I'm not entirely sure this is needed:
- How often do tests fail due to unsupported flags? I thought the -mem-path
problem was only due to a temporary typo (I'm not sure how temporary).
Once it's fixed, will we ever encounter that misspelling again?
- A typo is a bug too, so maybe we shouldn't make
On (Tue) Jul 28 2009 [08:42:32], Anthony Liguori wrote:
Amit Shah wrote:
Right; use virtio just as the transport and all the interesting
activity happens in userspaces. That was the basis with which I started.
I can imagine dbus doing the copy/paste, lock screen, etc. actions.
However for
On Wed, Jul 29, 2009 at 01:14:18PM +0530, Amit Shah wrote:
On (Tue) Jul 28 2009 [08:42:32], Anthony Liguori wrote:
Amit Shah wrote:
Right; use virtio just as the transport and all the interesting
activity happens in userspaces. That was the basis with which I started.
I can imagine dbus
Dne 29.7.2009 05:40, Lucas Meneghel Rodrigues napsal(a):
This patch adds a small setup script to set up huge memory
pages during the kvm tests execution. Also, added hugepage setup to the
fc8_quick sample.
Signed-off-by: Lukáš Doktorldok...@redhat.com
Signed-off-by: Lucas Meneghel
The Buildbot has detected a new failure of disable_kvm_i386_debian_5_0 on
qemu-kvm.
Full details are available at:
http://buildbot.b1-systems.de/qemu-kvm/builders/disable_kvm_i386_debian_5_0/builds/16
Buildbot URL: http://buildbot.b1-systems.de/qemu-kvm/
Buildslave for this Build:
The Buildbot has detected a new failure of default_x86_64_debian_5_0 on
qemu-kvm.
Full details are available at:
http://buildbot.b1-systems.de/qemu-kvm/builders/default_x86_64_debian_5_0/builds/17
Buildbot URL: http://buildbot.b1-systems.de/qemu-kvm/
Buildslave for this Build: b1_qemu_kvm_1
The Buildbot has detected a new failure of disable_kvm_x86_64_debian_5_0 on
qemu-kvm.
Full details are available at:
http://buildbot.b1-systems.de/qemu-kvm/builders/disable_kvm_x86_64_debian_5_0/builds/15
Buildbot URL: http://buildbot.b1-systems.de/qemu-kvm/
Buildslave for this Build:
The Buildbot has detected a new failure of default_i386_debian_5_0 on qemu-kvm.
Full details are available at:
http://buildbot.b1-systems.de/qemu-kvm/builders/default_i386_debian_5_0/builds/18
Buildbot URL: http://buildbot.b1-systems.de/qemu-kvm/
Buildslave for this Build: b1_qemu_kvm_2
Build
Hello Lucas,
I like your patch but I'm not entirely convinced about it necessity.
Stable version of KVM should have this fixed and unstable ones are for
developers, who are skilled enough to fix this using kvm_test.cfg.
On the other hand keep this patch somewhere. Eventually if qemu started
Anthony Liguori wrote:
Izik Eidus wrote:
You mean: when we later call for other madvise calls, if it will
remove the MADV_MERGEABLE from that memory?
if yes, the answer is no, it should be still l left in the
vma-vm_flags...
Excellent.
I'd suggest doing the following in osdep.h too:
On 07/29/2009 04:27 AM, Anthony Liguori wrote:
I also think that we could get away with getting rid of any checks for
!sync_mmu() since that was introduced in 2.6.27.
People are running kvm on 2.6.18.
--
error compiling committee.c: too many arguments to function
--
To unsubscribe from this
On 07/27/2009 05:30 PM, Joerg Roedel wrote:
Hello,
this is the latest version of the changes to KVM MMU code to support 1gb
pages. Pages with a size of 1gb are a feature of all Fam10h AMD CPUs
and this patch set makes them available for guests and allows to map the
nested page table with 1gb
On 07/29/2009 01:46 AM, Bartlomiej Zolnierkiewicz wrote:
From: Bartlomiej Zolnierkiewiczbzoln...@gmail.com
Subject: [PATCH] kvm: remove superfluous NULL pointer check in
kvm_inject_pit_timer_irqs()
This takes care of the following entries from Dan's list:
arch/x86/kvm/i8254.c +714
On 07/28/2009 09:26 PM, Marcelo Tosatti wrote:
See patches for details.
Applied, thanks.
Please add the failure scenario to the changelog in the future.
--
error compiling committee.c: too many arguments to function
--
To unsubscribe from this list: send the line unsubscribe kvm in
the
Alexander Graf wrote:
On 28.07.2009, at 23:28, Glauber Costa wrote:
On Tue, Jul 28, 2009 at 11:15:19PM +0200, Alexander Graf wrote:
On 28.07.2009, at 22:52, Glauber Costa glom...@redhat.com wrote:
Falling back to tcg has proven to be evil through time. The option is
to
do not try to
On 07/29/2009 01:17 PM, Jan Kiszka wrote:
I just wonder now if/when qemu-kvm will switch over to the
kvm-by-default-off policy of upstream?
