booting msm on kaiser

Hi Kevin,

as you may have seen in #htc-linux I have some problems booting pH5's/my version of the android msm tree on kaiser. We added up a machine type for kaiser removing the unnecessary hardware for halibut evaluation board and added vsfb support. Plus some other changes (I think earlier init of vsfb and some early debugging things). See the attached kaiser-msm.diff for detail.

When I hit the run button in haret (haret-20071122.exe) to start the bootscript (attached as default.txt) a loading bar appears. This reaches its end and then a Go Go Go.. appears. The screen of kaiser freezes at that state and the device does nothing at all and stays at the same screen with the haret loading bar( waited 10 minutes ). I attached the haretlog.txt (produced by creating empty earlharetlog.txt)

Now pH5 told me usually at least the screen should be painted black and some init lines should appear. So he cooked up the patches I mentioned above with early init of vsfb and debugging stuff. Also he let me input some commands from haretconsole, to see if haret has access to the framebuffer. Obviously it does:

on "VFH 0x51b00000 76800 0xffff" I get complete white screen on kaiser
on "print "%x" P2V(0x10010000)" Kaiser repeats "2a0000"

pH5 took from this that it is no problem with the framebuffer access and had no further ideas, telling me to contact you about the problem.
"19:01.31 pH5 after "go go go" it should disable interrupts in take_control() and the next thing that should happen is a fb_clear, which would turn the screen black.
19:02.38 pH5 dcordes_: maybe disabling interrupts doesn't work as easily on kaiser. or maybe haret just doesn't have the correct frame buffer information.
19:03.01 pH5 that is Kevin2's area of expertise"
(I attached the reffering log entriese from that day)

Maybe you have any ideas what might be wrong.
Thank you very much

Lukas

-- 
Psssst! Schon vom neuen GMX MultiMessenger gehört?
Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger