On 09/22/2009 04:54 AM, Marcelo Tosatti wrote:
On Mon, Sep 21, 2009 at 11:55:37AM +0300, Avi Kivity wrote:
Wait for 2000 cycles after the IPI to allow the host to schedule out.
Measures wake-from-idle overhead.
Signed-off-by: Avi Kivitya...@redhat.com
---
kvm/user/test/x86/vmexit.c |
On 09/22/2009 04:12 AM, Huang Ying wrote:
On Mon, 2009-09-21 at 18:08 +0800, Avi Kivity wrote:
On 09/21/2009 05:43 AM, Huang Ying wrote:
UCR (uncorrected recovery) MCE is supported in recent Intel CPUs,
where some hardware error such as some memory error can be reported
without PCC
On Fri, 18 Sep 2009 03:01:42 am Christoph Hellwig wrote:
Err, I'll take this one back for now pending some more discussion.
What we need more urgently is the writeback cache flag, which is now
implemented in qemu, patch following ASAP.
OK, still catching up on mail. I'll push them out of the
On 09/22/2009 02:37 AM, Marcelo Tosatti wrote:
get_dirty_log vs mark_page_dirty need to be mutually exclusive. Switch
to mmu_lock protection.
I'm not sure all archs use mmu_lock. Also, I'm unhappy about
introducing more use (especially as it's often unnecessary, if dirty
logging is
On 09/22/2009 02:37 AM, Marcelo Tosatti wrote:
Required for SRCU convertion later.
My feeling is that prepare/commit can be common code and only some small
bits (like allocating x86 rmaps) can be arch specific. That's for
another patchset though.
--
Do not meddle in the internals of
On 09/22/2009 02:37 AM, Marcelo Tosatti wrote:
Use two steps for memslot deletion: mark the slot invalid (which stops
instantiation of new shadow pages for that slot, but allows destruction),
then instantiate the new empty slot.
Also simplifies kvm_handle_hva locking.
unsigned int
On 09/22/2009 02:37 AM, Marcelo Tosatti wrote:
Using a similar two-step procedure as for memslots.
- gfn = unalias_gfn(kvm, gfn);
+ gfn = unalias_gfn(kvm, gfn, false);
To improve readability suggest two names, maybe unalias_gfn() and
unalias_gfn_instantiation(). boolean
On 09/22/2009 02:37 AM, Marcelo Tosatti wrote:
Signed-off-by: Marcelo Tosattimtosa...@redhat.com
I'd like an eleventh patch that converts slots_lock to a mutex; there
should be no readers left, yes?
--
Do not meddle in the internals of kernels, for they are subtle and quick to
panic.
On 09/22/2009 02:37 AM, Marcelo Tosatti wrote:
Improves vmexit latency by ~= 13% on UP guest.
Still requires decent stress testing and careful review, but posting for
early comments.
Okay, looks very nice. The only real problem I see is mmu_lock for
mark_dirty(), the other comments are
On 09/21/2009 06:33 PM, Mohammed Gamal wrote:
With emulate_invalid_guest_state=1 Windows XP exits with an invalid guest state
due to rflags not being in a VMX-compliant state. This patch fixes this issue,
although Windows XP doesn't boot yet with invalid state emulation on.
Also added GDT and
On Tue, Sep 22, 2009 at 9:13 AM, Avi Kivity a...@redhat.com wrote:
On 09/21/2009 06:33 PM, Mohammed Gamal wrote:
With emulate_invalid_guest_state=1 Windows XP exits with an invalid guest
state
due to rflags not being in a VMX-compliant state. This patch fixes this
issue,
although Windows XP
On Mon, Sep 21, 2009 at 08:30:26AM -0400, Michael Goldish wrote:
- Yolkfull Chow yz...@redhat.com wrote:
On Sun, Sep 20, 2009 at 06:16:28PM +0300, Michael Goldish wrote:
Some Windows programs behave badly when their STDIN is redirected to
a pipe
(most notably wmic). Therefore,
On 09/19/2009 09:40 AM, Avi Kivity wrote:
Add a general per-cpu notifier that is called whenever the kernel is
about to return to userspace. The notifier uses a thread_info flag
and existing checks, so there is no impact on user return or context
switch fast paths.
Ingo/Peter?
--
error
Hi,
I'm unable to build kvm from source on machine, where I run kvm virtualization.
I've applied your patch to kvm-88 source.
