[9fans] plug boot problem

[9fans] plug boot problem

[ron minnich]

I am having a simple problem which I know I've already solved -- but
the machine is elsewhere.

I am building a plug kernel. I've made the following change to plug
config file:


bootdir
	boot.fs boot
	/arm/bin/rc
	/rc/lib/rcmain

The boot.fs just has this:

#!/boot/rc -m /boot/rcmain

echo "IT's a beautiful day in the neighborhood!"
echo "HIHIHIHI"

/boot/rc -m/boot/rcmain -i

When I boot with this, I get:

Plan 9 from Bell Labs

l1 D: 16384 bytes, 4 ways 128 sets 32 bytes/line; write-through only
l1 I: 16384 bytes, 4 ways 128 sets 32 bytes/line; write-back type `reg
7 ops, format C' (016) possible
l2 cache: 256K or 512K: 4 ways, 32-byte lines, write-back, sdram only
cpu0: 1200MHz ARM Marvell 88F6281 A1; arm926ej-s arch v5te rev 2.1 part 131
#F0: kwnand: Samsung 2Gb 536,870,912 bytes pagesize 2048 erasesize
131,072 spares per page 64
#F0: nand addr 0xf9000000 len 536870912 width 1 interleave 0
#l0: 88e1116: 100Mbps port 0xf1072000 irq 11: 005043019de6
#l1: 88e1116: 100Mbps port 0xf1076000 irq 15: 005043019de7
#u/usb/ep1.0: ehci: portpanic: boot process died: sys: bad address in
syscall pc=0x1124
 0XF1050100 irq 19
504M memory: 54M kernel data, 450M user, 1952M swap
*init* 1: suicide: invalid address 0x2/1 in sys call pc=0x1124
*init* 1: suicide: sys: bad address in syscall pc=0x1124ktrace
/kernel/path 0x60806bb8 0x60b6efa8 0x60b6f003 # pc, sp, link

I am pretty sure this is some obvious  config mistake I'm making. I
can make one change to get it to do better:


bootdir
	boot$(CONF).out
	/arm/bin/rc
	/rc/lib/rcmain

and it boots fine. Any suggestions most welcome ...

thanks

ron

Re: [9fans] plug boot problem

[ron minnich]

On Sun, Dec 12, 2010 at 4:37 PM, ron minnich <rminnich@gmail.com> wrote:
> I am having a simple problem which I know I've already solved -- but
> the machine is elsewhere.


It's turning out not to be simple at all. The *init* is the init
process. It is the hand-crafted kernel process and it sets up an
initial name space and calls exec with a boot file name and argv. The
exec file name is /boot/boot; the argv is supplied by the kernel. It's
bogus; arg[1] is 2.

I'll let you now. No idea why this is an issue when I'm not using
standard boot (yet).

ron