[SCSI] add global timeout to the scsi mid-layer
There are certain rogue devices (and the aic7xxx driver) that return BUSY or QUEUE_FULL forever. This code will apply a global timeout (of the total number of retries times the per command timer) to a given command. If it is exceeded, the command is completed regardless of its state. The patch also removes the unused field in the command: timeout and timeout_total. This solves the problem of detecting an endless loop in the mid-layer because of BUSY/QUEUE_FULL bouncing, but will not recover the device. In the aic7xxx case, the driver can be recovered by sending a bus reset, so possibly this should be tied into the error handler? Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
This commit is contained in:
parent
f03a567054
commit
b21a413851
|
@ -9200,8 +9200,8 @@ asc_prt_scsi_cmnd(struct scsi_cmnd *s)
|
||||||
(unsigned) s->serial_number, s->retries, s->allowed);
|
(unsigned) s->serial_number, s->retries, s->allowed);
|
||||||
|
|
||||||
printk(
|
printk(
|
||||||
" timeout_per_command %d, timeout_total %d, timeout %d\n",
|
" timeout_per_command %d\n",
|
||||||
s->timeout_per_command, s->timeout_total, s->timeout);
|
s->timeout_per_command);
|
||||||
|
|
||||||
printk(
|
printk(
|
||||||
" scsi_done 0x%lx, done 0x%lx, host_scribble 0x%lx, result 0x%x\n",
|
" scsi_done 0x%lx, done 0x%lx, host_scribble 0x%lx, result 0x%x\n",
|
||||||
|
|
|
@ -268,6 +268,7 @@ struct scsi_cmnd *scsi_get_command(struct scsi_device *dev, int gfp_mask)
|
||||||
} else
|
} else
|
||||||
put_device(&dev->sdev_gendev);
|
put_device(&dev->sdev_gendev);
|
||||||
|
|
||||||
|
cmd->jiffies_at_alloc = jiffies;
|
||||||
return cmd;
|
return cmd;
|
||||||
}
|
}
|
||||||
EXPORT_SYMBOL(scsi_get_command);
|
EXPORT_SYMBOL(scsi_get_command);
|
||||||
|
@ -798,9 +799,23 @@ static void scsi_softirq(struct softirq_action *h)
|
||||||
while (!list_empty(&local_q)) {
|
while (!list_empty(&local_q)) {
|
||||||
struct scsi_cmnd *cmd = list_entry(local_q.next,
|
struct scsi_cmnd *cmd = list_entry(local_q.next,
|
||||||
struct scsi_cmnd, eh_entry);
|
struct scsi_cmnd, eh_entry);
|
||||||
|
/* The longest time any command should be outstanding is the
|
||||||
|
* per command timeout multiplied by the number of retries.
|
||||||
|
*
|
||||||
|
* For a typical command, this is 2.5 minutes */
|
||||||
|
unsigned long wait_for
|
||||||
|
= cmd->allowed * cmd->timeout_per_command;
|
||||||
list_del_init(&cmd->eh_entry);
|
list_del_init(&cmd->eh_entry);
|
||||||
|
|
||||||
disposition = scsi_decide_disposition(cmd);
|
disposition = scsi_decide_disposition(cmd);
|
||||||
|
if (disposition != SUCCESS &&
|
||||||
|
time_before(cmd->jiffies_at_alloc + wait_for, jiffies)) {
|
||||||
|
dev_printk(KERN_ERR, &cmd->device->sdev_gendev,
|
||||||
|
"timing out command, waited %ds\n",
|
||||||
|
wait_for/HZ);
|
||||||
|
disposition = SUCCESS;
|
||||||
|
}
|
||||||
|
|
||||||
scsi_log_completion(cmd, disposition);
|
scsi_log_completion(cmd, disposition);
|
||||||
switch (disposition) {
|
switch (disposition) {
|
||||||
case SUCCESS:
|
case SUCCESS:
|
||||||
|
|
|
@ -51,12 +51,16 @@ struct scsi_cmnd {
|
||||||
* printk's to use ->pid, so that we can kill this field.
|
* printk's to use ->pid, so that we can kill this field.
|
||||||
*/
|
*/
|
||||||
unsigned long serial_number;
|
unsigned long serial_number;
|
||||||
|
/*
|
||||||
|
* This is set to jiffies as it was when the command was first
|
||||||
|
* allocated. It is used to time how long the command has
|
||||||
|
* been outstanding
|
||||||
|
*/
|
||||||
|
unsigned long jiffies_at_alloc;
|
||||||
|
|
||||||
int retries;
|
int retries;
|
||||||
int allowed;
|
int allowed;
|
||||||
int timeout_per_command;
|
int timeout_per_command;
|
||||||
int timeout_total;
|
|
||||||
int timeout;
|
|
||||||
|
|
||||||
unsigned char cmd_len;
|
unsigned char cmd_len;
|
||||||
unsigned char old_cmd_len;
|
unsigned char old_cmd_len;
|
||||||
|
|
Loading…
Reference in New Issue