Re: Dom0: machine hang with cpuctl

[Christoph Egger <Christoph_Egger%gmx.de@localhost>]

Christoph Egger wrote:
> Hi!
>
>
> In a amd64 Dom0 (4.99.64), "cpuctl identify 0" works fine on VGA.
> In a ssh session on that machine, the same command makes the machine 
> hang - the command itself does not run and there's no reaction on the 
> shell.
>
>
> With a GENERIC kernel, "cpuctl identify 0" works both on VGA and
> (ssh) remote.
>
>
> When the machine hangs, I can ping the machine, I can break into ddb
> on VGA. Things I type on VGA don't appear on the display. But in ddb,
> the keyboard works normally.
>
> A backtrace in ddb is always the same:
>
> wskbd_translate() at netbsd:wskbd_translate+0xb1e
> wskbd_input() at netbsd:wskbd_input+0xc6
> pckbd_input() at netbsd:pckbd_input+0x54
> pckbcintr() at netbsd:pckbcintr+0x81
> Xresume_xenev6() at netbsd:Xresume_xenevt+0x55
> --- interrupt ---
> Xspllower() at netbsd:Xspllower+0xe
> idle_loop() at netbsd:idle_loop+0x86

I forgot to mention, when I type reboot in ddb then I get this panic:

assert_sleepable() at netbsd:assert_sleepable+0x75
_ftrans_start() at netbsd:_ftrans_start+0x29
ffs_sync() at netbsd:ffs_sync+0x6c
sys_sync() at netbsd:sys_sync+0xeb
vfs_shutdown() at netbsd:vfs_shutdown+0x4e
cpu_reboot() at netbsd:cpu_reboot+0xdc
db_reboot_cmd() at netbsd:db_reboot_cmd+0x47
db_command() at netbsd:db_command+0xb8
db_command_loop() at netbsd:db_command_loop+0xe9
db_trap() at netbsd:db_trap+0xdd
kdb_trap() at netbsd:kdb_trap+0xc2
trap() at netbsd:trap+0x3b5

> Can anyone reproduce this?

Christoph