Hi, I''m trying to boot a hvm guest directly on a linux kernel. Here is my config file: name=''test'' builder=''hvm'' memory = 512 vcpus=1 cpus = [''1''] disk = [''phy:/dev/iscsi/hosting-2253:0,hda,w''] device_model = ''/usr/lib/xen/bin/qemu-dm'' loader=''/usr/lib/xen/boot/hvmloader'' kernel=''/boot/bzImage'' serial=''pty'' apic=0 acpi=1 pae=0 videoram=4 stdvga=0 nographic=1 sdl=0 isa=0 display=0 vncdisplay=0 I''m using Xen 4.1.1 with an upstream linux 3.1-rc1 for the dom0. The console is not working and the qemu-dm process is dying after a while Here are the qemu-dm log: # cat /var/log/xen/qemu-dm-test.log domid: 58 -videoram option does not work with cirrus vga device model. Videoram set to 4M. Using file /dev/iscsi/hosting-2253:0 in read-write mode Watching /local/domain/0/device-model/58/logdirty/cmd Watching /local/domain/0/device-model/58/command Watching /local/domain/58/cpu char device redirected to /dev/pts/1 qemu_map_cache_init nr_buckets = 10000 size 4194304 shared page at pfn feffd buffered io page at pfn feffb Guest uuid = fa5d2eb8-d026-9f66-8ccc-3f89f1545517 Time offset set 0 populating video RAM at ff000000 mapping video RAM from ff000000 Register xen platform. Done register platform. platform_fixed_ioport: changed ro/rw state of ROM memory area. now is rw state. xs_read(/local/domain/0/device-model/58/xen_extended_power_mgmt): read error Log-dirty: no command yet. I/O request not ready: 0, ptr: 0, port: 0, data: 0, count: 0, size: 0 (qemu) vcpu-set: watch node error. char device redirected to /dev/pts/2 xen be: console-0: xen be: console-0: initialise() failed initialise() failed xs_read(/local/domain/58/log-throttling): read error qemu: ignoring not-understood drive `/local/domain/58/log-throttling'' medium change watch on `/local/domain/58/log-throttling'' - unknown device, ignored xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed xen be: console-0: xen be: console-0: initialise() failed initialise() failed cirrus vga map change while on lfb mode mapping vram to f0000000 - f0400000 platform_fixed_ioport: changed ro/rw state of ROM memory area. now is rw state. platform_fixed_ioport: changed ro/rw state of ROM memory area. now is ro state. I was wondering how could I fix the following errors: xs_read(/local/domain/0/device-model/58/xen_extended_power_mgmt): read error qemu: ignoring not-understood drive `/local/domain/58/log-throttling'' xen be: console-0: xen be: console-0: initialise() failed I can''t see what I forgot to make it boot on the kernel and make the console working. Regards, -- William _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Hi, On Thu, Aug 11, 2011 at 2:56 PM, William Dauchy <wdauchy@gmail.com> wrote:> I''m trying to boot a hvm guest directly on a linux kernel.I finally figured out that the direct kernel boot is broken since at least xen 4.x (both with qemu-dm and upstream qemu) -- William _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Thu, 11 Aug 2011, William Dauchy wrote:> Hi, > > I''m trying to boot a hvm guest directly on a linux kernel.That has never been supported AFAIK. Of course we could accept patches to make it work. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Hi Stefani, Thank you for your reply. On Tue, Aug 23, 2011 at 7:02 PM, Stefano Stabellini <stefano.stabellini@eu.citrix.com> wrote:> That has never been supported AFAIK. > Of course we could accept patches to make it work.Probably but there was config options to support this (loader + kernel) and code in both xend and qemu-dm. In upstream qemu, the code is present to support direct boot but the xen implementation is quite different from the usual "qemu way", at least for grabbing ram at start; see `pc_memory_init` which is bypassed by `xen_enabled`. I will see if I find time to patch the upstream qemu version. -- William _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Tue, 23 Aug 2011, William Dauchy wrote:> Hi Stefani, > > Thank you for your reply. > > On Tue, Aug 23, 2011 at 7:02 PM, Stefano Stabellini > <stefano.stabellini@eu.citrix.com> wrote: > > That has never been supported AFAIK. > > Of course we could accept patches to make it work. > > Probably but there was config options to support this (loader + > kernel) and code in both xend and qemu-dm. >It is still present for PV guests (remove builder=hvm), you can use kernel and ramdisk.> In upstream qemu, the code is present to support direct boot but the > xen implementation is quite different from the usual "qemu way", at > least for grabbing ram at start; see `pc_memory_init` which is > bypassed by `xen_enabled`. > I will see if I find time to patch the upstream qemu version.Unfortuntely it is not just a matter of editing qemu, you need to start from libxc and libxenlight: the linux kernel image should probably be passed by libxenlight to xc_hvm_build_target_mem, that should be able to read it and map it in the guest. _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
Hi Stefano, On Wed, Aug 24, 2011 at 1:07 PM, Stefano Stabellini <stefano.stabellini@eu.citrix.com> wrote:> It is still present for PV guests (remove builder=hvm), you can use > kernel and ramdisk.Yes I know that.> Unfortuntely it is not just a matter of editing qemu, you need to start > from libxc and libxenlight: the linux kernel image should probably be > passed by libxenlight to xc_hvm_build_target_mem, that should be able to > read it and map it in the guest.Thanks for the details. I will dig into it if I find time to. -- William _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Wed, Aug 24, 2011 at 02:08:44PM +0200, William Dauchy wrote:> Hi Stefano, > > On Wed, Aug 24, 2011 at 1:07 PM, Stefano Stabellini > <stefano.stabellini@eu.citrix.com> wrote: > > It is still present for PV guests (remove builder=hvm), you can use > > kernel and ramdisk. > > Yes I know that. > > > Unfortuntely it is not just a matter of editing qemu, you need to start > > from libxc and libxenlight: the linux kernel image should probably be > > passed by libxenlight to xc_hvm_build_target_mem, that should be able to > > read it and map it in the guest. > > Thanks for the details. I will dig into it if I find time to. >Hello, Some user on ##xen was asking for this feature.. did you have time to try it? -- Pasi _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel