[meta-freescale] I.MX6 SoloX M4 Problem

[Nathan Barrett]

Hello Everyone,

I believe I may have run into a pin multiplexing problem with the I.MX6SX SDB, 3.10, and imx6sx-sdb-m4.dtb

I have managed to boot Linux on the A9 and a demo application on the M4.
 
If I boot the M4 application via 'run m4boot' in u-boot before Linux is booted, the input/output on the M4 side seems fine at 115200 baud on the FTDI usb-serial 
debug port.

***** MCC Virtual TTY EXAMPLE *****
Please wait :
    1) A9 peer is ready
Then press "S" to start the demo
********************************
Press "S" to start the demo :
 
However -- As soon as the A9 core starts running Linux, any input/output on the M4 side looks like gibberish.  It's as if the baud rate is no longer set to 115200.
 
For example if I press 'S' before issuing the boot command in u-boot (to start Linux) the response is "Responder task started, MCC version is 002.000" from the 
M4
 
If I press 'S" after Linux is booting, I get no response and the S does not appear as an 'S' in Teraterm.  Again, it's as if the S isn't making it through because the 
baud rate is mangled.


To clarify this issue a little more.  I wrote a program on the M4 which prints "Hello World" every 100 ms.  I then set u-boot to automatically start the M4 and the 
A9.  The M4 starts first.
 
What I see when turning the board on is:
 
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
Hello World
ÈùÙloùÙ
ÈùÙÈùÙÈùÙÈùÙÈùÙÈùÙ
   ÈùÙÈùÙÈùÙÈùÙÈùÙÈùÙ
ÈùÙ                  ÈùÙÈùÙÈùÙÈùÙÈùÙÈùÙ
ÈùÙÈùÙÈùÙÈùÙÈùÙÈùÙ
   ÈùÙÈùÙÈùÙÈùÙÈùÙÈùÙ
ÈùÙÈùÙÈùÙÈùÙ         ÈùÙ
   ÈùÙÈùÙÈùÙÈùÙÈùÙÈùÙ
ÈùÙ                  ÈùÙÈùÙÈùÙÈùÙÈùÙ
ÈùÙÈùÙÈùÙÈùÙÈùÙÈùÙ
 
The gibberish repeats indefinitely.  Changing baud rates does not fix the issue.
 
Does anyone have any idea what's going on here?
 
Thanks