Contacts within AMD? AMD-756 USB host-controller blacklisted due to erratum #4.

[Miles Lane <miles@megapathdsl.net>]

Running 2.4.2-ac28, I get the following error:

usb-ohci.c: 00:07.4 (Advanced Micro Devices [AMD] AMD-756 [Viper] USB):
blacklisted, erratum #4

David Brownell recently added this check to the usb-ohci driver
since noone has gotten information from AMD for the workaround,
which is rumored to exist, for this bug.

Do any of you have contacts within AMD who might be able to
get an explanation of the workaround to David Brownell?

The bug is that the NDP value sent by the AMD-756 is sometimes
bogus.  The following examples, collected before the chip
was blacklisted, show the failure.  As you can see, the bogus 
value given varies.  Rereading NDP seems to give a valid value.  
I am not really clear why we don't simply read the value twice 
whenever the host-controller is detected to be an AMD-756.

Mar  4 17:20:52 aerie kernel: usb-ohci.c: bogus NDP=128 for OHCI
usb-00:07.4
Mar  4 17:20:52 aerie kernel: usb-ohci.c: rereads as NDP=4

Mar  4 17:50:29 aerie kernel: usb-ohci.c: bogus NDP=245 for OHCI
usb-00:07.4
Mar  4 17:50:29 aerie kernel: usb-ohci.c: rereads as NDP=4

Mar  6 21:11:07 aerie kernel: usb-ohci.c: bogus NDP=210 for OHCI
usb-00:07.4
Mar  6 21:11:07 aerie kernel: usb-ohci.c: rereads as NDP=4

Thanks,
	Miles