On 01/05/2011 05:48 PM, Avi Kivity wrote:
> On 01/05/2011 11:38 AM, Xiao Guangrong wrote:
>> We can interrupt the vcpu only when it's running in guest mode
>> to reduce IPI.
>>
>> It looks like only ia64 and x86 need to send IPI to other vcpus, so
>> i only add the implementation of 'vcpu->guest_mo
Hello,
On Wed, Jan 05, 2011 at 06:39:19PM +0100, Peter Zijlstra wrote:
> On Wed, 2011-01-05 at 19:35 +0200, Avi Kivity wrote:
>
> > Tejun, why did you end up not using preempt_notifiers in cmwq?
>
> Because I told him to use explicit function calls because that keeps the
> code easier to read.
Yeah, some other folks have suggested that too, but it looks like you need
kernel version 2.6.18-199 or higher to run iotop, and CentOS 5.5 only has
2.6.18-192 or something... Guess I need to wait for the kernel to be updated
there if I want to use iotop, which sounds like a very cool utility.
On Tue, Jan 4, 2011 at 11:19 AM, Erich Weiler wrote:
> Thanks for replying! I was able to figure it out - it was not the fault of
> KVM. One of the guests was running ganglia gmetad which was updating 30,000+
> rrd files every 15 seconds (thus generating load via disk I/O), I didn't spot
> th
We sometimes need to map between the virtio device and
the given pci device. One such use is OS installer that
gets the boot pci device from BIOS and needs to
find the relevant block device. Since it can't,
installation fails.
Supply softlinks between these to make it possible.
Signed-off-by: Mic
On Wed, Jan 05, 2011 at 08:55:44PM +0200, Michael Goldish wrote:
> On 01/05/2011 06:21 PM, Eduardo Habkost wrote:
> > By the way, I think we could make _new_context() and _pop_context() part
> > of the public interface (i.e. remove the "_" from their names). I see
> > @context_aware as just a help
On 01/05/2011 06:21 PM, Eduardo Habkost wrote:
> On Wed, Jan 05, 2011 at 06:12:05PM +0200, Avi Kivity wrote:
>> It would be nice to make the error context a stack, and to use the
>> with statement to manage the stack:
>>
>>
>>with error.context("main test"):
>>foo()
>>with error
On Wed, 2011-01-05 at 19:10 +0200, Avi Kivity wrote:
> On 01/04/2011 08:04 PM, Peter Zijlstra wrote:
> > This definitely wants to be EXPORT_SYMBOL_GPL() and if it were possible
> > I'd make it so only kvm.o could use it. It really sucks that kvm is a
> > module.
>
> Why does it suck? I mean apart
On Wed, 2011-01-05 at 19:35 +0200, Avi Kivity wrote:
> Tejun, why did you end up not using preempt_notifiers in cmwq?
Because I told him to use explicit function calls because that keeps the
code easier to read.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a
On 01/05/2011 07:28 PM, Peter Zijlstra wrote:
On Wed, 2011-01-05 at 19:19 +0200, Avi Kivity wrote:
> On 01/05/2011 07:15 PM, Peter Zijlstra wrote:
> > On Wed, 2011-01-05 at 19:10 +0200, Avi Kivity wrote:
> > > On 01/04/2011 08:04 PM, Peter Zijlstra wrote:
> > > > This definitely want
As I somehow missed to role out an -rc version, this kvm-kmod series
starts with the final 2.6.37 release. You can download it from
http://sourceforge.net/projects/kvm/files/kvm-kmod/2.6.37/kvm-kmod-2.6.37.tar.bz2/download
Major KVM changes since kvm-kmod-2.6.36.2:
- virtual TSC / kvmclock rewor
On Wed, 2011-01-05 at 19:19 +0200, Avi Kivity wrote:
> On 01/05/2011 07:15 PM, Peter Zijlstra wrote:
> > On Wed, 2011-01-05 at 19:10 +0200, Avi Kivity wrote:
> > > On 01/04/2011 08:04 PM, Peter Zijlstra wrote:
> > > > This definitely wants to be EXPORT_SYMBOL_GPL() and if it were
> > > possible
On Wed, 2011-01-05 at 18:04 +0100, Peter Zijlstra wrote:
> On Wed, 2011-01-05 at 17:57 +0100, Mike Galbraith wrote:
> > + p_cfs_rq = cfs_rq_of(pse);
> > + local = 1;
> > + }
> > +#endif
> > +
> > + /* Tell the scheduler that we'd really like pse to run next.
