Aaron Young wrote:
>
>
>
> I guess I'm good with it. Couple of nits:
>
> 1. The routine kdb_uhci_keyboard_urb() is really not uhci specific
> and so could be renamed to simply kdb_keyboard_urb(). Or
> kdb_get_urb() or kdb_get_urb_index() perhaps to be more descriptive.
> kdb_uhci_keyboard_urb() sounds like it should return a BOOLEAN.
>
> 2. I don't understand this comment and code:
>
>> + /* We don't know exactly if UHCI KDB QH was allocated,
>> + * so if kdb_usb_keyboard_detach(hid_inurb) failed
>> + * kdb_usb_keyboard_detach(kdb_urb) should be called.
>> + */
>> + kdb_usb_keyboard_detach(hid_inurb);
>> + kdb_usb_keyboard_detach(kdb_urb);
>
> I might not be following the diffs properly, but it seems like you
> should only be calling kdb_usb_keyboard_detach() for urbs which have
> already been attached via kdb_usb_keyboard_attach(). Is this the case?
>
>
> Once Jay gets all these patches integrated, I'll have to go
> back and retest OHCI and EHCI (on ia64) to make sure things still
> work okay. Jay, at this point, since there are several patches now
> built on top of the EHCI patch(es), I think it would be best to take
> the EHCI patch(es) and the UHCI patches. We can fix any bugs (if any) in
> subsequent patches. IF we were to redo the EHCI patches now (because of
> bugs or whatever), it could render these dependent patches as needing to
> be redone too. JMO...
I uploaded the EHCI patch and UHCI patch to the download area at
oss.sgi.com for those who want to explore USB ehci and uhci support
in kdb.
USB_EHCI_support-v2.bz2
USB_UHCI_support-v2.bz2
These two files can be applied on top of the kdb 2.6.24-rc4 patchset.
Note that the ehci patch should be applied first.
Caveats:
- The ehci patch has not been tested on x86_64 nor i386 platforms.
- The uhci patch will have compilation and link errors on ia64.
Please feedback, success or failure, to this list.
Thanks,
- jay
>
> -Aaron
>
---------------------------
Use http://oss.sgi.com/ecartis to modify your settings or to unsubscribe.
|