On Sun, Oct 12, 2008 at 2:15 PM, ron minnich rminnich@gmail.com wrote:
On Sun, Oct 12, 2008 at 11:04 AM, Corey Osgood corey.osgood@gmail.com wrote:
I'm attaching a minicom log file. From the looks of things, it also
looks
like I'm losing all output from stage0/1. The reason it boots twice is a GPIO that automatically resets the system if it isn't disabled, and it's
not
disabled yet.
the garbage may be your stage0/1 output. We've had trouble in the past with cheap uarts glitching badly when things like baud rate change. I actually don't think you need to worry about this garbage. I think you're doing very well indeed.
printk works, so your stack is working. So car is probably working. You ran initram. CAR is definitely working. You returned from initram. This is great.
Your biggest problem may be that you are not really getting initram right, which is why there is no stage2 output.
Oh, yeah, you are in trouble: LAR: Compression algorithm #1 (lzma) used Decoding error = 1
ram is no good. You need a test at the end of initram. But CAR is definitely solid I think. You need more debug in initram.
You're close.
I am going to be presenting CN700 at SC 08 I hope thanks to your fine work!
Ram init is currently a noop, I'm working on getting CS5536's SMBus functions (which are very clean) working in a more generic manner, and on vt8237. I'll get rid of any stage1 console output so the garbage will go away ;)
Thanks, Corey