On 01/05/2011 07:15 PM, Peter Zijlstra wrote:
On Wed, 2011-01-05 at 19:10 +0200, Avi Kivity wrote:
> On 01/04/2011 08:04 PM, Peter Zijlstra wrote:
> > This definitely wants to be EXPORT_SYMBOL_GPL() and if it were possible
> > I'd make it so only kvm.o could use it. It really sucks that kvm
On 01/04/2011 08:04 PM, Peter Zijlstra wrote:
This definitely wants to be EXPORT_SYMBOL_GPL() and if it were possible
I'd make it so only kvm.o could use it. It really sucks that kvm is a
module.
Why does it suck? I mean apart from the "virtualization is crap" song.
--
error compiling committ
On 01/05/2011 06:44 PM, Anthony Liguori wrote:
On 01/04/2011 03:39 PM, Marcelo Tosatti wrote:
On Tue, Jan 04, 2011 at 08:17:26AM -0600, Anthony Liguori wrote:
On 01/03/2011 04:01 AM, Avi Kivity wrote:
On 01/03/2011 11:46 AM, Jan Kiszka wrote:
Hi,
at least in kvm mode, the qemu_fair_mutex see
On Wed, 2011-01-05 at 17:57 +0100, Mike Galbraith wrote:
> + p_cfs_rq = cfs_rq_of(pse);
> + local = 1;
> + }
> +#endif
> +
> + /* Tell the scheduler that we'd really like pse to run next. */
> + p_cfs_rq->next = pse;
> +
> + /* We know whether we
Hi,
I don't see this patch in the git tree, nor a revert of the buggy
commit. Was any decision made on this?
thanks,
-serge
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/major
Take a look at Ganeti. It is designed to do exactly what you
describe: HA VMs sharing their LV-backed disks over DRBD.
-Ben
> I'm new to this list and i hope this is the right list for my question. I
> like to create several VM's using KVM on a SLES 11 SP1 host. What is
> better/faster ? Inst
It's Rik's patch to do with whatever he wants (I donated suggestion;),
but I took the liberty of cleaning it up a bit, see below. It's
essentially the same, but fixes a bug where the caller may have targeted
a task in it's thread group etc, but when preemption decision time came,
may have preempte
On 01/04/2011 03:39 PM, Marcelo Tosatti wrote:
On Tue, Jan 04, 2011 at 08:17:26AM -0600, Anthony Liguori wrote:
On 01/03/2011 04:01 AM, Avi Kivity wrote:
On 01/03/2011 11:46 AM, Jan Kiszka wrote:
Hi,
at least in kvm mode, the qemu_fair_mutex seems to have lost its
function o
On 01/05/2011 06:36 PM, Eduardo Habkost wrote:
On Wed, Jan 05, 2011 at 06:21:35PM +0200, Avi Kivity wrote:
> btw, you can have a decorator for enclosing an entire function in an
> error context:
>
> @function_error_context('migration test')
> def migration_test(...):
> ...
@con
On Wed, Jan 05, 2011 at 06:21:35PM +0200, Avi Kivity wrote:
> btw, you can have a decorator for enclosing an entire function in an
> error context:
>
>@function_error_context('migration test')
>def migration_test(...):
>...
