Avi Kivity wrote:
On 10/28/2009 11:16 PM, Erik Rull wrote:
Hi all,
after several tests it seems that the interrupt assignment on the host
systems influences the kvm performance drastically. If eth0, video and
usb
are on seperate interrupts the perfomance is way better than if they are
shared
Avi Kivity wrote:
On 11/03/2009 05:11 PM, Timur Safin wrote:
My totally noob in QEMU guess -
my bet it's CR4.OSFXSR which is controlled by presence of
cpuid.1.edx[24] - FXSR bit (FXSAVE and FXRSTOR) instructions.
That would affect floating point as well.
I'm curious - is there any way
On Thu, 2009-11-05 at 09:01 -0800, John Admanski wrote:
Well, the drop_caches configs I definitely like. As a general rule I
think it's reasonable to translate most global defaults into a global
config setting.
Ok, great, so 'drop_caches' will be kept as is.
The sysinfo changes I'm a
Right now autotest will drop caches between:
* Test executions
* Same test iterations
This change turns those into configurable options on
global_config.ini. Default configuration:
[CLIENT]
drop_caches: True
drop_caches_between_iterations: True
Signed-off-by: Lucas Meneghel Rodrigues
Linus, please pull from
git://git.kernel.org/pub/scm/virt/kvm/kvm.git kvm-updates/2.6.32
For two KVM fixes: 4GB TSS access in task switch emulation and MCE
ioctl overflow.
Gleb Natapov (1):
KVM: get_tss_base_addr() should return a gpa_t
Jan Kiszka (1):
KVM: x86: Catch potential
Fede wrote:
On Wed, Nov 4, 2009 at 11:00, Asdo a...@shiftmail.org wrote:
In case of a single video card in the system, it would be wonderful
to be able to suddenly give the video card as passthrough to the
guest removing it from the host. Since the video card internal state
is unknown to the
Avi Kivity wrote:
On 11/03/2009 12:09 AM, Alexander Graf wrote:
When we want to create a full VirtIO based machine, we're still missing
graphics output. Fortunately, Linux provides us with most of the frameworks
to render text and everything, we only need to implement a transport.
So this
We (the AlacrityVM team) are pleased to announce the availability of the
v0.2 release. There are numerous tweaks, fixes, and features that we
have added since the v0.1 days. Here are a few of the key highlights.
*) VENET support:
*) zero-copy transmits (guest memory is paged directly to the
Jamie Lokier wrote:
Avi Kivity wrote:
On 11/03/2009 12:09 AM, Alexander Graf wrote:
When we want to create a full VirtIO based machine, we're still missing
graphics output. Fortunately, Linux provides us with most of the frameworks
to render text and everything, we only need to
On Thu, Nov 05, 2009 at 12:01:10PM +0200, Michael Goldish wrote:
- Put the PCI device removal code in a finally clause.
Hi Michael,
I have a little concern with the removal procedure. Thinking about if
pci_add failed, the output will not contain right information including PCI
ID. The slice
On Thu, Nov 05, 2009 at 12:01:12PM +0200, Michael Goldish wrote:
This should slow the rate of accumulation of monitor files in /tmp.
Hi Michael,
I recommend we use TCP as monitor dev of VM. Two reasons:
1) we don't need to add extra code to remove monitor files
2) it's necessary for some
Perhaps its related to your kernel version?
With RHEL5 as the host OS I have not seen any problems with bonding and
dhcp in either the host or the guest. The stack is:
----
| tapX | ... | tapY |
----
\/
-
| br0 |
On Thu, 5 Nov 2009 02:27:24 am Michael S. Tsirkin wrote:
What it is: vhost net is a character device that can be used to reduce
the number of system calls involved in virtio networking.
Hi Michael,
Now everyone else has finally kicked all the tires and it seems to pass,
I've done a fairly
On Thu, 5 Nov 2009 03:55:42 am Paul E. McKenney wrote:
On Wed, Nov 04, 2009 at 01:57:29PM +0200, Michael S. Tsirkin wrote:
Can you ack this usage please?
I thought I had done so in my paragraph above, but if you would like
something a bit more formal...
snip verbose super-ack with
Is there anything I can do to help with this feature. It is probably one of the
most sort after requests.
I have been toying with the Xen component and would love to see this in KVM
-Original Message-
From: kvm-ow...@vger.kernel.org [mailto:kvm-ow...@vger.kernel.org] On Behalf Of
Han,
Basic graphics passthrough works in xen-unstable, we are working on iGFX
passthrough now (Xen Client already has iGFX passthrough support). After
complete it, we plan to do the same in KVM. Fede said he is porting xen code to
kvm.I think you can also think how to implement it in kvm cleanly.
On Thu, 5 Nov 2009 05:08:42 pm Stephen Rothwell wrote:
Hi Rusty,
On Thu, 5 Nov 2009 16:58:36 +1030 Rusty Russell ru...@rustcorp.com.au wrote:
Huh? virtio_has_feature does:
if (__builtin_constant_p(fbit))
BUILD_BUG_ON(fbit = 32);
else
BUG_ON(fbit
17 matches
Mail list logo