Le lundi 08 novembre 2010 à 16:03 +0800, xiaohui@intel.com a écrit :
> From: Xin Xiaohui
>
> >> Hmm, I suggest you read the comment two lines above.
> >>
> >> If destructor_arg is now cleared each time we allocate a new skb, then,
> >> please move it before dataref in shinfo structure, so tha
>-Original Message-
>From: Eric Dumazet [mailto:eric.duma...@gmail.com]
>Sent: Monday, November 08, 2010 4:25 PM
>To: Xin, Xiaohui
>Cc: net...@vger.kernel.org; kvm@vger.kernel.org; linux-ker...@vger.kernel.org;
>m...@redhat.com; mi...@elte.hu; da...@davemloft.net;
>herb...@gondor.apana.org
Hi,
I'm seeing lock-ups of the QEMU process on kvm.git as well as current
upstream kernels. This is a backtrace of the hanging VCPU thread:
[] __stop_cpus+0x184/0x1a7
[] try_stop_cpus+0x40/0x59
[] synchronize_sched_expedited+0x84/0x9d
[] __synchronize_srcu+0x33/0x72
[] synchronize_srcu_expedited+
On Mon, Nov 08, 2010 at 10:04:00AM +0100, Jan Kiszka wrote:
> Hi,
>
> I'm seeing lock-ups of the QEMU process on kvm.git as well as current
> upstream kernels. This is a backtrace of the hanging VCPU thread:
>
> [] __stop_cpus+0x184/0x1a7
> [] try_stop_cpus+0x40/0x59
> [] synchronize_sched_expedi
Am 08.11.2010 10:18, Markus Trippelsdorf wrote:
> On Mon, Nov 08, 2010 at 10:04:00AM +0100, Jan Kiszka wrote:
>> Hi,
>>
>> I'm seeing lock-ups of the QEMU process on kvm.git as well as current
>> upstream kernels. This is a backtrace of the hanging VCPU thread:
>>
>> [] __stop_cpus+0x184/0x1a7
>> [
Cosmetic change, but it helps to correlate IRQs with PCI devices.
Signed-off-by: Jan Kiszka
---
include/linux/kvm_host.h |1 +
virt/kvm/assigned-dev.c | 11 ++-
2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index
The guest may change states that pci_reset_function does not touch. So
we better save/restore the assigned device across guest usage.
Signed-off-by: Jan Kiszka
---
virt/kvm/assigned-dev.c |5 -
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/virt/kvm/assigned-dev.c b/virt/
Adds API documentation for KVM_[DE]ASSIGN_PCI_DEVICE,
KVM_[DE]ASSIGN_DEV_IRQ, KVM_SET_GSI_ROUTING, KVM_ASSIGN_SET_MSIX_NR, and
KVM_ASSIGN_SET_MSIX_ENTRY.
Signed-off-by: Jan Kiszka
---
Documentation/kvm/api.txt | 178 +
1 files changed, 178 insertions(
Ack and mask notifiers typically call back into kvm_set_irq, thus may
iterate over all VCPUs of a VM. Better keep this path preemptible to
prevent that user-space can massivle influence scheduling latencies. Use
sleepable RCU for the protection of irq_routing and the notfier lists.
Signed-off-by:
Clean up the srcu struct on vm destruction and refactor its release on
early errors.
Signed-off-by: Jan Kiszka
---
virt/kvm/kvm_main.c | 15 +++
1 files changed, 7 insertions(+), 8 deletions(-)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 4111a4b..6cfcde7 100644
--
Nine patches (yeah, it's getting more and more) to improve "classic"
device assigment /wrt IRQs. Highlight is the last one that resolves the
host IRQ sharing issue for all PCI 2.3 devices. Quite essential when
passing non-MSI-ready devices like many USB host controllers.
As there were concerns reg
This improves the IRQ forwarding for assigned devices: By using the
kernel's threaded IRQ scheme, we can get rid of the latency-prone work
queue and simplify the code in the same run.
Moreover, we no longer have to hold assigned_dev_lock while raising the
guest IRQ, which can be a lenghty operatio
PCI 2.3 allows to generically disable IRQ sources at device level. This
enables us to share IRQs of such devices between on the host side when
passing them to a guest. This feature is optional, user space has to
request it explicitly. Moreover, user space can inform us about its view
of PCI_COMMAND
Any arch not supporting device assigment will also not build
assigned-dev.c. So testing for KVM_CAP_DEVICE_DEASSIGNMENT is pointless.
KVM_CAP_ASSIGN_DEV_IRQ is unconditinally set. Moreover, add a default
case for dispatching the ioctl.
