Hi,
I am working on linux kernel 2.4.18. OHCI HCD. I have
You mean, "usb-ohci". "ohci-hcd" is a different driver, which is available on 2.4.21pre5 (separate patch) ut not on 2.4.18 ...
the following doubt:
When the device enumeration fails (usb.c: USB device
not accepting new address), usb.c sets the devnum
field ( in the device data structure) to -1 and calls
the dev->bus->op->deallocate(dev); This inturn calls sohci_free_dev() for the OHCI which
checks for the devnum >=0 (usb_dev->devnum >=0 in the
usb-ohci.c)
Yeah, all that "devnum = -1" logic has always looked error prone to me. Not just in the OHCI code.
In the above case the OHCI HCD doesn't clean up the dummy TD for the Control ED (which is in ED_UNLINK state). Doesn't this cause a td leakage????
Maybe. Does "ohci-hcd" do the same?
- Dave
------------------------------------------------------- This SF.net email is sponsored by: The Definitive IT and Networking Event. Be There! NetWorld+Interop Las Vegas 2003 -- Register today! http://ads.sourceforge.net/cgi-bin/redirect.pl?keyn0001en _______________________________________________ [EMAIL PROTECTED] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