./configure
Error: libpci check failed
Disable KVM Device Assignment capability.
rm: cannot remove `include/asm': Is a directory
rm: cannot remove `include-compat/asm':
On Tue, 22 Sep 2009 12:25:33 +0300
Avi Kivity a...@redhat.com wrote:
On 09/19/2009 09:40 AM, Avi Kivity wrote:
Add a general per-cpu notifier that is called whenever the kernel is
about to return to userspace. The notifier uses a thread_info flag
and existing checks, so there is no impact
On 09/22/2009 12:43 AM, Ira W. Snyder wrote:
Sure, virtio-ira and he is on his own to make a bus-model under that, or
virtio-vbus + vbus-ira-connector to use the vbus framework. Either
model can work, I agree.
Yes, I'm having to create my own bus model, a-la lguest, virtio-pci, and
On 09/22/2009 12:37 PM, Arjan van de Ven wrote:
On Tue, 22 Sep 2009 12:25:33 +0300
Avi Kivitya...@redhat.com wrote:
On 09/19/2009 09:40 AM, Avi Kivity wrote:
Add a general per-cpu notifier that is called whenever the kernel is
about to return to userspace. The notifier uses a
On Mon, Sep 21, 2009 at 09:27:18AM -0700, Chris Wright wrote:
* Stephen Hemminger (shemmin...@vyatta.com) wrote:
On Mon, 21 Sep 2009 16:37:22 +0930
Rusty Russell ru...@rustcorp.com.au wrote:
Actually this framework can apply to traditional network adapters
which have
just
Resending with Cc: added
On Mon, 2009-09-21 at 20:37 -0300, Marcelo Tosatti wrote:
plain text document attachment (introduce-srcu-and-use-for-slots)
Use two steps for memslot deletion: mark the slot invalid (which stops
instantiation of new shadow pages for that slot, but allows
Hi
Earlier post was one big mistake. Everything is compiled now (from kvm-88
sources).
I don't want to corrupt present kvm installation, or better to say, I want to
modify it as small as possible. Could you suggest me what should I replace or
how to run new compilation instead of old one
On Tuesday 22 September 2009, Michael S. Tsirkin wrote:
More importantly, when virtualizations is used with multi-queue
NIC's the virtio-net NIC is a single CPU bottleneck. The virtio-net
NIC should preserve the parallelism (lock free) using multiple
receive/transmit queues. The number
On (Tue) Sep 22 2009 [13:31:11], Michal Filka wrote:
I don't want to corrupt present kvm installation, or better to say, I want to
modify it as small as possible. Could you suggest me what should I replace or
how to run new compilation instead of old one without installing?
You can
On Tue, Sep 22, 2009 at 09:37:51AM +0300, Avi Kivity wrote:
On 09/22/2009 02:37 AM, Marcelo Tosatti wrote:
get_dirty_log vs mark_page_dirty need to be mutually exclusive. Switch
to mmu_lock protection.
I'm not sure all archs use mmu_lock. Also, I'm unhappy about
introducing more use
On 09/22/2009 03:44 PM, Marcelo Tosatti wrote:
I think you can use rcu for this as well. When you read the log,
allocate a new empty bitmap, switch the memslots pointer to include it,
and synchronize_srcu(). Now we are certain everyone is using the new
bitmap we can copy the old one to
On Tue, Sep 22, 2009 at 07:40:10AM -0300, Fernando Carrijo wrote:
Resending with Cc: added
On Mon, 2009-09-21 at 20:37 -0300, Marcelo Tosatti wrote:
- kvm_arch_flush_shadow(kvm);
+ flush_shadow = 1;
}
#else /* not defined CONFIG_S390 */
Btw, what's the state of getting compatfd upstream? It's a pretty
annoying difference between qemu upstream and qemu-kvm.
--
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
Christoph Hellwig h...@infradead.org wrote:
Btw, what's the state of getting compatfd upstream? It's a pretty
annoying difference between qemu upstream and qemu-kvm.
I haven't tried. I can try to send a patch. Do you have any use case
that will help the cause?
Later, Juan.
--
To unsubscribe
On Tue, Sep 22, 2009 at 03:25:13PM +0200, Juan Quintela wrote:
Christoph Hellwig h...@infradead.org wrote:
Btw, what's the state of getting compatfd upstream? It's a pretty
annoying difference between qemu upstream and qemu-kvm.