Signed-off-by: Jan Kiszka
---
virt/kvm/assigned-dev.c |9
When we deassign a guest IRQ, clear the potentially asserted guest line.
There might be no chance for the guest to do this, specifically if we
switch from INTx to MSI mode.
Signed-off-by: Jan Kiszka
---
virt/kvm/assigned-dev.c |3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --
Make use of the new KVM feature that allows legacy interrupt sharing
for PCI-2.3-compliant devices. As exclusive mode (with IRQ masking at
interrupt controller level) is generally faster, the new mode has to be
enabled explicitly via the property "host_pci_2_3".
Signed-off-by: Jan Kiszka
---
hw/
On Wed, Oct 06, 2010 at 02:58:57PM -0600, Alex Williamson wrote:
> Our code paths for saving or migrating a VM are full of functions that
> return void, leaving no opportunity for a device to cancel a migration,
> either from error or incompatibility. The ivshmem driver attempted to
> solve this w
I run my application both on xen and kvm, and I found the performance
on kvm is worse than xen.
kvm: 2.6.32-60.el6.x86_64 redhat 6 bate 2
xen: xen3.4.2
then i stat the vm exit reason, and i found the ept violation reason,
kvm is much more than xen. the static result is :
(XEN) 10 times on cp
Hi,
I'm running a Windows XP guest in KVM. After enabling the audio support
(-soundhw es1370) the CPU load of the KVM process in the host system
goes up to 100%.
Tested with: qemu-kvm-0.13.0 Debian 0.13.0+dfsg-1
KVM start command: kvm -m 512 -hda xp.img -soundhw es1370
Did you experience simil
On Mon, Nov 08, 2010 at 10:14:35AM +0800, Xiao Guangrong wrote:
> On 11/05/2010 06:31 PM, Gleb Natapov wrote:
> > On Fri, Nov 05, 2010 at 04:03:28PM +0800, Xiao Guangrong wrote:
> >> On 11/05/2010 03:45 PM, Gleb Natapov wrote:
> >>
>
> It looks like something broken: apfs can generated in
On 11/08/2010 02:15 PM, lidong chen wrote:
I run my application both on xen and kvm, and I found the performance
on kvm is worse than xen.
kvm: 2.6.32-60.el6.x86_64 redhat 6 bate 2
xen: xen3.4.2
then i stat the vm exit reason, and i found the ept violation reason,
kvm is much more than xen. the
Am Montag 08 November 2010 07:32:39 schrieb Stefan Hajnoczi:
> If you have the time, you can use perf probes to trace I/O requests in
> the host kernel. Perhaps completion interrupts are being dropped.
> You may wish to start by tracing requests issued and completed by the
> SATA driver.
Fortunat
Please send in any agenda items you are interested in covering.
For once I start:
- report from linux plumbers virt track (jes?, anthony?)
- report from linux kernel summit (avi?)
Cheers, Juan.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord..
On Mon, 2010-11-08 at 13:40 +0200, Michael S. Tsirkin wrote:
> On Wed, Oct 06, 2010 at 02:58:57PM -0600, Alex Williamson wrote:
> > Our code paths for saving or migrating a VM are full of functions that
> > return void, leaving no opportunity for a device to cancel a migration,
> > either from erro
> What workload is this? Are you using virtio?
yes,i using virtio and vhost. the cpu rate is 70%.
I found it maybe caused by 2M HugePages. if didn't use 2M
HugePages, there are no vm exits caused by ept violation.
I will test the performance without 2M Hugepages later, and see why
HugePage ca
you can use "perf top" to collect some information.
2010/11/8 Stefan Pietsch :
> Hi,
>
> I'm running a Windows XP guest in KVM. After enabling the audio support
> (-soundhw es1370) the CPU load of the KVM process in the host system
> goes up to 100%.
>
> Tested with: qemu-kvm-0.13.0 Debian 0.13.0+
On Mon, Nov 08, 2010 at 07:59:57AM -0700, Alex Williamson wrote:
> On Mon, 2010-11-08 at 13:40 +0200, Michael S. Tsirkin wrote:
> > On Wed, Oct 06, 2010 at 02:58:57PM -0600, Alex Williamson wrote:
> > > Our code paths for saving or migrating a VM are full of functions that
> > > return void, leavin
I mean the second case, the problem is in "d:", now, A-apf is on
L0->L1 mapping, and A-apf.gva is nested fault address(L2's physics
address in L0->L2 mapping), you use this address to retry #pf in
L0->L1 mapping? it is just waste.
On 11/8/10, Gleb Natapov wrote:
> On Mon, Nov 08, 2010 at 10:14:35
On Mon, Nov 08, 2010 at 12:21:45PM +0100, Jan Kiszka wrote:
> Clean up the srcu struct on vm destruction and refactor its release on
> early errors.
