I started this as a thread on current-users:
http://mail-index.netbsd.org/current-users/2013/11/19/msg023704.html
It turned out the "hanging" is actually more like playing roulette - sometimes
you do get through -> timing?
Stephen pointed out that "apm0 at mainbus0" does compile on NetBSD 6.1.2/i386,
so I got a little further:
apm0 at mainbus0 compiles, but doesn't attach, so running the laptop without
acpi doesn't seem to be an option. (Although boot -2 gives a working laptop
which can't power-off nor check its battery state.)
I also enabled PCI_*_FIXUP just in case.
I popped "vmstat -w 1" into /etc/rc.local based on the "interrupt storm"
theory. vmstat was printing much faster than every 1s (scrolling off
screen). This matches the -current/i386 observations of:
- lots of hard disk timeouts
- external keyboard as a ridiculously high repeat rate
- logging in via ssh has to be rapid to beat the login timeout
- laptop keyboard doesn't work
kern.timecounter.choice = TSC(q=-100, f=32248970 Hz) clockinterrupt(q=0, f=100
Hz) ichlpcib0(q=1000, f=3579545 Hz) ACPI-Fast(q=1000, f=3579545 Hz)
lapic(q=-100, f=437599 Hz) i8254(q=100, f=1193182 Hz) dummy(q=-1000000,
f=1000000 Hz)
kern.timecounter.hardware = ACPI-Fast
kern.timecounter.timestepwarnings = 0
I tried
kern.timecounter.hardware: ACPI-Fast -> ichlpcib0
(and lapic, TSC)
but the vmstat -w 1 test still runs too fast
kern.timecounter.hardware = ichlpcib0
kern.clockrate: tick = 10000, tickadj = 40, hz = 100, profhz = 100, stathz = 100
Any thoughts on what to try next?