Re: 7043-B50 (PowerPC 604e) Re: PegasosII(ofppc) can't boot

[Frank Wille <frank%phoenix.owl.de@localhost>]

Makoto Fujiwara wrote:

> By removing the line of  ofppc/conf/std.ofppc:
> -options        PPC_OEA64_BRIDGE
> +# options      PPC_OEA64_BRIDGE
>
> Things went further as:
>   http://www.ki.nu/~makoto/NetBSD/ofppc/dmesg/INSTALL-ISI-trap
>
> For the line,
>     trap: pid 1.1 (init): user ISI trap @ 0xfffffffc (SRR1=0x1000d032)
> after reverting the changes
>     sys/arch/powerpc/oea/oea_machdep.c -r 1.12 to -r 1.11

I have tested 6.99.23 on my Pegasos2 today, and indeed it looks very
similar. Here it hangs after:

gt0 at mainbus0

After inserting some printf()s here and there, it seems that the system will
hang on the first bus_space_write(). It makes no difference whether I write
to the PCI config register or to a ISA serial port register. There is
something terribly broken...


I tried your approach to remove the PPC_OEA64_BRIDGE define, but the Pegasos
behaved differently. Now it crashed much earlier:

---8<---
NetBSD 6.99.23 (GENERIC) #0: Fri Aug 30 19:54:26 CEST 2013
       
frank%albireo.wendik.de@localhost:/home/frank/netbsd/current/src/objofppc/sys/arch/ofppc/compile/GENERIC
Model: Pegasos2
total memory = 512 MB
avail memory = 481 MB
sysctl_createv: sysctl_create(delay) returned 1
sysctl_createv: sysctl_create(filedelay) returned 1
sysctl_createv: sysctl_create(dirdelay) returned 1
sysctl_createv: sysctl_create(metadelay) returned 1
trap: kernel read DSI trap @ 0xb893c105 by 0x3d9268 (DSISR 0x40000000,
err=14), lr 0x3d924c
Press a key to panic.
panic: trap
Stopped in pid 0.1 (system) at  netbsd:cpu_Debugger+0x10:       lwz     r0,
0x14
(r1)
db> bt
0x006c0830: at db_panic+0xe8
0x006c0850: at vpanic+0x13c
0x006c0880: at panic+0x4c
0x006c08c0: at trap+0x120
0x006c0950: kernel DSI read trap @ 0xb893c105 by pool_cache_put_paddr+0x4c:
srr1
=0x1032
            r1=0x6c0a20 cr=0x24222044 xer=0x20000000 ctr=0 dsisr=0x40000000
0x006c0a20: at 0x6c0cbc
0x006c0a40: at vmem_free+0x50
0x006c0a50: at uvm_km_kmem_free+0x58
0x006c0a70: at kmem_intr_free+0x94
0x006c0a80: at kern_free+0x20
0x006c0a90: at sysctl_create+0xa48
0x006c0bd0: at sysctl_createv+0x2c0
0x006c0d70: at kernfs_modcmd+0x16c
0x006c0dc0: at module_do_builtin+0x1ec
0x006c0e30: at module_init_class+0x208
0x006c0e70: at vfsinit+0x1fc
0x006c0ec0: at main+0x1b0
0x006c0f10: at 0x1000d0
---8<---

sysctl_create() failed? Followed by a crash in vmem_free()? Very strange.

-- 
Frank Wille