Package: qemu-system-x86
Version: 2.1+dfsg-2
Severity: important
acpi is broken
I'm using the kvm command instance with the -kernel parameter,
The kernels 486 and 686 were used showing the same issue
Some output from dmesg in VM,
"e820: BIOS-provided physical RAM map:"
"Linux version 3.14-2-486"
'
[ 0.000000] ACPI: RSDP 000f0bf0 000014 (v00 BOCHS )
[ 0.000000] ACPI: ?E\xffffffbah 07fe1854 993C30F (v26 ?R?/??
\xffffffc9\xffffffc4\xffffffc6?/{?? 4F0B0C0B
?\xffffffd7\xffffffa0\xffffffa0\xffffffe8 475E9827)
[ 0.000000] ------------[ cut here ]------------
[ 0.000000] WARNING: CPU: 0 PID: 0 at
/build/linux-neEfbl/linux-3.14.15/arch/x86/mm/ioremap.c:548
__early_ioremap+0x11c/0x19d()
[ 0.000000] Modules linked in:
[ 0.000000] CPU: 0 PID: 0 Comm: swapper Not tainted 3.14-2-486 #1
Debian 3.14.15-2
[ 0.000000] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996),
BIOS debian/1.7.5-1-0-g506b58d-dirty-20140812_231322-gandalf 04/01/2014
[ 0.000000] c13dab08 c103597a c14a4c68 00000000 00000000 c14a3378
00000224 c15a272d
[ 0.000000] c15a272d 00000000 0000993d 00000854 07fe1000 c1035a2b
00000009 00000000
[ 0.000000] c15a272d c102f5cc 0000015d c102f607 00000000 c15a2a2b
3f3f7b2f 00000000
[ 0.000000] Call Trace:
[ 0.000000] [<c13dab08>] ? dump_stack+0xa/0x13
[ 0.000000] [<c103597a>] ? warn_slowpath_common+0x7a/0x90
[ 0.000000] [<c15a272d>] ? __early_ioremap+0x11c/0x19d
[ 0.000000] [<c15a272d>] ? __early_ioremap+0x11c/0x19d
[ 0.000000] [<c1035a2b>] ? warn_slowpath_null+0x1b/0x20
[ 0.000000] [<c15a272d>] ? __early_ioremap+0x11c/0x19d
[ 0.000000] [<c102f5cc>] ? __native_set_fixmap+0x1c/0x30
[ 0.000000] [<c102f607>] ? native_set_fixmap+0x27/0x30
[ 0.000000] [<c15a2a2b>] ? early_iounmap+0xdb/0xf8
[ 0.000000] [<c15a2944>] ? early_ioremap+0xa/0xb
[ 0.000000] [<c15be30b>] ? acpi_tb_parse_root_table+0x1ba/0x316
[ 0.000000] [<c15bc6a7>] ? acpi_table_init+0x14/0x98
[ 0.000000] [<c12dd1c7>] ? dmi_check_system+0x17/0x40
[ 0.000000] [<c159bfa0>] ? acpi_boot_table_init+0x19/0x78
[ 0.000000] [<c15940ce>] ? setup_arch+0xb48/0xd74
[ 0.000000] [<c13d908e>] ? printk+0x37/0x3b
[ 0.000000] [<c1591727>] ? start_kernel+0x6d/0x385
[ 0.000000] ---[ end trace d9827e46b22eb0c7 ]---
[ 0.000000] 0MB HIGHMEM available.
'
(On the native host,
ACPI became broken after kvm was upgraded from Wheezy(stable) to Jessie
)
I tried updating the VM guest which is Wheezy to Jessie, and copied
later kernels from it to use with -kernel, as ACPI was working with the
previous KVM edition. This didn't help..
fwiw, the command I'm using with kvm..
/usr/bin/qemu-system-i386
-cpu kvm32 -m 128 -machine type=pc,accel=kvm
-display none -enable-kvm
-drive
file=<path>/debian.drive.raw.bin,media=disk,if=virtio,cache=none,aio=native
-device virtio-net-pci,mac=<macaddress>,netdev=nic1
-netdev type=tap,id=nic1,ifname=atapdev,script=no,downscript=no
-monitor unix:<path>/mymonitor.socket,server,nowait
-daemonize -runas auser -chroot <path>
-kernel <path>/vmlinuz-3.14-2-486 -initrd <path>/initrd.img-3.14-2-486
-append root=/dev/vda1 quiet ipv6.disable=1
vmlinuz-3.14-2-486 and initrd.img-3.14-2-486 were copied from the VM.
The system can boot and work but I cannot do a full shutdown unless I
type "halt" inside the VM and then later kill the kvm instance.
The native host is Debian Jessie x86_64, the VM is Jessie 32-bit, ..
Both were upgraded but the reason for upgrading the native host was for
me to check out accelerating features of kvm. The VM guest was then
upgraded thinking it would fix the acpi issue but couldn't.
Should I try to to use another bios file?
thanks
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