@context_aware does that, but it doesn't let you set the
On Tue, Jan 04, 2011 at 09:32:12AM +0100, Jan Kiszka wrote:
> Next round, addressing review comments and new findings:
> - new: Flush coalesced mmio buffer on IO window exits
> - new: Do not use qemu_fair_mutex in kvm mode
> - new: Implicitly clear nmi_injected/pending on reset
> - new: Fix !CO
On Wed, 2011-01-05 at 15:26 +, Daniel P. Berrange wrote:
> On Wed, Jan 05, 2011 at 05:14:55PM +0200, Avi Kivity wrote:
> > On 01/05/2011 04:57 PM, Alex Williamson wrote:
> > >A valid argument. I think it could also be argued that the user is
> > >providing ownership of the file and writing to
On Wed, 2011-01-05 at 17:14 +0200, Avi Kivity wrote:
> On 01/05/2011 04:57 PM, Alex Williamson wrote:
> > A valid argument. I think it could also be argued that the user is
> > providing ownership of the file and writing to the file is part of the
> > low level details of the sysfs rom file API an
On 01/05/2011 06:21 PM, Eduardo Habkost wrote:
On Wed, Jan 05, 2011 at 06:12:05PM +0200, Avi Kivity wrote:
> It would be nice to make the error context a stack, and to use the
> with statement to manage the stack:
>
>
> with error.context("main test"):
> foo()
> with error.c
On 01/05/2011 06:12 PM, Avi Kivity wrote:
On 01/05/2011 05:45 PM, Michael Goldish wrote:
In complex tests (KVM) an exception string is often not informative
enough and
the traceback and source code have to be examined in order to figure
out what
caused the exception. Context strings are a way
On Wed, Jan 05, 2011 at 06:12:05PM +0200, Avi Kivity wrote:
> It would be nice to make the error context a stack, and to use the
> with statement to manage the stack:
>
>
>with error.context("main test"):
>foo()
>with error.context("before reboot"):
>bar()
>
> If
On 01/05/2011 05:45 PM, Michael Goldish wrote:
In complex tests (KVM) an exception string is often not informative enough and
the traceback and source code have to be examined in order to figure out what
caused the exception. Context strings are a way for tests to provide
information about what
On Wed, Jan 05, 2011 at 05:45:26PM +0200, Michael Goldish wrote:
> In complex tests (KVM) an exception string is often not informative enough and
> the traceback and source code have to be examined in order to figure out what
> caused the exception. Context strings are a way for tests to provide
>
This is an example of a case where context information is useful.
Signed-off-by: Michael Goldish
---
.../kvm/tests/migration_with_file_transfer.py | 12 +---
1 files changed, 9 insertions(+), 3 deletions(-)
diff --git a/client/tests/kvm/tests/migration_with_file_transfer.py
b/cl
For consistency with other exception classes, keep just a single blank line
between __init__() and __str__().
Signed-off-by: Michael Goldish
---
client/common_lib/error.py |1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/client/common_lib/error.py b/client/common_lib/erro
Exceptions look for a ._context attribute and embed it in the string returned
by __str__(). If the attribute isn't set (or if it's an empty string) nothing
is embedded.
Signed-off-by: Michael Goldish
---
client/common_lib/error.py | 10 +-
1 files changed, 9 insertions(+), 1 deletions
Signed-off-by: Michael Goldish
---
client/tests/kvm/kvm_preprocessing.py | 13 +++--
1 files changed, 11 insertions(+), 2 deletions(-)
diff --git a/client/tests/kvm/kvm_preprocessing.py
b/client/tests/kvm/kvm_preprocessing.py
index 5ce0a3b..e5b8906 100644
--- a/client/tests/kvm/kvm_pr
In recent Python version, when exceptions are unpickled their __init__() method
is called again. The parameters passed to __init__() upon unpickling are
*self.args. Because the first execution of __init__() sets self.args to
something different from the parameters passed to __init__() (by passing
In complex tests (KVM) an exception string is often not informative enough and
the traceback and source code have to be examined in order to figure out what
caused the exception. Context strings are a way for tests to provide
information about what they're doing, so that when an exception is raise
On Wed, Jan 05, 2011 at 05:14:55PM +0200, Avi Kivity wrote:
> On 01/05/2011 04:57 PM, Alex Williamson wrote:
> >A valid argument. I think it could also be argued that the user is
> >providing ownership of the file and writing to the file is part of the
> >low level details of the sysfs rom file AP
On 01/05/2011 04:57 PM, Alex Williamson wrote:
A valid argument. I think it could also be argued that the user is
providing ownership of the file and writing to the file is part of the
low level details of the sysfs rom file API and should be handled by the
user of that API. We basically have 3
On 01/05/2011 04:26 PM, Reeted wrote:
On 01/05/2011 02:08 PM, Avi Kivity wrote:
My aim is that i want to create a HA-Cluster. For every service
(currently 3) i'd like to create a VM. I want to have two
hosts/nodes, both running SLES 11 SP1. I want to use DRBD to
replicate the File/LV to the
On Wed, 2011-01-05 at 10:57 +0200, Avi Kivity wrote:
> On 01/04/2011 08:45 PM, Alex Williamson wrote:
> > The PCI sysfs rom file is exposed read-only by default, but we need
> > to write to it to enable and disable the ROM around the read. When
> > running as root, the code works fine as is, but w
Hello,
I was able to pass-through an audio card to the VM. I could see the
card connected to the VM, but it seems like it is not working in the
VM.