I haven't tried. I can try to send a patch. Do you have
* Avi Kivity a...@redhat.com wrote:
On 09/19/2009 09:40 AM, Avi Kivity wrote:
Add a general per-cpu notifier that is called whenever the kernel is
about to return to userspace. The notifier uses a thread_info flag
and existing checks, so there is no impact on user return or context
switch
On 09/22/2009 05:32 PM, Ingo Molnar wrote:
* Avi Kivitya...@redhat.com wrote:
On 09/19/2009 09:40 AM, Avi Kivity wrote:
Add a general per-cpu notifier that is called whenever the kernel is
about to return to userspace. The notifier uses a thread_info flag
and existing checks, so
Signed-off-by: Gleb Natapov g...@redhat.com
---
qemu-kvm-x86.c |9 ++---
1 files changed, 6 insertions(+), 3 deletions(-)
diff --git a/qemu-kvm-x86.c b/qemu-kvm-x86.c
index 32561f0..acb1b91 100644
--- a/qemu-kvm-x86.c
+++ b/qemu-kvm-x86.c
@@ -1673,9 +1673,12 @@ uint32_t
Each caller of the function already calls it.
Signed-off-by: Gleb Natapov g...@redhat.com
---
hw/apic.c |1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/hw/apic.c b/hw/apic.c
index 3a2e128..a9d1fb8 100644
--- a/hw/apic.c
+++ b/hw/apic.c
@@ -488,7 +488,6 @@ void
Otherwise cpu is always unhalted by call to kvm_arch_get_registers()
Signed-off-by: Gleb Natapov g...@redhat.com
---
qemu-kvm.c | 10 ++
qemu-kvm.h |9 +
2 files changed, 11 insertions(+), 8 deletions(-)
diff --git a/qemu-kvm.c b/qemu-kvm.c
index 6511cb6..2a7fe3d 100644
On Tue, Sep 22, 2009 at 12:43:36PM +0300, Avi Kivity wrote:
On 09/22/2009 12:43 AM, Ira W. Snyder wrote:
Sure, virtio-ira and he is on his own to make a bus-model under that, or
virtio-vbus + vbus-ira-connector to use the vbus framework. Either
model can work, I agree.
Yes,
Ingo Molnar wrote:
* Avi Kivity a...@redhat.com wrote:
On 09/19/2009 09:40 AM, Avi Kivity wrote:
Add a general per-cpu notifier that is called whenever the kernel is
about to return to userspace. The notifier uses a thread_info flag
and existing checks, so there is no impact on user return
On 09/22/2009 05:45 PM, Avi Kivity wrote:
Would be nice to convert some existing open-coded return-to-user-space
logic to this facility. One such candidate would be lockdep_sys_exit?
I only implemented this for x86, while lockdep is arch independent.
If arch support is added, it should be
On 09/22/2009 06:25 PM, Ira W. Snyder wrote:
Yes. vbus is more finely layered so there is less code duplication.
The virtio layering was more or less dictated by Xen which doesn't have
shared memory (it uses grant references instead). As a matter of fact
lguest, kvm/pci, and kvm/s390 all
Amit Shah schrieb:
On (Tue) Sep 22 2009 [13:31:11], Michal Filka wrote:
I don't want to corrupt present kvm installation, or better to say, I want
to modify it as small as possible. Could you suggest me what should I
replace or how to run new compilation instead of old one without
On Tue, Sep 22, 2009 at 09:59:04AM +0300, Avi Kivity wrote:
On 09/22/2009 02:37 AM, Marcelo Tosatti wrote:
Use two steps for memslot deletion: mark the slot invalid (which stops
instantiation of new shadow pages for that slot, but allows destruction),
then instantiate the new empty slot.
Alpar Torok
[...]
On really new kvms this is no longer true, you can now load a snapshot saved
on a processor from another vendor.
I am really happy to hear that, another good thing about the latest
release is that loadvm from console works. Previously Win XP BSOD-ed.
[..]
In deed, this
On Tue, 22 Sep 2009 13:50:54 +0200
Arnd Bergmann a...@arndb.de wrote:
On Tuesday 22 September 2009, Michael S. Tsirkin wrote:
More importantly, when virtualizations is used with multi-queue
NIC's the virtio-net NIC is a single CPU bottleneck. The virtio-net
NIC should preserve the
On Tue, 2009-09-22 at 16:32 +0200, Ingo Molnar wrote:
* Avi Kivity a...@redhat.com wrote:
On 09/19/2009 09:40 AM, Avi Kivity wrote:
Add a general per-cpu notifier that is called whenever the kernel is
about to return to userspace. The notifier uses a thread_info flag
and existing
On 09/22/2009 07:50 PM, Peter Zijlstra wrote:
Would be nice to convert some existing open-coded return-to-user-space
logic to this facility. One such candidate would be lockdep_sys_exit?
