEBUG3(printk("qla2xxx_eh_abort: searching sp %p in retry "
"queue.\n", sp);)
- spin_lock_irqsave(&ha->list_lock, flags);
+ spin_lock(&ha->list_lock);
list_for_each_safe(list, temp, &ha->retry_queue) {
rp = list_entry(list, srb_t, list);
@@ -1168,7 +1164,7 @@
break;
}
- spin_unlock_irqrestore(&ha->list_lock, flags);
+ spin_unlock(&ha->list_lock);
/*
@@ -1179,7 +1175,7 @@
DEBUG3(printk("qla2xxx_eh_abort: searching sp %p "
"in pending queue.\n", sp);)
- spin_lock_irqsave(&vis_ha->list_lock, flags);
+ spin_lock(&vis_ha->list_lock);
list_for_each_safe(list, temp, &vis_ha->pending_queue) {
rp = list_entry(list, srb_t, list);
@@ -1203,14 +1199,14 @@
found++;
break;
} /* list_for_each_safe() */
- spin_unlock_irqrestore(&vis_ha->list_lock, flags);
+ spin_unlock(&vis_ha->list_lock);
} /*End of if !found */
if (!found) { /* find the command in our active list */
DEBUG3(printk("qla2xxx_eh_abort: searching sp %p "
"in outstanding queue.\n", sp);)
- spin_lock_irqsave(&ha->hardware_lock, flags);
+ spin_lock(&ha->hardware_lock);
for (i = 1; i < MAX_OUTSTANDING_COMMANDS; i++) {
sp = ha->outstanding_cmds[i];
@@ -1229,8 +1225,8 @@
/* Get a reference to the sp and drop the lock.*/
sp_get(ha, sp);
- spin_unlock_irqrestore(&ha->hardware_lock, flags);
- spin_unlock(ha->host->host_lock);
+ spin_unlock(&ha->hardware_lock);
+ spin_unlock_irq(ha->host->host_lock);
if (qla2x00_abort_command(ha, sp)) {
DEBUG2(printk("qla2xxx_eh_abort: abort_command "
@@ -1245,7 +1241,7 @@
sp_put(ha,sp);
spin_lock_irq(ha->host->host_lock);
- spin_lock_irqsave(&ha->hardware_lock, flags);
+ spin_lock(&ha->hardware_lock);
/*
* Regardless of mailbox command status, go check on
@@ -1254,11 +1250,11 @@
break;
}/*End of for loop */
- spin_unlock_irqrestore(&ha->hardware_lock, flags);
+ spin_unlock(&ha->hardware_lock);
} /*End of if !found */
- /*Waiting for our command in done_queue to be returned to OS.*/
+ /* Waiting for our command in done_queue to be returned to OS.*/
if (qla2x00_eh_wait_on_command(ha, cmd) != 0) {
DEBUG2(printk("qla2xxx_eh_abort: cmd returned back to OS.\n");)
return_status = SUCCESS;
@@ -1273,7 +1269,7 @@
DEBUG2(printk("qla2xxx_eh_abort: Exiting. return_status=0x%x.\n",
return_status));
- return(return_status);
+ return return_status;
}
/**************************************************************************
@@ -1295,7 +1291,6 @@
{
int cnt;
int status;
- unsigned long flags;
srb_t *sp;
struct scsi_cmnd *cmd;
@@ -1306,11 +1301,11 @@
* array
*/
for (cnt = 1; cnt < MAX_OUTSTANDING_COMMANDS; cnt++) {
- spin_lock_irqsave(&ha->hardware_lock, flags);
+ spin_lock(&ha->hardware_lock);
sp = ha->outstanding_cmds[cnt];
if (sp) {
cmd = sp->cmd;
- spin_unlock_irqrestore(&ha->hardware_lock, flags);
+ spin_unlock(&ha->hardware_lock);
if (cmd->device->id == t) {
if (!qla2x00_eh_wait_on_command(ha, cmd)) {
status = 1;
@@ -1319,7 +1314,7 @@
}
}
else {
- spin_unlock_irqrestore(&ha->hardware_lock, flags);
