Re: Probable module bug in linux-2.6.5-1.358

[Richard B. Johnson]

On Thu, 7 Oct 2004, Stephen Hemminger wrote:

> Still haven't full source so this is still guess work.
> But assuming it is a character device, did you forget to add an owner
> field to the file ops structure?
>
> static struct file_operations xxx_fops = {
> 	.owner	= THIS_MODULE,
> 	.read	= my_read,
> ...
>
> The owner field is used by the character device layer to maintain module
> ref counts in 2.6.

No. The owner field is properly filled in and I did provide
the complete source code and build files for anybody who didn't
delete the email. Scanner.tar.gz was attached.

Further, unregister_chrdev() is called. Its return value is
checked. Everything is fine. I can manually remove and
reinstall the module multiple times and the resources are
always released and re-acquired.

The problem is that if you install the module and then
remove it, if you attempt to open the device-file, the
kernel will crash because it calls where the properly-
removed module used to be. It isn't there anymore.


Cheers,
Dick Johnson
Penguin : Linux version 2.6.5-1.358-noreg on an i686 machine (5537.79 BogoMips).
            Note 96.31% of all statistics are fiction.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/