>
> Signed-off-by: Jan Kiszka
Yay, I suspected something's wrong with error handling in srcu.
Acked-by: Michael S. Tsirkin
This one actually has
On Tue, Nov 09, 2010 at 12:58:24AM +0800, Xiao Guangrong wrote:
> I mean the second case, the problem is in "d:", now, A-apf is on
> L0->L1 mapping, and A-apf.gva is nested fault address(L2's physics
> address in L0->L2 mapping), you use this address to retry #pf in
> L0->L1 mapping? it is just was
On 11/08/2010 08:23 AM, Juan Quintela wrote:
Please send in any agenda items you are interested in covering.
For once I start:
- report from linux plumbers virt track (jes?, anthony?)
- report from linux kernel summit (avi?)
Good thinking.
Regards,
Anthony Liguori
Cheers, Juan.
--
To u
On Sun, Oct 31, 2010 at 01:40:08PM +0200, Gleb Natapov wrote:
> Domain should be determined form parent bus and bus number is configured
> by guest and should not be used in qemu internally.
>
> Signed-off-by: Gleb Natapov
> ---
> hw/pci.c | 11 ++-
> 1 files changed, 6 insertions(+),
On Mon, 2010-11-08 at 18:54 +0200, Michael S. Tsirkin wrote:
> On Mon, Nov 08, 2010 at 07:59:57AM -0700, Alex Williamson wrote:
> > On Mon, 2010-11-08 at 13:40 +0200, Michael S. Tsirkin wrote:
> > > On Wed, Oct 06, 2010 at 02:58:57PM -0600, Alex Williamson wrote:
> > > > Our code paths for saving o
On Sun, Oct 31, 2010 at 01:40:08PM +0200, Gleb Natapov wrote:
> Domain should be determined form parent bus and bus number is configured
> by guest and should not be used in qemu internally.
>
> Signed-off-by: Gleb Natapov
> ---
> hw/pci.c | 11 ++-
> 1 files changed, 6 insertions(+),
On Mon, Nov 08, 2010 at 07:17:37PM +0200, Michael S. Tsirkin wrote:
> On Sun, Oct 31, 2010 at 01:40:08PM +0200, Gleb Natapov wrote:
> > Domain should be determined form parent bus and bus number is configured
> > by guest and should not be used in qemu internally.
> >
> > Signed-off-by: Gleb Natap
Am 08.11.2010 18:00, Michael S. Tsirkin wrote:
> On Mon, Nov 08, 2010 at 12:21:45PM +0100, Jan Kiszka wrote:
>> Clean up the srcu struct on vm destruction and refactor its release on
>> early errors.
>>
>> Signed-off-by: Jan Kiszka
>
> Yay, I suspected something's wrong with error handling in src
On Mon, Nov 08, 2010 at 07:29:50PM +0200, Gleb Natapov wrote:
> On Mon, Nov 08, 2010 at 07:17:37PM +0200, Michael S. Tsirkin wrote:
> > On Sun, Oct 31, 2010 at 01:40:08PM +0200, Gleb Natapov wrote:
> > > Domain should be determined form parent bus and bus number is configured
> > > by guest and sho
On Mon, Nov 08, 2010 at 08:12:55PM +0200, Michael S. Tsirkin wrote:
> On Mon, Nov 08, 2010 at 07:29:50PM +0200, Gleb Natapov wrote:
> > On Mon, Nov 08, 2010 at 07:17:37PM +0200, Michael S. Tsirkin wrote:
> > > On Sun, Oct 31, 2010 at 01:40:08PM +0200, Gleb Natapov wrote:
> > > > Domain should be de
Hi,
after booting 2.6.36 on one of my testing machines, modprobe kvm-intel
produces:
[ 1349.991299] general protection fault: [#1] PREEMPT SMP
[ 1349.997199] last sysfs file: /sys/module/drbd/parameters/cn_idx
[ 1350.003201] CPU 2
[ 1350.005081] Modules linked in: kvm(+) ocfs2 ocfs2_nodeman
Hi, Anthony,
On Mon, 2010-11-08 at 11:39 +0800, Anthony Liguori wrote:
> On 11/07/2010 07:29 PM, Huang Ying wrote:
> > On Sun, 2010-11-07 at 00:24 +0800, Avi Kivity wrote:
> >
> >> On 11/01/2010 03:22 PM, Anthony Liguori wrote:
> >>
> >>> On 11/01/2010 02:20 PM, Huang Ying wrote:
> >>>
On Mon, Nov 08, 2010 at 10:20:46AM -0700, Alex Williamson wrote:
> On Mon, 2010-11-08 at 18:54 +0200, Michael S. Tsirkin wrote:
> > On Mon, Nov 08, 2010 at 07:59:57AM -0700, Alex Williamson wrote:
> > > On Mon, 2010-11-08 at 13:40 +0200, Michael S. Tsirkin wrote:
> > > > On Wed, Oct 06, 2010 at 02:
On Mon, Nov 08, 2010 at 07:29:50PM +0200, Gleb Natapov wrote:
> > This will make it compatible with existing code and generally
> > what users expect.
