Previously when I was testing the jetway target, I found that the IDE devices were always enabled, as the mainboard's Config.lb seems to suggest would be the case.
Today, after running an 'svn up' and reverting the src/southbridge/via/vt8237r to before the breakage in rev 3567, I found that the IDE devices aren't ever enabled, no matter what I set for the values of ide0_enable and ide1_enable in the mainboard's Config.lb. And of course, the board won't boot.
I looked in the built target's image/static.c and found the struct southbridge_via_vt8237r_config with ide0_enable and ide1_enable both set to 1 as they should be.
But back in src/southbridge/via/vt8237r/vt8237r_ide.c (around line 41), sb->ide0_enable and sb->ide1_enable are both 0.
Forcing them to 1 there causes the devices to be enabled, and then the system will boot.
But the question is, what's going on there? Why are these values set to 0? Is ide_init accessing chip_info for the wrong device? Does anyone have any ideas?
-Alex Mauer "hawke"