I tried passing card to Ubuntu 10.04 and Windows 7 as well. Windows7
detected the hardware as 'High Definition Audio Controller' but shows
the device
On 01/05/2011 02:08 PM, Avi Kivity wrote:
My aim is that i want to create a HA-Cluster. For every service
(currently 3) i'd like to create a VM. I want to have two
hosts/nodes, both running SLES 11 SP1. I want to use DRBD to
replicate the File/LV to the passive node. Do you thing that's a goo
Avi Kivity wrote:
> On 01/05/2011 02:27 PM, Lentes, Bernd wrote:
> > Hi,
> >
> > I'm new to this list and i hope this is the right list for
> my question. I like to create several VM's using KVM on a
> SLES 11 SP1 host. What is better/faster ?
>
> Logical volumes are faster.
>
> > Install the V
On 01/05/2011 03:35 PM, Lentes, Bernd wrote:
Thanks for the quick answer. Just to be sure:
From the host's point of view, the LV has no filesystem and does not have to
be mounted ?
Correct. You assign it with the command
qemu-kvm -drive file=/dev/mapper/volgroupname-lvname,cache=none
--
Second and likely last maintenance release for the kvm-kmod-2.6.36
series. I'm working on a 2.6.37 release as well.
KVM changes since kvm-kmod-2.6.36.1:
- VMX: Fix host userspace gsbase corruption
- Correct ordering of ldt reload wrt fs/gs reload
- fix information leak to userland
- Write prot
On 01/05/2011 02:27 PM, Lentes, Bernd wrote:
Hi,
I'm new to this list and i hope this is the right list for my question. I like
to create several VM's using KVM on a SLES 11 SP1 host. What is better/faster ?
Logical volumes are faster.
Install the VM into a file or a logical Volume ? If i c
sorry for the subjectless first message 0:-)
Hi all
I'm a computer science student from germany interested in
virtualization technologies and given I have some free time I'd
like to dig into kvm modding, maybe even actual development ;-)
I've been trying to understand the kvm mmu, with a special
On Tue, Jan 04, 2011 at 06:43:10PM -1000, Zachary Amsden wrote:
> On 01/04/2011 05:36 AM, Marcelo Tosatti wrote:
> >On Tue, Dec 28, 2010 at 07:38:18PM -1000, Zachary Amsden wrote:
> >>During a host suspend, TSC may go backwards, which KVM interprets
> >>as an unstable TSC. Technically, KVM should
Hi,
I'm new to this list and i hope this is the right list for my question. I like
to create several VM's using KVM on a SLES 11 SP1 host. What is better/faster ?
Install the VM into a file or a logical Volume ? If i chosse to install it into
a LV: does the LV needs a Filesystem, and does it ha
On Tue, Dec 28, 2010 at 07:38:16PM -1000, Zachary Amsden wrote:
> A further set of improvements to KVM clock. Now it actually can
> stay synchronized on SMP systems with a stable TSC, does not get
> destabilized by host suspend, and is resistant to migration.
>
> I did look a bit into the second
From: Jan Kiszka
If the kernel does not support KVM_CAP_ASYNC_PF, it also does not know
about the related MSR. So skip it during state synchronization in that
case. Fixes annoying kernel warnings.
