On Sun, 2003-08-24 at 22:00, David Brownell wrote: > > > Any chance of splitting out the "remove ohci->sleeping" bits > > > by themselves? If you don't, I'll want to do that. > > > > > > np. I didn't really intent to submit this as a candidate patch, more > > as a way for you to comment on the direction I'm going. > > OK. I'm looking at some stuff there, so far no overlapping. > And I won't worry too much about moving the PMAC #ifdefs up > towards the USB core, instead of down towards PMAC hardware.
I want to clean these by correctly probing the controller type, and setting a feature bit somewhere when it's a keylargo built-in, that would avoid calling the pmac feature code when not necessary, ie on PCI controllers on macs. Also, I need to dig into some of Apple's code ,they have some workarounds for broken controllers (typically bogus queue handling on old Opti's and problems with which ports we can suspend safely on lucents iirc) that may be worth adapting. > Do you know how much the registers have changed after resume? > > I've found the ohci_dump_status() call in the resume path to be > informative. Or were you getting an IRQ even earlier than that? I think it's just the brown paper bag thing in hcd_irq, but I haven't re-tested yet. Ben. ------------------------------------------------------- This SF.net email is sponsored by: VM Ware With VMware you can run multiple operating systems on a single machine. WITHOUT REBOOTING! Mix Linux / Windows / Novell virtual machines at the same time. Free trial click here:http://www.vmware.com/wl/offer/358/0 _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
