Re: amd64 xen3_dom0 failing to boot - stalls (generic boots fine)

[Sarton O'Brien <sarton%crebrokers.com@localhost>]

Sarton O'Brien wrote:
> Manuel, if you are reading this (sorry for the cc), I generally don't 
> have many problems booting xen (outside of the bootdev/rootdev 
> changes) or in the ability to provide data for debugging but I find 
> this really odd. Not getting a panic and not being able to break into 
> ddb makes it even stranger. Have you any ideas?

I was wrong, ddb works from serial:

++++fatal breakpoint trap in supervisor mode
trap type 1 code 0 rip ffffffff8020312d cs e030 rflags 202 cr2  
7f7ffd971160 cpl
6 rsp ffffa000126689f8
Stopped in pid 0.2 (system) at  netbsd:breakpoint+0x5:  leave
breakpoint() at netbsd:breakpoint+0x5
xencons_tty_input() at netbsd:xencons_tty_input+0xc7
xencons_intr() at netbsd:xencons_intr+0x5f
evtchn_do_event() at netbsd:evtchn_do_event+0xaa
call_evtchn_do_event() at netbsd:call_evtchn_do_event+0xd
hypervisor_callback() at netbsd:hypervisor_callback+0xa3
idle_loop() at netbsd:idle_loop+0xcb
ds          0x8072
es          0
fs          0x6fe0
gs          0xcba0
rdi         0xffffa00012851000
rsi         0xffffffff80b6f0d0  rbuf.11636
rbp         0xffffa000126689f8
rbx         0xffffffff80b6f0d0  rbuf.11636
rdx         0x2b
rcx         0xfe2b
rax         0x1
r8          0
r9          0x7d
r10         0xffffffff80b8dde0  sleeptab+0x6e0
r11         0x246
r12         0xffffa00012851800
r13         0
r14         0x1
r15         0xffffa00012851000
rip         0xffffffff8020312d  breakpoint+0x5
cs          0xe030
rflags      0x202
rsp         0xffffa000126689f8
ss          0xe02b
netbsd:breakpoint+0x5:  leave
db> show events
No such command: events
db> show
callout         arptab          malloc          ncache          uvmexp
pages           breaks          map             object          vnode
procs           buf             module          page            vmem
pools           event           mount           pool            vmems
aio_jobs        files           mqueue          registers       watches
all             lock            mbuf            sched_qs
db> show event
evcnt type 0: bus_dma loads = 1358
evcnt type 0: bus_dma nbouncebufs = 42
evcnt type 0: vmcmd kills = 18
evcnt type 0: vmcmd calls = 149
evcnt type 0: softint net/0 = 27
evcnt type 0: softint bio/0 = 1319
evcnt type 0: softint clk/0 = 2483
evcnt type 0: softint ser/0 = 1
evcnt type 0: crosscall unicast = 2
evcnt type 0: namecache entries collected = 170
evcnt type 0: namecache under scan target = 47
evcnt type 1: vcpu0 xencons = 5
evcnt type 1: vcpu0 ioapic0 pin 21 = 28
evcnt type 1: vcpu0 ioapic0 = 17
evcnt type 1: vcpu0 ioapic0 pin 20 = 1336
evcnt type 1: vcpu0 clock = 5011
db>

pin 20 happens to be piixide1 -> atabus1 -> wd0 ... though I have no 
idea what kind of number is 'high'.

ioapic0: pin 20 attached to pci0 device 31 INT_C (type 0x0<type=0x0> 
flags 0xf<p
ol=0x3=Act Lo,trig=0x3=Level>)
x0> 0x3000000<target=0x3>
piixide1: using ioapic0 pin 20, event channel 10 for native-PCI interrupt
atabus1 at piixide1 channel 0

Mean anything? :)

Sarton