On Wed, 9 Jun 2004, David Brownell wrote:
> Alan Stern wrote:
> > Greg:
> >
> > After my last few changesets there were a few small items that needed to
> > be tidied up.
>
> On top of the "one big tidy up" you've been doing to that driver? :)
>
> It's needed further attention for quite a while, and I'm glad you're
> doing this. I take it these all came up as you were diving into the
> "make reset work properly" problems?
Yes. Resets can't work right without proper locking (which is still to
come). Proper locking can't be implemented if the logical control flow
isn't adequate. And on top of all that, there were all these little
things I kept noticing about the hub driver over the last year or more...
This looked like the right time to finally get it all straightened out.
I'm going to be busy for a few days, so the contributions will slow down
for a while. Up on deck are:
Remove the khubd private semaphore and simply use hdev->serialize
instead.
Store bitmasks for hub ports with pending changes and ports
with logical connect changes pending.
Other miscellaneous fixups for the hub driver.
Replace direct access to udev->serialize with new routines
usb_lock_device() and usb_unlock_device(). This is needed to
make the locking work right when drivers are loaded and
unloaded.
Figure out some way to make sense of resets of root hubs...?
Once that's all done we should see resets, locking, and everything else
all working like magic! :-)
Alan Stern
-------------------------------------------------------
This SF.Net email is sponsored by: GNOME Foundation
Hackers Unite! GUADEC: The world's #1 Open Source Desktop Event.
GNOME Users and Developers European Conference, 28-30th June in Norway
http://2004/guadec.org
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
