[Trac #1147] USB HardDisk is not recognized in OFW , but recognized after booting to system.

[Zarro Boogs per Child]

#1147: USB HardDisk is not recognized in OFW , but recognized after booting to
system.
---------------------------------+------------------------------------------
 Reporter:  chiaying.lin         |        Owner:  rsmith 
     Type:  defect               |       Status:  new    
 Priority:  normal               |    Milestone:  BTest-3
Component:  ofw - open firmware  |   Resolution:         
 Keywords:                       |  
---------------------------------+------------------------------------------
Comment (by wmb at firmworks.com):

 I have a similar external USB hard disk enclosure that works just fine
 under OFW.

 It is of a different make and model - a no-name one that I bought at
 Fry's.  The hard disk in it is Hitachi IC25N040ATMR04-0.  The interface
 chip inside is Cypress CY7C68300B-56PVXC.

 The web page for the PCI PL-25UA device mentioned above says:
 This product cannot be started via the installed hard disk.

 I'm not sure what that means, but one possible interpretation is that
 using it as a boot device doesn't work very well, perhaps because it takes
 a long time to start.  But that is just a wild conjecture...

 On the web page for the Zynet HD-D4 device, there are links to a download
 area for product manuals.  The table of manuals for the HD-D4 series has
 variants labeled Cypress, Genesys, and Ali.  That suggests to me that
 there are three possible interface chips that might be in this product, so
 there is no guarantee that if you order one, you will get the same thing
 that this bug report cites.

 Instead of ordering one of the devices, it might be better just to try the
 most likely firmware fix, which is to increase the delay between applying
 power and enumerating the USB bus.  That can be done with a trivial mod to
 the firmware.

 Another test would be to type this at the ok prompt:

 ok probe-usb2

 If that works, the delay hypothesis is almost certainly correct.

 I would point out that these "parasitically powered" USB devices are dodgy
 in general.  They require two USB connections to the host, one for data
 and the other to supply extra power, because they need more power than one
 standard USB port can deliver.  (Alternatively, they can be used with an
 external power adapter.)  This "dual USB cable" situation violates the
 spirit of the USB spec, in which ports are supposed to be logically
 independent.  In particular, turning off the power to port A is not
 supposed to affect port B in any way.

-- 
Ticket URL: <http://dev.laptop.org/ticket/1147#comment:6>
One Laptop Per Child <http://laptop.org/>