On Llu, 2003-06-23 at 19:11, Tuukka Toivonen wrote:
> Let's suppose user calls open() on my device. The first line of open()
> call could lock semaphores, increase reference counts, whatever, but all is
> useless if usb_disconnect() happens _after_ open() system call but _before_
> the first line of the open()-function has been executed.
Firstly you are smart enough to notice the open thing. Thats a good sign
not a bad one 8)
For module unload you can set
owner: THIS_MODULE
and the kernel bumps your module count
> So I can not think of any way to fix it... usb_disconnect() can not trust
> that the open() call has had time to do any locking, or anything. How can
> it know whether resources may be freed?
Because the open handler can do
take a lock shared with your disconnect
check the "is it still here" static info
realise it isnt
abort
[or realise it is and ref count it so that the unplug wont free it just
yet]
-------------------------------------------------------
This SF.Net email is sponsored by: INetU
Attention Web Developers & Consultants: Become An INetU Hosting Partner.
Refer Dedicated Servers. We Manage Them. You Get 10% Monthly Commission!
INetU Dedicated Managed Hosting http://www.inetu.net/partner/index.php
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