That will surely inconvenience/surprise a lot of users.
A migration path could be:
- add -accel
- start warning when -accel is not used, encouraging
On 07/24/2009 08:23 AM, Glauber Costa wrote:
it is not needed.
Applied, thanks.
--
error compiling committee.c: too many arguments to function
--
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
On 07/24/2009 08:57 AM, Glauber Costa wrote:
Creation of those devices can be a lot simpler. In particular,
if we start by assuming they will be on, and disabling later if
anything fails (user request counts as failing here), we can remove
two variables that really serves no reason.
The
On 07/24/2009 09:01 PM, Glauber Costa wrote:
Marcelo,
these two patches move the control of coalescing regions to upstream code.
Note the first one won't apply in your tip. I'm assuming you have already
applied the ones who remove some pit/irqchip related variables in kvm_context
Applied,
On 07/24/2009 08:46 PM, Glauber Costa wrote:
they have it, and it does the same as we do.
Applied, thanks.
--
error compiling committee.c: too many arguments to function
--
To unsubscribe from this list: send the line unsubscribe kvm in
the body of a message to majord...@vger.kernel.org
On 07/24/2009 08:46 PM, Glauber Costa wrote:
Some more cleanup from the libkvm era. Functions that starts with
kvm_qemu_yyy just to wrap a kvm_yyy. I'm removing it, and changing
the name of one of them to match upstream.
Applied, thanks.
--
error compiling committee.c: too many arguments
On 07/21/2009 09:51 AM, Gleb Natapov wrote:
On Tue, Jul 21, 2009 at 02:46:44PM +0800, Yang, Sheng wrote:
Happen to see this:
include/linux.kvm.h
503 #define KVM_IRQ_LINE_STATUS _IOWR(KVMIO, 0x67, struct kvm_irq_level)
504 #define KVM_REGISTER_COALESCED_MMIO \
505
On Wednesday 29 July 2009 10:10:13 am qemu-...@buildbot.b1-systems.de wrote:
BUILD FAILED: failed git
JFYI, i disabled now my two debian buildslaves until i found the issue with
rm -rf.
Avi's buildslave is still running ...
Sorry for the noise.
Best Regards,
Daniel
--
Daniel Gollub
On Wed, Jul 29, 2009 at 4:18 AM, Michael Goldishmgold...@redhat.com wrote:
I'm not entirely sure this is needed:
I wasn't sure either... I guess I ended up spending time to fix a
problem that didn't need to be fixed...
- How often do tests fail due to unsupported flags? I thought the -mem-path
On Wed, Jul 29, 2009 at 5:11 AM, Lukáš Doktorldok...@redhat.com wrote:
Hello Lucas,
I like your patch but I'm not entirely convinced about it necessity.
Stable version of KVM should have this fixed and unstable ones are for
developers, who are skilled enough to fix this using kvm_test.cfg.
On 07/16/2009 01:26 PM, Yolkfull Chow wrote:
Signed-off-by: Yolkfull Chowyz...@redhat.com
---
client/tests/kvm/kvm_vm.py | 11 +++
1 files changed, 11 insertions(+), 0 deletions(-)
diff --git a/client/tests/kvm/kvm_vm.py b/client/tests/kvm/kvm_vm.py
index 503f636..895049e 100644
On Thu, Jul 23, 2009 at 06:45:53PM -0300, Marcelo Tosatti wrote:
On Wed, Jul 22, 2009 at 11:53:26PM +0200, Jan Kiszka wrote:
Release and re-acquire preemption and IRQ lock in the same order as
vcpu_enter_guest does.
This should happen in vcpu_enter_guest, before it decides to disable
Avi Kivity wrote:
On 07/29/2009 01:17 PM, Jan Kiszka wrote:
I just wonder now if/when qemu-kvm will switch over to the
kvm-by-default-off policy of upstream?
That will surely inconvenience/surprise a lot of users.
A migration path could be:
- add -accel
- start warning when -accel
On Wed, Jul 29, 2009 at 9:18 AM, Avi Kivitya...@redhat.com wrote:
On 07/16/2009 01:26 PM, Yolkfull Chow wrote:
Signed-off-by: Yolkfull Chowyz...@redhat.com
---
client/tests/kvm/kvm_vm.py | 11 +++
1 files changed, 11 insertions(+), 0 deletions(-)
diff --git
On 07/29/2009 03:24 PM, Marcelo Tosatti wrote:
On Thu, Jul 23, 2009 at 06:45:53PM -0300, Marcelo Tosatti wrote:
On Wed, Jul 22, 2009 at 11:53:26PM +0200, Jan Kiszka wrote:
Release and re-acquire preemption and IRQ lock in the same order as
vcpu_enter_guest does.
