[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] xm/xl shutdown does not work with HVM guest



WR is WindRiver distribution. I don't see any interrupt count going up for acpi.

   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi
   9:          0          0          0          0   IO-APIC-fasteoi   acpi

I don't see any ACPI events in the HVM guest either.

root@lc-6:/root>
root@lc-6:/root> ps -eaf | grep acpid
root      3771     1  0 21:57 ?        00:00:00 /usr/sbin/acpid
root      6684  3882  0 22:00 ttyS0    00:00:00 grep acpid
root@lc-6:/root> kill -9 3771
root@lc-6:/root> cat /proc/acpi/event

root@lc-6:/root> service acpid restart
Stopping acpi daemon: [FAILED]
Starting acpi daemon: [  OK  ]
root@lc-6:/root> service acpid restart
Stopping acpi daemon: [  OK  ]
Starting acpi daemon: [  OK  ]
root@lc-6:/root> 

I issued trigger like this :-

xm trigger pvm-01-6 power


HVM guest cfg file is :-

boot = "c"
memory = 8192
vcpus = 4
disk = [ 'file:/root/PSVs/mnt_local_ssd/local_ssd/pvm-6/ssc_pvm_01.img,hda,w', ',hdc:cdrom,r' ]
vif = [ 'model=e1000, mac=00:16:3e:00:05:00, bridge=br0', 'model=e1000, mac=00:16:3f:00:05:01, bridge=br1' ]
pci = [  '0000:08:00.0=0@0b', '0000:01:00.0=0@0c', '0000:07:11.6=0@1a', '0000:07:11.7=0@1b', '0000:88:11.6=0@1c', '0000:88:11.7=0@1d' ]
cpus = [  '36', '37', '38', '39' ]
 
#
# --- Mandatory config file entries ---
#

# HVM specific
kernel = "hvmloader"
builder = "hvm"
device_model = "qemu-dm"

# Enable ACPI support
acpi = 1

# Enable serial console
serial = "pty"

# Enable VNC
vnc = 1
vnclisten = "0.0.0.0"

pci_msitranslate = 0

# Default behavior for following events
>
 
# Enable Xen Platform PCI device for Platform VM
xen_platform_pci=0


Thanks,
/Saurabh

On Fri, Feb 21, 2014 at 1:13 PM, Konrad Rzeszutek Wilk <konrad.wilk@xxxxxxxxxx> wrote:
On Fri, Feb 21, 2014 at 09:51:52AM -0800, Saurabh Mishra wrote:
> Hi --
>
> I tried enabling CONFIG_APM but it still didn't work. Let me know if you
> guys happen to know what all needs to be enabled in guest HVM for ACPI
> events to work (xm trigger <vm> power). Since SuSE HVM VM works with 'xm
> trigger <vm> power', I'm suspecting we have not enabled something in our WR
> distro.

What is 'WR'? Anyhow, you might also need ACPI. Do you get any ACPI
events at all in your guest?

Aka, does /proc/interrupts for acpi show an increasing number as you
do 'xl trigger poweroff' ?

>
> Thanks,
> /Saurabh
>
>
> On Thu, Feb 20, 2014 at 4:26 PM, Saurabh Mishra <saurabh.globe@xxxxxxxxx>wrote:
>
> > Hi Ian --
> >
> > So enabling 'CONFIG_APM=y' in WR kernel for HVM guest should be enough?
> > Let me try that out.
> >
> > Thanks,
> > /Saurabh
> >
> >
> > On Thu, Feb 20, 2014 at 1:15 AM, Ian Campbell <Ian.Campbell@xxxxxxxxxx>wrote:
> >
> >> On Wed, 2014-02-19 at 11:02 -0800, Saurabh Mishra wrote:
> >> > config or driver do I need to enable in WR HVM guest such that it
> >> > accepts 'xl/xm trigger <vm> power'?
> >>
> >> Support for ACPI power events.
> >>
> >> Ian.
> >>
> >>
> >>
> >

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.