On Sun, 9 Oct 2005 17:22:39 -0400 (EDT), Alan Stern <[EMAIL PROTECTED]> wrote:
> But when would you call kthread_start?

> What happens if the thread calls request_module too soon?

I did not see an oops or deadlock, but the storage module fails to
resolve symbols which it needs from libusual, such as
usb_usual_set_present. They are only available once the module
has finished loading. So, there is a (very small, perhaps) possiblity
that it would happen even now. I even used to have a little delay
right after daemonize().

> BTW, it's a little surprising that the hotplug support programs can work
> okay with modules using external references for their usb_id tables.  Do
> they interpret that the same as having an empty table (or no table)?

The depmod does not know about any references from struct usb_driver.
It generates the lists using the hook in MODULE_DEVICE_TABLE. If libusual
is enabled, storage drivers do not have MODULE_DEVICE_TABLE, and thus,
they do not have any entries in modules.alias. If libusual is disabled,
the reference is not external and the code is exactly as it was before.

I have verified that modules.alias is generated correctly in all cases.

-- Pete


-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
linux-usb-devel@lists.sourceforge.net
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel

Reply via email to