Signed-off-by: Jan Kiszka
---
target-i386/kvm.c | 12 ++--
1 files changed, 10 insertio
From: Jan Kiszka
This is required to support keeping VCPU states across a system reset.
If we do not read the current state before the reset,
cpu_synchronize_all_post_reset may write back incorrect state
information.
The first user of this will be MCE MSR synchronization which currently
works ar
Hi all
I'm a computer science student from germany interested in
virtualization technologies and given I have some free time I'd
like to dig into kvm modding, maybe even actual development ;-)
I've been trying to understand the kvm mmu, with a special regard
to paging structures.
I'm having a har
On Wed, 2011-01-05 at 17:40 +0900, KOSAKI Motohiro wrote:
> > > If you are interesting GIL mess and issue, please feel free to ask more.
> >
> > I suggest looking into an explicit round-robin scheme, where each thread
> > adds itself to a queue and an unlock wakes up the first waiter.
>
> I'm su
On 01/05/2011 05:54 PM, Avi Kivity wrote:
>> +kvm_make_request(req, vcpu);
>> cpu = vcpu->cpu;
>> if (cpus != NULL&& cpu != -1&& cpu != me&&
>> atomic_read(&vcpu->guest_mode))
>> @@ -163,7 +161,7 @@ static bool make_all_cpus_request(struct kvm *kvm,
>> unsi
On 01/05/2011 05:48 PM, Avi Kivity wrote:
>> srcu_read_unlock(&vcpu->kvm->srcu, idx);
>> +atomic_set(&vcpu->guest_mode, 1);
>> kvm_guest_enter();
>
> I think it needs an smp_wmb() (technically x86 needs it too, but x86 is
> strongly ordered)
..
>> -if (cpus != NULL&
On Wed, 2011-01-05 at 11:39 +0900, KOSAKI Motohiro wrote:
> After calling pthread_cond_signal(), T1 which cond_signal caller and T2
> which waked start to GIL grab race. But usually T1 is always win because
> lock variable is in T1's cpu cache. Why kernel and userland have so much
> different resu
On 01/05/2011 11:40 AM, Xiao Guangrong wrote:
Now, we have 'vcpu->guest_mode' to judge whether need to send
ipi to other cpus, this way is very exact, so checking request
bit is needless, then we can drop the spinlock let it's collateral
Clever.
@@ -147,11 +147,9 @@ static bool make_all_cpus_
On 01/05/2011 11:38 AM, Xiao Guangrong wrote:
We can interrupt the vcpu only when it's running in guest mode
to reduce IPI.
It looks like only ia64 and x86 need to send IPI to other vcpus, so
i only add the implementation of 'vcpu->guest_mode' in ia64, but i
don't know ia64 well, please point ou
On Wed, 2011-01-05 at 00:38 +0100, Tommaso Cucinotta wrote:
> Il 04/01/2011 19:15, Dario Faggioli ha scritto:
> >
> > Forwarded Message
> > From: Peter Zijlstra
> > To: Rik van Riel
> > Cc: Hillf Danton,kvm@vger.kernel.org,
> > linux-ker...@vger.kernel.org, Avi Kiviti, Srivatsa
>
Now, we have 'vcpu->guest_mode' to judge whether need to send
ipi to other cpus, this way is very exact, so checking request
bit is needless, then we can drop the spinlock let it's collateral
There is the simple test result of kernbench:
before patch:
real5m5.100s
We can interrupt the vcpu only when it's running in guest mode
to reduce IPI.
It looks like only ia64 and x86 need to send IPI to other vcpus, so
i only add the implementation of 'vcpu->guest_mode' in ia64, but i
don't know ia64 well, please point out the right way for me if the
implementation is
Fix:
[ 1001.499596] ===
[ 1001.499599] [ INFO: suspicious rcu_dereference_check() usage. ]
[ 1001.499601] ---
..
