For those of you who would like to assist in testing without waiting for
the next release, yet are uncomfortable with git, I've made available
daily development snapshots of kvm in
http://people.qumranet.com/avi/snapshots/
If you report a bug in a snaphot, please specify both the snapshot nam
Jeremy Fitzhardinge wrote:
> Nakajima, Jun wrote:
> > The hypervisor detection machanism is generic, and the signature
> > returned is implentation specific. Having a list of all hypervisor
> > signatures sounds fine to me as we are detecting vendor-specific
> > processor(s) in the native. And I do
Nakajima, Jun wrote:
> The hypervisor detection machanism is generic, and the signature
> returned is implentation specific. Having a list of all hypervisor
> signatures sounds fine to me as we are detecting vendor-specific
> processor(s) in the native. And I don't expect the list is large.
>
>
On Fri, 2007-09-14 at 16:44 -0500, Anthony Liguori wrote:
> So then each module creates a hypercall page using this magic MSR and
> the hypervisor has to keep track of it so that it can appropriately
> change the page on migration. The page can only contain a single
> instruction or else it ca
On Sat, 2007-09-15 at 01:31 +0300, Dor Laor wrote:
> Second, regardless of the channel signal notification, there are still
> real necessities for userspace hypercall handling:
> 1. For virtio drivers there is also registration hypercall for passing
> the shared memory pfns.
> Sure there are
On Fri, 2007-09-14 at 13:53 -0700, Jeremy Fitzhardinge wrote:
> Anthony Liguori wrote:
> > This patch refactors the current hypercall infrastructure to better support
> > live
> > migration and SMP. It eliminates the hypercall page by trapping the UD
> > exception that would occur if you used the
Jeremy Fitzhardinge wrote:
> Nakajima, Jun wrote:
> > Today, 3 CPUID leaves starting from 0x4000_ are defined in a
generic
> > fashion (hypervisor detection, version, and hypercall page), and
those
> > are the ones used by Xen today. We should extend those leaves (e.g.
> > starting from 0x4000_
Nakajima, Jun wrote:
> Today, 3 CPUID leaves starting from 0x4000_ are defined in a generic
> fashion (hypervisor detection, version, and hypercall page), and those
> are the ones used by Xen today. We should extend those leaves (e.g.
> starting from 0x4000_0003) for the vmm-independent feature
>
>The push model may require parallelization of the disk and memory
>convergence depending on how much disk activity the guest is doing.
>That's a pretty interesting problem but it shouldn't be too
>hard to solve.
>
Embed this kind of cache like disk ops (write and read hit always
goes to loc
Jeremy Fitzhardinge wrote:
> Nakajima, Jun wrote:
> > > > one. Start the kvm leaves at 0x40001000 or something?
> > > >
> > > >
> > > Yeah, that works with me.
> > >
> >
> > To me this is the beginning of fragmentation. Why do we need
different
> > and VMM-specific Linux paravirtualization for
Nakajima, Jun wrote:
>>> one. Start the kvm leaves at 0x40001000 or something?
>>>
>>>
>> Yeah, that works with me.
>>
>
> To me this is the beginning of fragmentation. Why do we need different
> and VMM-specific Linux paravirtualization for hardware-assisted
> virtualization? That wou
Anthony Liguori wrote:
> Jeremy Fitzhardinge wrote:
> > Anthony Liguori wrote:
> >
> > > Yeah, see, the initial goal was to make it possible to use the KVM
> > > paravirtualizations on other hypervisors. However, I don't think
this
> > > is really going to be possible in general so maybe it's bet
Avi Kivity wrote:
> Anthony Liguori wrote:
>
>> Dor Laor wrote:
>>
>>
Hi all,
I've finally started looking at Dor's git tree, and it struck me
>>> that
>>>
>>>
>>>
it conflicts with Anthony's hypercall patches. FW
Jeremy Fitzhardinge wrote:
> Anthony Liguori wrote:
>
>> Yeah, see, the initial goal was to make it possible to use the KVM
>> paravirtualizations on other hypervisors. However, I don't think this
>> is really going to be possible in general so maybe it's better to just
>> use leaf 0. I'll let
Anthony Liguori wrote:
> Yeah, see, the initial goal was to make it possible to use the KVM
> paravirtualizations on other hypervisors. However, I don't think this
> is really going to be possible in general so maybe it's better to just
> use leaf 0. I'll let others chime in before sending a new
Jeremy Fitzhardinge wrote:
> Anthony Liguori wrote:
>
>> The whole point of using the instruction is to allow hypercalls to be
>> used in many locations. This has the nice side effect of not
>> requiring a central hypercall initialization routine in the guest to
>> fetch the hypercall page. A
Zachary Amsden wrote:
> On Fri, 2007-09-14 at 16:02 -0500, Anthony Liguori wrote:
>
>> Jeremy Fitzhardinge wrote:
