[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20061011133600.GE9305@skybase>
Date: Wed, 11 Oct 2006 15:36:00 +0200
From: Martin Schwidefsky <schwidefsky@...ibm.com>
To: linux-kernel@...r.kernel.org, cornelia.huck@...ibm.com
Subject: [S390] cio: Remove grace period for vary off chpid.
From: Cornelia Huck <cornelia.huck@...ibm.com>
[S390] cio: Remove grace period for vary off chpid.
The grace period handling introduced needless complexity. It didn't
help the dasd driver (which can handle terminated I/O just well),
and it doesn't help for long running channel programs (which won't
complete during the grace period anyway). Terminating I/O using a
path that just disappeared immediately is much more consistent with
what the user expects.
Signed-off-by: Cornelia Huck <cornelia.huck@...ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@...ibm.com>
---
drivers/s390/cio/chsc.c | 17 +++-------
drivers/s390/cio/css.h | 2 -
drivers/s390/cio/device.h | 1
drivers/s390/cio/device_fsm.c | 68 ++++--------------------------------------
drivers/s390/cio/device_ops.c | 2 -
5 files changed, 14 insertions(+), 76 deletions(-)
diff -urpN linux-2.6/drivers/s390/cio/chsc.c linux-2.6-patched/drivers/s390/cio/chsc.c
--- linux-2.6/drivers/s390/cio/chsc.c 2006-10-11 15:23:22.000000000 +0200
+++ linux-2.6-patched/drivers/s390/cio/chsc.c 2006-10-11 15:23:38.000000000 +0200
@@ -707,8 +707,7 @@ chp_process_crw(int chpid, int on)
return chp_add(chpid);
}
-static inline int
-__check_for_io_and_kill(struct subchannel *sch, int index)
+static inline int check_for_io_on_path(struct subchannel *sch, int index)
{
int cc;
@@ -718,10 +717,8 @@ __check_for_io_and_kill(struct subchanne
cc = stsch(sch->schid, &sch->schib);
if (cc)
return 0;
- if (sch->schib.scsw.actl && sch->schib.pmcw.lpum == (0x80 >> index)) {
- device_set_waiting(sch);
+ if (sch->schib.scsw.actl && sch->schib.pmcw.lpum == (0x80 >> index))
return 1;
- }
return 0;
}
@@ -750,12 +747,10 @@ __s390_subchannel_vary_chpid(struct subc
} else {
sch->opm &= ~(0x80 >> chp);
sch->lpm &= ~(0x80 >> chp);
- /*
- * Give running I/O a grace period in which it
- * can successfully terminate, even using the
- * just varied off path. Then kill it.
- */
- if (!__check_for_io_and_kill(sch, chp) && !sch->lpm) {
+ if (check_for_io_on_path(sch, chp))
+ /* Path verification is done after killing. */
+ device_kill_io(sch);
+ else if (!sch->lpm) {
if (css_enqueue_subchannel_slow(sch->schid)) {
css_clear_subchannel_slow_list();
need_rescan = 1;
diff -urpN linux-2.6/drivers/s390/cio/css.h linux-2.6-patched/drivers/s390/cio/css.h
--- linux-2.6/drivers/s390/cio/css.h 2006-10-11 15:23:38.000000000 +0200
+++ linux-2.6-patched/drivers/s390/cio/css.h 2006-10-11 15:23:38.000000000 +0200
@@ -170,7 +170,7 @@ void device_trigger_reprobe(struct subch
/* Helper functions for vary on/off. */
int device_is_online(struct subchannel *);
-void device_set_waiting(struct subchannel *);
+void device_kill_io(struct subchannel *);
/* Machine check helper function. */
void device_kill_pending_timer(struct subchannel *);
diff -urpN linux-2.6/drivers/s390/cio/device_fsm.c linux-2.6-patched/drivers/s390/cio/device_fsm.c
--- linux-2.6/drivers/s390/cio/device_fsm.c 2006-10-11 15:23:38.000000000 +0200
+++ linux-2.6-patched/drivers/s390/cio/device_fsm.c 2006-10-11 15:23:38.000000000 +0200
@@ -59,18 +59,6 @@ device_set_disconnected(struct subchanne
cdev->private->state = DEV_STATE_DISCONNECTED;
}
-void
-device_set_waiting(struct subchannel *sch)
-{
- struct ccw_device *cdev;
-
- if (!sch->dev.driver_data)
- return;
- cdev = sch->dev.driver_data;
- ccw_device_set_timeout(cdev, 10*HZ);
- cdev->private->state = DEV_STATE_WAIT4IO;
-}
-
/*
* Timeout function. It just triggers a DEV_EVENT_TIMEOUT.
*/
@@ -947,7 +935,7 @@ ccw_device_killing_irq(struct ccw_device
cdev->private->state = DEV_STATE_ONLINE;
if (cdev->handler)
cdev->handler(cdev, cdev->private->intparm,
- ERR_PTR(-ETIMEDOUT));
+ ERR_PTR(-EIO));
if (!sch->lpm) {
PREPARE_WORK(&cdev->private->kick_work,
ccw_device_nopath_notify, (void *)cdev);
@@ -984,51 +972,15 @@ ccw_device_killing_timeout(struct ccw_de
cdev->private->state = DEV_STATE_ONLINE;
if (cdev->handler)
cdev->handler(cdev, cdev->private->intparm,
- ERR_PTR(-ETIMEDOUT));
+ ERR_PTR(-EIO));
}
-static void
-ccw_device_wait4io_irq(struct ccw_device *cdev, enum dev_event dev_event)
-{
- struct irb *irb;
- struct subchannel *sch;
-
- irb = (struct irb *) __LC_IRB;
- /*
- * Accumulate status and find out if a basic sense is needed.