[ 1001.499633] stack backtrace:
[ 1001.499636] Pid: 6035, comm: qemu-system-x8
On 01/05/2011 11:30 AM, KOSAKI Motohiro wrote:
> On 01/05/2011 10:40 AM, KOSAKI Motohiro wrote:
> > > On 01/05/2011 04:39 AM, KOSAKI Motohiro wrote:
> > > > >On 01/04/2011 08:14 AM, KOSAKI Motohiro wrote:
> > > > >>Also, If pthread_cond_signal() call sys_yield_to
iml
> On 01/05/2011 10:40 AM, KOSAKI Motohiro wrote:
> > > On 01/05/2011 04:39 AM, KOSAKI Motohiro wrote:
> > > > > On 01/04/2011 08:14 AM, KOSAKI Motohiro wrote:
> > > > > > Also, If pthread_cond_signal() call sys_yield_to imlicitly,
> > > we can
> > > > > > avoid almost Nehalem (and
On 01/05/2011 10:40 AM, KOSAKI Motohiro wrote:
> On 01/05/2011 04:39 AM, KOSAKI Motohiro wrote:
> > > On 01/04/2011 08:14 AM, KOSAKI Motohiro wrote:
> > > > Also, If pthread_cond_signal() call sys_yield_to imlicitly, we can
> > > > avoid almost Nehalem (and other P2P cache arch)
Am 05.01.2011 09:33, Huang Ying wrote:
> On Wed, 2011-01-05 at 16:07 +0800, Jan Kiszka wrote:
>> Am 05.01.2011 07:42, Huang Ying wrote:
>>> On Tue, 2011-01-04 at 16:32 +0800, Jan Kiszka wrote:
From: Jan Kiszka
There is no need to restrict writing back MCE MSRs to reset or full
On 01/04/2011 08:45 PM, Alex Williamson wrote:
The PCI sysfs rom file is exposed read-only by default, but we need
to write to it to enable and disable the ROM around the read. When
running as root, the code works fine as is, but when running
de-privileged via libvirt, the fopen("r+") will fail
On 01/04/2011 05:43 PM, Anthony Liguori wrote:
The fact that the iothread drops the global lock during sleep is a
detail that shouldn't affect correctness. The IO thread is
absolutely allowed to run for arbitrary periods of time without
dropping the qemu mutex.
No, it's not, since it will st
> On 01/05/2011 04:39 AM, KOSAKI Motohiro wrote:
> > > On 01/04/2011 08:14 AM, KOSAKI Motohiro wrote:
> > > > Also, If pthread_cond_signal() call sys_yield_to imlicitly, we can
> > > > avoid almost Nehalem (and other P2P cache arch) lock unfairness
> > > > problem. (probaby creating pthread_
On 01/05/2011 04:39 AM, KOSAKI Motohiro wrote:
> On 01/04/2011 08:14 AM, KOSAKI Motohiro wrote:
> > Also, If pthread_cond_signal() call sys_yield_to imlicitly, we can
> > avoid almost Nehalem (and other P2P cache arch) lock unfairness
> > problem. (probaby creating pthread_condattr_setauto
On Wed, 2011-01-05 at 16:07 +0800, Jan Kiszka wrote:
> Am 05.01.2011 07:42, Huang Ying wrote:
> > On Tue, 2011-01-04 at 16:32 +0800, Jan Kiszka wrote:
> >> From: Jan Kiszka
> >>
> >> There is no need to restrict writing back MCE MSRs to reset or full
> >> state updates as setting their values has
Am 05.01.2011 07:45, Huang Ying wrote:
> On Fri, 2010-12-31 at 17:10 +0800, Jan Kiszka wrote:
>> Am 31.12.2010 06:22, Huang Ying wrote:
>>> In Linux kernel HWPoison processing implementation, the virtual
>>> address in processes mapping the error physical memory page is marked
>>> as HWPoison. So
Am 05.01.2011 07:42, Huang Ying wrote:
> On Tue, 2011-01-04 at 16:32 +0800, Jan Kiszka wrote:
>> From: Jan Kiszka
>>
>> There is no need to restrict writing back MCE MSRs to reset or full
>> state updates as setting their values has no side effects.
>
> Sorry for late.
Don't worry.
>
> The MCE
75 matches
Mail list logo