>>
>>> Anthony Liguori wrote:
>>>
>>>
This patch refactors the current hypercall infrastructure to better
support live
migration and SMP. It eliminate
Anthony Liguori wrote:
> The whole point of using the instruction is to allow hypercalls to be
> used in many locations. This has the nice side effect of not
> requiring a central hypercall initialization routine in the guest to
> fetch the hypercall page. A PV driver can be completely independen
On Fri, 2007-09-14 at 16:02 -0500, Anthony Liguori wrote:
> Jeremy Fitzhardinge wrote:
> > Anthony Liguori wrote:
> >
> >> This patch refactors the current hypercall infrastructure to better
> >> support live
> >> migration and SMP. It eliminates the hypercall page by trapping the UD
> >> exce
Haydn Solomon wrote:
> I see that vmware just demonstrated migration of virtual disks. I know
> that kvm is very young but I think has come a long way in a very short
> period. I just wanted to get some feedback/discussion on how difficult
> this would be to implement in kvm/qemu and if this wou
Jeremy Fitzhardinge wrote:
> Anthony Liguori wrote:
>
>> This patch refactors the current hypercall infrastructure to better support
>> live
>> migration and SMP. It eliminates the hypercall page by trapping the UD
>> exception that would occur if you used the wrong hypercall instruction for
Anthony Liguori wrote:
> This patch refactors the current hypercall infrastructure to better support
> live
> migration and SMP. It eliminates the hypercall page by trapping the UD
> exception that would occur if you used the wrong hypercall instruction for the
> underlying architecture and repla
I see that vmware just demonstrated migration of virtual disks. I know that
kvm is very young but I think has come a long way in a very short period. I
just wanted to get some feedback/discussion on how difficult this would be
to implement in kvm/qemu and if this would be anywhere in future plans f
Hi Jeremy,
I gave this patch a try today against the latest git
1edeb9c05aa034633978f31e2c773a1be55ed337 and I cannot get past the
"Press F10..." prompt. I'm on an Intel Core 2 Duo processor. It does
work though with -no-kvm.
Have you tested this patch with KVM?
Regards,
Anthony Liguori
J
This adds a CPUID entry for the paravirtualization feature bitmap. We can do
this unconditionally because the guest requires that both the feature CPUID
entry and the signature CPUID entry exists to enable paravirtualization.
This means that guest will never enable paravirtualization if either us
Attached is the patch I'm using to test it. I'm going to take a look at
Rusty's pv_ops implementation for kvm-lite so that we can submit a
single common one instead of introducing mine and having unnecessary churn.
Regards,
Anthony Liguori
Subject: [PATCH 2/3] KVM paravirt-ops implementation
This patch refactors the current hypercall infrastructure to better support live
migration and SMP. It eliminates the hypercall page by trapping the UD
exception that would occur if you used the wrong hypercall instruction for the
underlying architecture and replacing it with the right one lazily.
On 9/14/07, Cam Macdonell <[EMAIL PROTECTED]> wrote:
>
> Hi Ron,
>
> This would be interesting to try, especially if booting is noticeably
> faster that bochs.
It would definitely be interesting to try.
Cheers,
Jorge
-
This
Hi Ron,
This would be interesting to try, especially if booting is noticeably
faster that bochs.
Did you get that how-to up?
Thanks,
Cam
ron minnich wrote:
> Somebody here at OLS was asking me about linuxbios on kvm. Well, that
> was too much to resist.
>
> Short form: it works, uses a grub-
My main virtual machine I work with is windows xp. Thanks for the reply and
I'm looking forward to future releases!
On 9/14/07, Luca <[EMAIL PROTECTED]> wrote:
>
> On 9/14/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> > Haydn Solomon wrote:
> > > You guys are quick! Ok, kvm-39 allows me to run my wi
Greetings,
I am trying to duplicate what I have done with the Free version of
VmWare server, with KVM. Linux, Fedora 7 host and Fedora Core 3 guest..
Since the KVM with fedora 7 is not current I decided to try and build
KVM from source.
After discovering the gcc 3.4. requirement, I
Hi Avi,
Attached is the patch to implement emulation of instruction:
dec reg
opcodes: 0x48 - 0x4f
Please apply.
--
Thanks & Regards,
Nitin
Open Source Technology Center, Intel Corporation
-
The mind is like a parach
Hi Avi,
Attached is the patch to implement instructions:
inc reg
opcode: 0x40 - 0x47
Please apply.
--
Thanks & Regards,
Nitin
Open Source Technology Center, Intel Corporation
-
The mind is like a parachute; it work
Hi Avi,
Attached is the patch to implement instruction:
jump absolute
opcode: 0xff /4
Please apply.