- * This is fine since we have already adapted the lpm.
- */
- ccw_device_accumulate_irb(cdev, irb);
- if (cdev->private->flags.dosense) {
- if (ccw_device_do_sense(cdev, irb) == 0) {
- cdev->private->state = DEV_STATE_W4SENSE;
- }
- return;
- }
-
- /* Iff device is idle, reset timeout. */
- sch = to_subchannel(cdev->dev.parent);
- if (!stsch(sch->schid, &sch->schib))
- if (sch->schib.scsw.actl == 0)
- ccw_device_set_timeout(cdev, 0);
- /* Call the handler. */
- ccw_device_call_handler(cdev);
- if (!sch->lpm) {
- PREPARE_WORK(&cdev->private->kick_work,
- ccw_device_nopath_notify, (void *)cdev);
- queue_work(ccw_device_notify_work, &cdev->private->kick_work);
- } else if (cdev->private->flags.doverify)
- ccw_device_online_verify(cdev, 0);
-}
-
-static void
-ccw_device_wait4io_timeout(struct ccw_device *cdev, enum dev_event dev_event)
+void device_kill_io(struct subchannel *sch)
{
int ret;
- struct subchannel *sch;
+ struct ccw_device *cdev;
- sch = to_subchannel(cdev->dev.parent);
- ccw_device_set_timeout(cdev, 0);
+ cdev = sch->dev.driver_data;
ret = ccw_device_cancel_halt_clear(cdev);
if (ret == -EBUSY) {
ccw_device_set_timeout(cdev, 3*HZ);
@@ -1047,12 +999,12 @@ ccw_device_wait4io_timeout(struct ccw_de
}
if (cdev->handler)
cdev->handler(cdev, cdev->private->intparm,
- ERR_PTR(-ETIMEDOUT));
+ ERR_PTR(-EIO));
if (!sch->lpm) {
PREPARE_WORK(&cdev->private->kick_work,
ccw_device_nopath_notify, (void *)cdev);
queue_work(ccw_device_notify_work, &cdev->private->kick_work);
- } else if (cdev->private->flags.doverify)
+ } else
/* Start delayed path verification. */
ccw_device_online_verify(cdev, 0);
}
@@ -1289,12 +1241,6 @@ fsm_func_t *dev_jumptable[NR_DEV_STATES]
[DEV_EVENT_TIMEOUT] = ccw_device_killing_timeout,
[DEV_EVENT_VERIFY] = ccw_device_nop, //FIXME
},
- [DEV_STATE_WAIT4IO] = {
- [DEV_EVENT_NOTOPER] = ccw_device_online_notoper,
- [DEV_EVENT_INTERRUPT] = ccw_device_wait4io_irq,
- [DEV_EVENT_TIMEOUT] = ccw_device_wait4io_timeout,
- [DEV_EVENT_VERIFY] = ccw_device_delay_verify,
- },
[DEV_STATE_QUIESCE] = {
[DEV_EVENT_NOTOPER] = ccw_device_quiesce_done,
[DEV_EVENT_INTERRUPT] = ccw_device_quiesce_done,
diff -urpN linux-2.6/drivers/s390/cio/device.h linux-2.6-patched/drivers/s390/cio/device.h
--- linux-2.6/drivers/s390/cio/device.h 2006-09-20 05:42:06.000000000 +0200
+++ linux-2.6-patched/drivers/s390/cio/device.h 2006-10-11 15:23:38.000000000 +0200
@@ -21,7 +21,6 @@ enum dev_state {
/* states to wait for i/o completion before doing something */
DEV_STATE_CLEAR_VERIFY,
DEV_STATE_TIMEOUT_KILL,
- DEV_STATE_WAIT4IO,
DEV_STATE_QUIESCE,
/* special states for devices gone not operational */
DEV_STATE_DISCONNECTED,
diff -urpN linux-2.6/drivers/s390/cio/device_ops.c linux-2.6-patched/drivers/s390/cio/device_ops.c
--- linux-2.6/drivers/s390/cio/device_ops.c 2006-10-11 15:23:38.000000000 +0200
+++ linux-2.6-patched/drivers/s390/cio/device_ops.c 2006-10-11 15:23:38.000000000 +0200
@@ -50,7 +50,6 @@ ccw_device_clear(struct ccw_device *cdev
if (cdev->private->state == DEV_STATE_NOT_OPER)
return -ENODEV;
if (cdev->private->state != DEV_STATE_ONLINE &&
- cdev->private->state != DEV_STATE_WAIT4IO &&
cdev->private->state != DEV_STATE_W4SENSE)
return -EINVAL;
sch = to_subchannel(cdev->dev.parent);
@@ -155,7 +154,6 @@ ccw_device_halt(struct ccw_device *cdev,
if (cdev->private->state == DEV_STATE_NOT_OPER)
return -ENODEV;
if (cdev->private->state != DEV_STATE_ONLINE &&
- cdev->private->state != DEV_STATE_WAIT4IO &&
cdev->private->state != DEV_STATE_W4SENSE)
return -EINVAL;
sch = to_subchannel(cdev->dev.parent);
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists