[NTLK] Designing an Internal WiFi Board for MP2x00

[Matthias Melcher]

> On Apr 1, 2015, at 2:04 AM, Jake Bordens <jake at allaboutjake.com> wrote:
> 
> 
>> I have a feeling that there is something slightly wrong on the MessagePad hardware side. I have seen the domino modem line and I wonder if even Apple knew that something is not working. Was the internal modem ever released? If not, maybe that's exactly why it could not be released?
> 
> I hadn't seen that video.  Funny and sad at the same time.
> 
> I was excited today to rediscover a constant in the DDK headers:
> 
> #define kHWLocBuiltInExtra			'tblt'
> 
> 
> I say "rediscovered" because Simon mentions it in a post in 2010.  Sadly, though it doesn't do what I want it to.  Based on some scope probing, this is the hardware location of the 
> Serial2.  I guess that's useful information to know.  Serial 2 is exposed on the internal connector.
> 
> I don't know why, but I have it in my head that Serial Port 2 is used for the keyboard.  I don't know why, so <citation needed>
> 
> Its one more piece of info about the internal serial slot, but doesn't get me closer to keeping the SerialPortSel3 high on open.
> 
> My latest thinking is that the SerPortSel pins may be configured in VoyagerPlatform::PowerOnSubsystem().  This function does call SerialPort0LineDriverConfig and SerialPort3LineDriverConfig.  I haven't reverse engineered the entire function though to see what it is actually doing.  Nor can I find anyhwere that it is actually called... But my disassembly is pretty incomplete.

Look for TPlatformDriver::PowerOnSubsystem(unsigned long). This is a virtual function. When loading the REx, it gets overridden by VoyagerPlatform::PowerOnSubsystem(). It's called by  0x00192534: IOPowerOn(unsigned long), which in turn is called by 0x001D6E58: TSerialChipVoyager::PowerOn(void) and 0x0026C898: TVoyagerPlatform::PowerOnSystem(void) and a few others.

It's time to put the disassembled ROM online somewhere, or at least the comments we write, combined with a disassembler (mine is quite complete by now and proved that the resulting assembly file can be reassembled into an identical ROM).

I will start a new thread.