Mark Brinicombe wrote:

> On Fri, 28 May 1999, Dave wrote:
>
> > Any idea when I will be able to 'roll my own' from ksrc?
> Should be able to do it now. The change was picked up in last nights sup
> scan so sup will pick it up and it should be in the source tars on
> ftp.netbsd.org.

Thanks, I'll do a 'sup' tonight.

There is a slightly strange behaviour which is puzzling.

When I first booted the kernel that you posted, it booted OK as /netbsd, so
I installed the 1.4 base, comp, text and misc sets and edited all the files
in my /etc, /var et al that had changed and rebooted.

All the boot messages looked OK until:

icside0 at podulebus0 [ podule0 ]: ARCIN V5
panic: no map for fault adress

Stopped in  at _Debugger+0x10: ldmb  r11, {r11, r13, r15}

db>trace
_Debugger(_Debugger+0x10)
_panic(_panic+0x14)
_data_abort_handler(_data_abort_handler+0x10)
___wdccommand_intro(___wdccommand_intro+0x10)
___wdccommand_start(___wdccommand_start+0x10)
_wdcstart(_wdcstart+0x10)
_wdc_exec_xfer(_wdc_exec_xfer+0x10)
_wdc_exec_command(_wdc_exec_command+0x10)
_ata_get_params(_ata_get_params+0x10)
_icside_attach(_icside_attach+0x10)
_config_attach(_config_attach+0x10)
_config_found_sm(_config_found_sm+0x10)
_podulebusattach(_podulebusattach+0x10)
_config_attach(_config_attach+0x10)
_config_rootfound(_config_rootfound+0x10)
_configure(_configure+0x10)
_main(_main+0x10)
_kernel_text(kernel_text+0x3c)

db>show map
MAP 0xf0144560: [0xe1a0c00d->0xe92dd800]
          #ent=-476053503, sz=-335544581, ref=-369098748, version=-476053504

panic: no map for fault adress

Stopped in  at _Debugger+0x10: ldmb  r11, {r11, r13, r15}

I tried loading the kernel from UnixFS:$.netbsd on wd0a and the result was
much the same, but I think the initial panic was 'polled interrupt not
found'.

I then copied /netbsd to an ADFS partition and discovered that, when loaded
from there, all goes well! However any attempt to boot with a native kernel
fails as above.

The only thing that does not seem to work as expected is ps, which has a
proc size mismatch, but does list the processes. I suppose that's because
the kernel is out of sync with the base set. I'll sup and compile a new one.

regards,
Dave
--
Make it idiot-proof and someone will make a better idiot!

e-mail: webmaster@largesalad.co.uk
net: http://www.largesalad.co.uk/DJMsoft/