2002-11-07 20:44:20

by Dave Olien

Subject: [BUG] 2.5.46 reloading a block device module doesn't probe devices

I just converted the DAC960 driver in 2.5.46 to use the pci_register_driver()
to discover devices.

I've been experimenting loading the DAC960 driver module, unloading
the module, and then loading it again. I get curious behavior.
The load, unload seem to work fine.

The reload also APPEARS to succeed. The insmod command completes
with a successful status.

But the second loading of the driver never calls the driver's probe routines.
I added printfs, and determined that the module_init() function
DOES get called the second time. But a printf in the probe() routine
doesn't show up the second time.

Has there been a patch recently for this problem?

Thanks!

Dave Olien

2002-11-07 21:54:42

by Mike Anderson

Subject: Re: [BUG] 2.5.46 reloading a block device module doesn't probe devices

Dave Olien [[email protected]] wrote:
>
>
> I just converted the DAC960 driver in 2.5.46 to use the pci_register_driver()
> to discover devices.
>
> I've been experimenting loading the DAC960 driver module, unloading
> the module, and then loading it again. I get curious behavior.
> The load, unload seem to work fine.
>
> The reload also APPEARS to succeed. The insmod command completes
> with a successful status.
>
> But the second loading of the driver never calls the driver's probe routines.
> I added printfs, and determined that the module_init() function
> DOES get called the second time. But a printf in the probe() routine
> doesn't show up the second time.
>
> Has there been a patch recently for this problem?

Dave,

In scsi we where having issues of insmod / rmmod not cleaning up sysfs
correctly resulting in secondary insmod problems and shutdown oops.

The problems where:
- If no probe routine than a device is attach, but error is returned
allowing more than one attach.
- Parent always +2 on device_register.
- driver_unregister lacks full cleanup.

The patch below contains a few workarounds patmans and I have been using
to get our scsi insmod / rmmod plus shutdown working. Mochel has been
offline for a few days so I have not got any feedback from him on the
correctness. YMMV.

-andmike
--
Michael Anderson
[email protected]

bus.c | 4 +++-
core.c | 2 --
driver.c | 2 ++
3 files changed, 5 insertions(+), 3 deletions(-)
------

--- 1.22/drivers/base/bus.c Thu Oct 31 08:20:23 2002
+++ edited/drivers/base/bus.c Wed Nov 6 22:06:39 2002
@@ -209,8 +209,10 @@
attach(dev);
else
dev->driver = NULL;
- } else
+ } else {
attach(dev);
+ error = 0;
+ }
}
return error;
}
--- 1.50/drivers/base/core.c Thu Oct 31 08:20:23 2002
+++ edited/drivers/base/core.c Wed Nov 6 13:07:47 2002
@@ -173,8 +173,6 @@
return -EINVAL;

device_initialize(dev);
- if (dev->parent)
- get_device(dev->parent);
error = device_add(dev);
if (error && dev->parent)
put_device(dev->parent);
--- 1.14/drivers/base/driver.c Wed Oct 30 16:35:48 2002
+++ edited/drivers/base/driver.c Wed Nov 6 21:41:48 2002
@@ -127,6 +127,8 @@
drv->present = 0;
spin_unlock(&device_lock);
pr_debug("driver %s:%s: unregistering\n",drv->bus->name,drv->name);
+ bus_remove_driver(drv);
+ kobject_unregister(&drv->kobj);
put_driver(drv);
}