--
Thanks & Regards,
Nitin
Open Source Technology Center, Intel Corporation
-
The mind is like a parachute; it wor
Nitin A Kamble wrote:
> On Fri, 2007-09-14 at 10:08 -0700, Avi Kivity wrote:
>
>> This seems like it is papering over other bugs. Some instructions use
>> src.val or dst.val without having decoded the src or dst operand.
>>
>> Which instructions are these? Can we fix them instead?
>>
>
>
On Fri, 2007-09-14 at 10:08 -0700, Avi Kivity wrote:
> This seems like it is papering over other bugs. Some instructions use
> src.val or dst.val without having decoded the src or dst operand.
>
> Which instructions are these? Can we fix them instead?
Instructions using 8bit operands such as
On 9/14/07, Avi Kivity <[EMAIL PROTECTED]> wrote:
> Haydn Solomon wrote:
> > You guys are quick! Ok, kvm-39 allows me to run my windows xp machine
> > now. I just have one question that I'm not sure of.. should reboot be
> > working now? I tried rebooting the windows xp and got the following
> > o
A guest context switch to an uncached cr3 can require allocation of
shadow pages, but we only recycle shadow pages in kvm_mmu_page_fault().
Move shadow page recycling to mmu_topup_memory_caches(), which is called
from both the page fault handler and from guest cr3 reload.
Signed-off-by: Avi Kivit
Laurent Vivier wrote:
> Avi Kivity wrote:
>
>> Laurent Vivier wrote:
>>
>>> These patches split the emulator in two parts: one to decode the
>>> instruction,
>>> the other to execute it. The decode part is then called only when needed.
>>>
>>>
>>>
>> Patchset looks good, but fails
Wei Huang wrote:
> The compilation errors also showed up on my boxes. The attached patch
> fixes the errors by renaming variables.
>
> Signed-off-by: Wei Huang <[EMAIL PROTECTED]>
> diff --git a/drivers/kvm/kvm_main.c b/drivers/kvm/kvm_main.c
> index 99e4917..f04ca6b 100644
> --- a/drivers/kvm/kvm_
Haydn Solomon wrote:
> You guys are quick! Ok, kvm-39 allows me to run my windows xp machine
> now. I just have one question that I'm not sure of.. should reboot be
> working now? I tried rebooting the windows xp and got the following
> output.
>
Reboot was improved in kvm-39, but not completely
Sebastian Färber wrote:
> Hi,
>
> i just upgraded to lastet git and now get the following error in dmesg:
> --
> emulation failed (mmio) rip c2008430 60 17 97 f8
> --
>
Sorry, I can't really track down what 'latest git' means.
Can you re-fetch and try again, and if it fails post the HEAD commi
Nitin A Kamble wrote:
> Hi Avi,
> Attached is the patch to initialize src.val & dst.val. Without this,
> certain instructions are getting affected in their emulation.
>
> Please apply.
>
>
This seems like it is papering over other bugs. Some instructions use
src.val or dst.val without havi
Gildas wrote:
> Hi,
>
> Some updates since kvm-37
>
> Compilation worked fine for me on an intel Core2 64bits machine
> running ubuntu feisty with kernel 2.6.20 x86_64.
>
> Loading modules went ok.
>
> * Running a windows xp image with "/usr/local/bin/qemu-system-x86_64
> -smp 2 -m 768 -hda win_xp.
Avi Kivity wrote:
> Rusty Russell wrote:
>
>> Hi all,
>>
>> I've finally started looking at Dor's git tree, and it struck me that
>> it conflicts with Anthony's hypercall patches. FWIW I like Anthony's
>> patching thing, and don't really care about arg order. It'd be nice if
>> we could p
Anthony Liguori wrote:
> Dor Laor wrote:
>
>>> Hi all,
>>>
>>> I've finally started looking at Dor's git tree, and it struck me
>>>
>>>
>> that
>>
>>
>>> it conflicts with Anthony's hypercall patches. FWIW I like Anthony's
>>> patching thing, and don't really care about
Rusty Russell wrote:
> Hi all,
>
> I've finally started looking at Dor's git tree, and it struck me that
> it conflicts with Anthony's hypercall patches. FWIW I like Anthony's
> patching thing, and don't really care about arg order. It'd be nice if
> we could pull in the same direction tho
Avi Kivity wrote:
> Laurent Vivier wrote:
>> These patches split the emulator in two parts: one to decode the
>> instruction,
>> the other to execute it. The decode part is then called only when needed.
>>
>>
>
> Patchset looks good, but fails booting FC6 x86-64 on Intel. It may be a
> merge e
Hi,
We run some tests against latest kvm.git
6617597ecdbdff1419d2e942b6aa301f82fe4598 and kvm-userspace.git
c109b253be21f951cf4e86d6135e50e65c2eb4f6.
One issue fixed:
The 64bit crash issue has been fixed.
Following Issues still exist
1. guest timer
49 matches
Mail list logo