> >
> It will drop domain and bus number part anyway.
Yes but the proposal was to add :00: in front to
make it have the expected format for mo
On Mon, Nov 08, 2010 at 11:00:17PM +0200, Michael S. Tsirkin wrote:
> On Mon, Nov 08, 2010 at 07:29:50PM +0200, Gleb Natapov wrote:
> > > This will make it compatible with existing code and generally
> > > what users expect.
> > >
> > It will drop domain and bus number part anyway.
>
> Yes but th
On Mon, 2010-11-08 at 22:59 +0200, Michael S. Tsirkin wrote:
> On Mon, Nov 08, 2010 at 10:20:46AM -0700, Alex Williamson wrote:
> > On Mon, 2010-11-08 at 18:54 +0200, Michael S. Tsirkin wrote:
> > > On Mon, Nov 08, 2010 at 07:59:57AM -0700, Alex Williamson wrote:
> > > > On Mon, 2010-11-08 at 13:40
I'm trying to convert a physical Windows XP machine into a KVM guest.
All the guides so far mentions using dd to create a flat image file,
then using qemu-img to convert that to qcow2. Since I've been making
mistake here and there, retrying the process several times (initially
converting each logic
09.11.2010 01:48, Emmanuel Noobadmin wrote:
> I'm trying to convert a physical Windows XP machine into a KVM guest.
> All the guides so far mentions using dd to create a flat image file,
> then using qemu-img to convert that to qcow2. Since I've been making
> mistake here and there, retrying the pr
08.11.2010 15:40, Stefan Pietsch wrote:
> Hi,
>
> I'm running a Windows XP guest in KVM. After enabling the audio support
> (-soundhw es1370) the CPU load of the KVM process in the host system
> goes up to 100%.
>
> Tested with: qemu-kvm-0.13.0 Debian 0.13.0+dfsg-1
>
> KVM start command: kvm -m
Thanks for the confirmation and just for the benefit of anybody else
who subsequently searches for KVM QEMU convert physical
drive virtual machine image , yes it works :)
On 11/9/10, Michael Tokarev wrote:
> 09.11.2010 01:48, Emmanuel Noobadmin wrote:
>> I'm trying to convert a physical Windows
On Mon, 2010-11-08 at 11:39 +0800, Anthony Liguori wrote:
> On 11/07/2010 07:29 PM, Huang Ying wrote:
> > On Sun, 2010-11-07 at 00:24 +0800, Avi Kivity wrote:
> >
> >> On 11/01/2010 03:22 PM, Anthony Liguori wrote:
> >>
> >>> On 11/01/2010 02:20 PM, Huang Ying wrote:
> >>>
>
"Michael S. Tsirkin" wrote on 10/26/2010 02:27:09 PM:
> Re: [v3 RFC PATCH 0/4] Implement multiqueue virtio-net
>
> On Mon, Oct 25, 2010 at 09:20:38PM +0530, Krishna Kumar2 wrote:
> > > Krishna Kumar2/India/i...@ibmin wrote on 10/20/2010 02:24:52 PM:
> >
> > Any feedback, comments, objections, iss
sorry,i take mistake,
the ept violation is not caused by 2M hugepages, but caused by e1000
emulated by qemu.
I test uesd 2M hugepages and virio, the performance is better than
e1000. but is still worse than xen. I will continue to find out other
reason.
so, there are two question:
1.why have so
On Tue, Nov 09, 2010 at 03:37:44PM +0800, lidong chen wrote:
> sorry,i take mistake,
>
> the ept violation is not caused by 2M hugepages, but caused by e1000
> emulated by qemu.
> I test uesd 2M hugepages and virio, the performance is better than
> e1000. but is still worse than xen. I will contin
09.11.2010 05:54, Emmanuel Noobadmin wrote:
> Thanks for the confirmation and just for the benefit of anybody else
> who subsequently searches for KVM QEMU convert physical
> drive virtual machine image , yes it works :)
Heh. Well, it is not something unexpected really. Just a few more
comments
53 matches
Mail list logo