+ spin_unlock(&ha->hardware_lock);
}
}
return (status);
@@ -1354,7 +1349,6 @@
os_tgt_t *tq;
os_lun_t *lq;
fc_port_t *fcport_to_reset;
- unsigned long flags;
srb_t *rp;
struct list_head *list, *temp;
@@ -1402,7 +1396,7 @@
ha->dpc_flags, cmd->result, cmd->allowed, cmd->state));
/* Clear commands from the retry queue. */
- spin_lock_irqsave(&ha->list_lock, flags);
+ spin_lock(&ha->list_lock);
list_for_each_safe(list, temp, &ha->retry_queue) {
rp = list_entry(list, srb_t, list);
@@ -1416,11 +1410,10 @@
rp->cmd->result = DID_RESET << 16;
__add_to_done_queue(ha, rp);
}
- spin_unlock_irqrestore(&ha->list_lock, flags);
+ spin_unlock(&ha->list_lock);
spin_unlock_irq(ha->host->host_lock);
- /* Blocking call-Does context switching if abort isp is active etc */
if (qla2x00_wait_for_hba_online(ha) != QLA_SUCCESS) {
DEBUG2(printk(KERN_INFO
"%s failed:board disabled\n",__func__));
@@ -1429,7 +1422,6 @@
goto eh_dev_reset_done;
}
- /* Blocking call-Does context switching if loop is Not Ready */
if (qla2x00_wait_for_loop_ready(ha) == QLA_SUCCESS) {
if (qla2x00_device_reset(ha, fcport_to_reset) == 0) {
return_status = SUCCESS;
@@ -1453,7 +1445,7 @@
if (return_status == FAILED) {
DEBUG3(printk("%s(%ld): device reset failed\n",
- __func__,ha->host_no));
+ __func__, ha->host_no));
qla_printk(KERN_INFO, ha, "%s: device reset failed\n",
__func__);
@@ -1511,7 +1503,6 @@
{
int cnt;
int status;
- unsigned long flags;
srb_t *sp;
struct scsi_cmnd *cmd;
@@ -1522,17 +1513,17 @@
* array
*/
for (cnt = 1; cnt < MAX_OUTSTANDING_COMMANDS; cnt++) {
- spin_lock_irqsave(&ha->hardware_lock, flags);
+ spin_lock(&ha->hardware_lock);
sp = ha->outstanding_cmds[cnt];
if (sp) {
cmd = sp->cmd;
- spin_unlock_irqrestore(&ha->hardware_lock, flags);
+ spin_unlock(&ha->hardware_lock);
status = qla2x00_eh_wait_on_command(ha, cmd);
if (status == 0)
break;
}
else {
- spin_unlock_irqrestore(&ha->hardware_lock, flags);
+ spin_unlock(&ha->hardware_lock);
}
}
return (status);
@@ -1570,14 +1561,12 @@
spin_unlock_irq(ha->host->host_lock);
- /* Blocking call-Does context switching if abort isp is active etc*/
if (qla2x00_wait_for_hba_online(ha) != QLA_SUCCESS) {
DEBUG2(printk("%s failed:board disabled\n",__func__));
spin_lock_irq(ha->host->host_lock);
return FAILED;
}
- /* Blocking call-Does context switching if loop is Not Ready */
if (qla2x00_wait_for_loop_ready(ha) == QLA_SUCCESS) {
if (qla2x00_loop_reset(ha))
rval = SUCCESS;
@@ -1637,7 +1626,6 @@
spin_unlock_irq(ha->host->host_lock);
- /* Blocking call-Does context switching if abort isp is active etc*/
if (qla2x00_wait_for_hba_online(ha) != QLA_SUCCESS)
goto board_disabled;
reply other threads:[~2004-06-22 5:51 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20040622055502.GA8446@linux.local.home \
--to=andrew.vasquez@qlogic.com \
--cc=James.Bottomley@steeleye.com \
--cc=linux-scsi@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.