#11292 NORM 1.75-fi: OFW should report selftest detection failures differently than major errors

[Zarro Boogs per Child]

#11292: OFW should report selftest detection failures differently than major errors
-------------------------------------------+--------------------------------
           Reporter:  greenfeld            |       Owner:  wmb at firmworks.com                
               Type:  enhancement          |      Status:  new                              
           Priority:  normal               |   Milestone:  1.75-firmware                    
          Component:  ofw - open firmware  |     Version:  Development build as of this date
         Resolution:                       |    Keywords:                                   
        Next_action:  never set            |    Verified:  0                                
Deployment_affected:                       |   Blockedby:                                   
           Blocking:                       |  
-------------------------------------------+--------------------------------
Description changed by greenfeld:

Old description:

> Unlike earlier models, the XO-1.75 may or may not have components
> attached to the motherboard, and some early motherboards allow things to
> be attached in different ways.  The keyboard and mouse may be hooked to
> the /ap-sp or /ec-spi interface, and the touchscreen attachment for
> post-1.75 experimentation usually is not present but occasionally is.
>
> When we encounter the PS/2 interface on the wrong port, the keyboard test
> appears in unusable form but disappears after a delay, and the mouse test
> currently fails.  The test-all command currently tries both keyboard
> ports, while "test /keyboard" seems to hit the current one on the first
> try and not move on.
>
> If possible we should show a "not found" message which is different than
> an outright "failure" message for these cases, if not making the tests
> somewhat smarter (i.e. look at both PS/2 interfaces, and only test the
> one that is connected).  Otherwise teachers in the field may be confused
> by false failure reports.
>
> This may be muted somewhat by the fact that production XO-1.75s are
> unlikely to have different attachment points and optional components, and
> if they do we may be able to pick up on them via the SKU or similar.

New description:

 Unlike earlier models, the XO-1.75 may or may not have components attached
 to the motherboard, and some early motherboards allow things to be
 attached in different ways.  The keyboard and mouse may be hooked to the
 /ap-sp or /ec-spi interface, and the touchscreen attachment for post-1.75
 experimentation usually is not present but occasionally is.

 When we encounter the PS/2 interface on the wrong port, the keyboard test
 appears in unusable form but disappears after a delay, and the mouse test
 currently fails.  The test-all command currently tries both keyboard
 ports, while "test /keyboard" seems to hit the current one on the first
 try and not attempt to use the other.

 If possible we should show a "not found" message which is different than
 an outright "failure" message for these cases, if not making the tests
 somewhat smarter (i.e. look at both PS/2 interfaces, and only test the one
 that is connected).  Otherwise teachers in the field may be confused by
 false failure reports.

 This may be muted somewhat by the fact that production XO-1.75s are
 unlikely to have different attachment points and optional components, and
 if they do we may be able to pick up on them via the SKU or similar.

--

-- 
Ticket URL: <http://dev.laptop.org/ticket/11292#comment:1>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system