Well, now that the list is back up...
Randy rejected a patch of mine because, while it worked fine on OHCI, it
locks up a system using UHCI controllers (either driver). This happens
when my driver is using the interrupt endpoint of a device.
While I can't reproduce _exactly_ what Randy reported, I can come pretty
close. And what I've found it this: I load the driver once and it uses
usb_request_irq(). Everything seems fine. I rmmod the driver and it uses
usb_release_irq(). Everything still seems fine. I load the driver again,
and while the debugging messages seem okay, I never get the interrupt
callback. If I unplug the device and then re-attach it, it works fine
again (of course, that calls usb_release_irq() and usb_request_irq()
again).
Note that this works fine on OHCI. Does anyone have any idea what's wrong
on UHCI? I really need this cleared up, because the patch that Randy
rejected is very important.
Matt Dharm
--
Matthew Dharm Home: [EMAIL PROTECTED]
Engineer, Qualcomm, Inc. Work: [EMAIL PROTECTED]
What, are you one of those Microsoft-bashing Linux freaks?
-- Customer to Greg
User Friendly, 2/10/1999
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
