PowerBook 12" hangs right after attaching ams0

[Nathan Whitehorn]

Ed Schouten wrote:
> Hello Nathan,
> 
> * Nathan Whitehorn <nwhitehorn at freebsd.org> wrote:
>> Drat. Could you try the patch at
>> http://people.freebsd.org/~nwhitehorn/adb-delayed-attach.diff?
> 
> After applying the patch, it still gets stuck during boot.

Could you try the latest HEAD? I poked around at my machines this 
morning to try to make them hang, succeeded, and then fixed some 
problems that make mouse probes hang after the keyboard is probed 
successfully.

>> Unfortunately, ams0's attach doesn't have any debug information in it.
>> It is stuck somewhere between lines 171 and 183 of adb_mouse.c, if you
>> want to fill it with printf().
> 
> I can confirm it indeed gets stuck in adb_mouse_attach(), namely during
> its call to adb_set_device_handler(). I've just added a `return (ENXIO)'
> to adb_mouse_probe(), so I can at least boot my PowerBook for now. The
> keyboard support rocks. Thanks! :-)

That's about what I expected. The change I added make it retry commands 
like that if they get stuck, inside of looping forever on a reply. Glad 
your keyboard works, though!

> There are some random things I still have to look into:
> 
> - I've dd'd the boot loader to /dev/ad0s2, but for some reason my Mac
>   still refuses to boot from it.

What is the partition type (gpart show) for ad0s2? I think it might need 
to be set to !Apple_Bootstrap. The partition size may also need to be 
exactly 800K. You should be able to change both of these things with the 
system up, though you might need to enable foot shooting in geom.

> - The processor still runs at 533 MHz. I think Linux also did this when
>   I tried it 3-4 years ago.

OF sets it to this at boot. I think there is something you can change at 
the firmware prompt to bring it up to full speed, but my memory is foggy.

I recently picked up a G4 iBook, and am going to look at cpufreq once my 
work calms down a little in a month or so. Feel free to beat me to it, 
though :)
-Nathan