Subject: Re: 1.6 miniroot won't boot
To: None <bouyer@antioche.lip6.fr, grant@grunta.com>
From: None <eeh@netbsd.org>
List: port-sparc64
Date: 06/07/2002 00:34:02
| On Thu, Jun 06, 2002 at 12:20:21PM +0200, Manuel Bouyer wrote:
|
| > Rebooting with command: boot disk1:b                                  
| > Boot device: /sbus/espdma@e,8400000/esp@e,8800000/sd@1,0:b  File and args: 
| > NetBSD IEEE 1275 Bootblock
| > .
| > and it hangs there.
| > 
| > Has anyone sucessfully used the 1.6 miniroot ?

I've stopped using miniroots for a while now.  Usually netboot an
INSTALL kernel or use an ISO image.

|
| Manuel,
|
| I haven't yet built 1.6 on sparc64, but I am seeing the same here with
| -current from yesterday's source. Does netbooting work?
|
| Sun Enterprise 220R (2 X UltraSPARC-II 450MHz), No Keyboard
| OpenBoot 3.29, 1024 MB memory installed, Serial #15536159.
| Ethernet address 8:0:20:ed:10:1f, Host ID: 80ed101f.
|
|
|
| Rebooting with command: boot disk:b
| Boot device: /pci@1f,4000/scsi@3/disk@0,0:b  File and args:
| NetBSD IEEE 1275 Bootblock
| .

Hm.  Sounds like a booblock issue.  I tried the miniroot from
the latest 1.6_BETA build and got a bit farther:


Rebooting with command: boot disk:b                                   
Boot device: /pci@1f,0/ide@d/disk@0,0:b  File and args: 
NetBSD IEEE 1275 Bootblock
..>> NetBSD/sparc64 OpenFirmware Boot, Revision 1.5
>> (autobuild@tgm.daemon.org, Tue Jun  4 14:53:32 UTC 2002)
devopen: getdisklabel sez no disk label
loadfile: reading header
elf64_exec: Booting /pci@1f,0/ide@d/disk@0,0:b/netbsd
4561384@0x1000000+7316640@0x1800000+1071968@0x1efa4a0 
symbols @ 0xfef82300 74+299064+157490 start=0x1000000
chain: calling OF_chain(800000, e478, 1000000, fff63a80, 18)
[ using 457408 bytes of netbsd ELF symbol table ]
pmap_bootstrap: could not claim physical dseg extension at ec00000 size 400000
Kernel size exceeds 4MB
Setting DTLB entry 00000000 01000000 data e0000000 0f800074
Setting DTLB entry 00000000 01800000 data e0000000 0d400076
Setting ITLB entry 00000000 01000000 data e0000000 0f800074
Setting CPUINFO mappings...
Setting TSB pointer 00000000 01f86000
consinit()
setting up stdin
chosen = f002d8a4, stdin @ 0x1818b58
stdin instance = fff5b988
stdin node = f007285c
setting up stdout
stdout instance = fff5bde0
stdout package = f007285c
buffer @ 0x2009d90
data fault: pc=f0008ff0 addr=665ffffe000
kernel trap 30: data access exception
1 tt=1ff tstate=0 tpc=0x0 tnpc=0x0
Stopped in pid 0 () at  0xf0008ff0:     ld              [%g4 + 0x0], %l0
db> mach prom
Type  'go' to resume
ok 


Can you try the following:

ok setenv fcode-debug? true
ok boot disk:b -V

That should give us some useful debug info.

Eduardo