Re: [RFC] kvm-s390: userspace snapshot

[Oliver Paukstadt <pstadt@sourcentral.org>]

On Thu, 2008-06-12 at 00:14 +0200, Christian Borntraeger wrote:

> Ok, I got an idea.
> Does that patch fix the handle_should_not_happen PANIC?
> 
Patch does not fit, because my code contains
 vcpu->arch.sie_block->gmsor = 0x000000000000;
so I changed this before I applied the patch.
The console patch you mentioned was applied too.

Now I am able to get the kernel running a little further:
[...]
PID hash table entries: 256 (order: 8, 2048 bytes)
console [hvc0] enabled
sclp vt220 tty driver: could not register vt220 - sclp_register returned
-5
list_del corruption. prev->next should be 00000000003d72a8, but was
0000000000000000
------------[ cut here ]------------
kernel BUG at lib/list_debug.c:67!
illegal operation: 0001 [#1] SMP 
Modules linked in:
CPU: 0 Not tainted 2.6.26-rc5-guest-20080609-01433-gdf4245d-dirty #2
Process swapper (pid: 0, task: 00000000003ada00, ksp: 00000000003e8000)
Krnl PSW : 0400000180000000 0000000000198c64 (list_del+0x50/0xb8)
           R:0 T:1 IO:0 EX:0 Key:0 M:0 W:0 P:0 AS:0 CC:0 PM:0 EA:3
Krnl GPRS: 0000000000000479 00000000003b2c28 0000000000000058
0000000000000001
           0000000000041246 0000000000000000 00000000003e8594
0000000000414000
           00000000003a2000 0000000000005000 00000000003d72a8
0000000000454418
           00000000003d72a8 00000000002bd820 0000000000198c60
00000000003e7dd8
Krnl Code: 0000000000198c54: e34040000004       lg      %r4,0(%r4)
           0000000000198c5a: c0e5fff542cf       brasl   %r14,411f8
           0000000000198c60: a7f40001           brc     15,198c62
          >0000000000198c64: e310c0000004       lg      %r1,0(%r12)
           0000000000198c6a: b904003c           lgr     %r3,%r12
           0000000000198c6e: c020000be1b9       larl    %r2,314fe0
           0000000000198c74: e3c010080020       cg      %r12,8(%r1)
           0000000000198c7a: a784000a           brc     8,198c8e
Call Trace:
([<0000000000198c60>] list_del+0x4c/0xb8)
 [<00000000001e9a72>] sclp_unregister+0x3a/0x5c
 [<0000000000401410>] __sclp_vt220_cleanup+0x98/0xb4
 [<0000000000401594>] __sclp_vt220_init+0x168/0x17c
 [<00000000004016e8>] sclp_vt220_con_init+0x3c/0x60
 [<00000000003fddd0>] console_init+0x48/0x60
 [<00000000003e8bd0>] start_kernel+0x37c/0x4c4
 [<0000000000012020>] _ehead+0x20/0x80
Last Breaking-Event-Address:
 [<0000000000000000>] 0x0
 <4>---[ end trace 31fd0ba7d8756001 ]---
Kernel panic - not syncing: Attempted to kill the idle task!

Looks like the sclp_vt220 stuff has a problem to unregister an
unitialized nonexistant console.

kuli.log reports some unknown diags and instructions, but I guess this
is the runtime feature detection:

launch_cpu_ipl: starting guest (ipl)
run_cpu: cpu 0: activated, running work...
handle_diag: cpu 0: unknown diagnose 9c at addr 3f136c, sending prog 1
enter_pgmcheck: cpu: 0: sending program check 1
handle_diag: cpu 0: unknown diagnose 260 at addr 3f1048, sending prog 1
enter_pgmcheck: cpu: 0: sending program check 1
handle_priv: cpu 0: unknown privileged instruction b216 at addr
400200180000000, sending prog 1
enter_pgmcheck: cpu: 0: sending program check 1
sclp_service_call: cpu 0: unknown sclp service call 0x780005, sccb
0x452000,addr 0x1e8780
sclp_service_call: cpu 0: unknown sclp service call 0x780005, sccb
0x452000,addr 0x1e8780
handle_waitpsw: cpu 0: entered disabled wait PSW at 1f312

Regards,
Oliver Paukstadt