And here I was thinking this was one of the hottest code paths in the
whole kernel...
If you're
On Tue, 2009-09-22 at 19:52 +0300, Avi Kivity wrote:
On 09/22/2009 07:50 PM, Peter Zijlstra wrote:
Would be nice to convert some existing open-coded return-to-user-space
logic to this facility. One such candidate would be lockdep_sys_exit?
And here I was thinking this was one of the
On 09/22/2009 07:55 PM, Peter Zijlstra wrote:
If you're using lockdep, surely that's not your biggest worry?
No, but that's all under #ifdef and fully disappears when not enabled.
Generic return-tu-user notifiers don't sound like they will though.
They will if not selected. If
On 09/22/2009 06:50 PM, Avi Kivity wrote:
On 09/22/2009 05:45 PM, Avi Kivity wrote:
Would be nice to convert some existing open-coded return-to-user-space
logic to this facility. One such candidate would be lockdep_sys_exit?
I only implemented this for x86, while lockdep is arch independent.
On Tue, 2009-09-22 at 16:32 +0200, Ingo Molnar wrote:
Would be nice to convert some existing open-coded return-to-user-space
logic to this facility. One such candidate would be lockdep_sys_exit?
I don't really like lockdep_sys_exit() in such a call,
lockdep_sys_exit() is currently placed such
On Tuesday 22 September 2009, Stephen Hemminger wrote:
My idea for that was to open multiple file descriptors to the same
macvtap device and let the kernel figure out the right thing to
do with that. You can do the same with raw packed sockets in case
of vhost_net, but I wouldn't want to
Bugs item #2864630, was opened at 2009-09-22 17:05
Message generated for change (Tracker Item Submitted) made by aolney
You can respond by visiting:
https://sourceforge.net/tracker/?func=detailatid=893831aid=2864630group_id=180599
Please note that this message will contain a full copy of the
On Fri, Sep 18, 2009 at 03:00:32PM +0200, Alexander Graf wrote:
Normally when event_inj is valid the host CPU would write the contents to
exit_int_info, so the hypervisor knows that the event wasn't injected.
We failed to do so so far, so let's model closer to the CPU.
Signed-off-by:
On Fri, Sep 18, 2009 at 03:00:29PM +0200, Alexander Graf wrote:
SVM has some cleanup code, that tries to reinject interrupts and exceptions
when the guest didn't manage to deal with them yet. It basically transfers
them to KVM internal state.
Unfortunately, the internal state is reserved for
On Fri, Sep 18, 2009 at 03:00:28PM +0200, Alexander Graf wrote:
When injecting an NMI to the l1 guest while it was running the l2 guest, we
didn't #VMEXIT but just injected the NMI to the l2 guest.
Let's be closer to real hardware and #VMEXIT if we're supposed to do so.
Signed-off-by:
On Fri, Sep 18, 2009 at 03:00:30PM +0200, Alexander Graf wrote:
Real hardware would first process the event_inj field and then notify the
host that an interrupt is waiting.
Does it really? I couldn't find this in the SVM spec.
Let's do the same and just not EXIT_INTR if we have an event
09/23 09:50:27 ERROR| traceback:0013| File
/usr/lib64/python2.4/logging/__init__.py, line 744, in emit
09/23 09:50:27 ERROR| traceback:0013| self.handleError(record)
09/23 09:50:27 ERROR| traceback:0013| File
/root/project/autotest/client/setup_modules.py, line 86, in
Sorry for lost autot...@test.kernel.org. So send again. Thanks.
--
fix an error of undefine variable
09/23 09:50:27 ERROR| traceback:0013| File
/usr/lib64/python2.4/logging/__init__.py, line 744, in emit
09/23 09:50:27 ERROR| traceback:0013|
Hello,
Guests which use Message Signalled Interrupts at least with virtio block device
hang the guest with the latest qemu-kvm tree.
Tested on AMD Phenom II. Should be easy to reproduce.
Regards,
Will Trives
This email was sent
Hi,
I did a couple of tests.
1. test
Kernel modules from distribution (kvm-72) new user space application
(qemu-system-x86_64 - patched kvm-88)
It seems that it is not working
Possibilities:
1) ru is not ready because of other reason than switching to suspended state
2) patch is not working
57 matches
Mail list logo