This should
On 07/29/2009 03:28 PM, Alexander Graf wrote:
Avi Kivity wrote:
On 07/29/2009 01:17 PM, Jan Kiszka wrote:
I just wonder now if/when qemu-kvm will switch over to the
kvm-by-default-off policy of upstream?
That will surely inconvenience/surprise a lot of users.
A migration
On Wed, Jul 29, 2009 at 9:36 AM, Lucas Meneghel Rodriguesl...@redhat.com
wrote:
On Wed, Jul 29, 2009 at 9:18 AM, Avi Kivitya...@redhat.com wrote:
On 07/16/2009 01:26 PM, Yolkfull Chow wrote:
Signed-off-by: Yolkfull Chowyz...@redhat.com
---
client/tests/kvm/kvm_vm.py | 11 +++
1
- Lucas Meneghel Rodrigues l...@redhat.com wrote:
On Wed, Jul 29, 2009 at 9:18 AM, Avi Kivitya...@redhat.com wrote:
On 07/16/2009 01:26 PM, Yolkfull Chow wrote:
Signed-off-by: Yolkfull Chowyz...@redhat.com
---
client/tests/kvm/kvm_vm.py | 11 +++
1 files changed, 11
On Wed, Jul 29, 2009 at 9:46 AM, Michael Goldishmgold...@redhat.com wrote:
- Lucas Meneghel Rodrigues l...@redhat.com wrote:
On Wed, Jul 29, 2009 at 9:18 AM, Avi Kivitya...@redhat.com wrote:
On 07/16/2009 01:26 PM, Yolkfull Chow wrote:
Signed-off-by: Yolkfull Chowyz...@redhat.com
When caching guest intercepts there is no need anymore for the
nested_svm_exit_handled_real function. So move its code into
nested_svm_exit_handled.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 108 +++-
1 files
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 19 ---
1 files changed, 8 insertions(+), 11 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 381ed38..f81de35 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -1366,18
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 36 ++--
1 files changed, 34 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index b85da3f..0a0b3e1 100644
--- a/arch/x86/kvm/svm.c
+++
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 30 +++---
1 files changed, 23 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 31467b1..9192c9a 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 26 --
1 files changed, 24 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 1a44e43..381ed38 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@
Hi,
here is a first round of patches to clean up the code for SVM virtualization
for KVM on AMD. There is more to clean up but since I am on vacation the rest
of the week here is what I have queued up so far. Maybe this saves me from
rebasing this code next week ;-)
I tested these patches with
This makes it more clear for which purpose these members in the vcpu_svm
exist.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 62 +++
1 files changed, 33 insertions(+), 29 deletions(-)
diff --git a/arch/x86/kvm/svm.c
If this function returns true a nested vmexit is required. Move that
vmexit into the nested_svm_exit_handled function. This also simplifies
the handling of nested #pf intercepts in this function.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 41
On 07/29/2009 03:46 PM, Michael Goldish wrote:
I think I already commented on this and Yolkfull posted a new patch,
and that patch is the one that got applied. That's what I see in my
local tree anyway.
Yeah, I have a long backlog (and don't read all autotest patches
anyway). Sorry
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 31 +++
1 files changed, 23 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 27d26c3..9f69b25 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
On 07/29/2009 03:56 PM, Joerg Roedel wrote:
Hi,
here is a first round of patches to clean up the code for SVM virtualization
for KVM on AMD. There is more to clean up but since I am on vacation the rest
of the week here is what I have queued up so far. Maybe this saves me from
rebasing this
The interrupt completion code must run after nested exits are handled
because not injected interrupts or exceptions may be handled by the l1
guest first.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c |5 +++--
1 files changed, 3 insertions(+), 2 deletions(-)
diff
On Mon, Jul 27, 2009 at 6:10 PM, Lucas Meneghel Rodriguesl...@redhat.com
wrote:
This patch adds a small setup script to set up huge memory
pages during the kvm tests execution. Also, added hugepage setup to the
fc8_quick sample.
Applied.
Signed-off-by: Lukáš Doktor ldok...@redhat.com
On Wed, Jul 29, 2009 at 03:18:51PM +0300, Avi Kivity wrote:
On 07/16/2009 01:26 PM, Yolkfull Chow wrote:
Signed-off-by: Yolkfull Chowyz...@redhat.com
---
client/tests/kvm/kvm_vm.py | 11 +++
1 files changed, 11 insertions(+), 0 deletions(-)
diff --git
On 07/29/2009 03:56 PM, Joerg Roedel wrote:
Signed-off-by: Joerg Roedeljoerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 30 +++---
1 files changed, 23 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 31467b1..9192c9a 100644
---
On Wed, Jul 29, 2009 at 04:05:50PM +0300, Avi Kivity wrote:
On 07/29/2009 03:56 PM, Joerg Roedel wrote:
Hi,
here is a first round of patches to clean up the code for SVM virtualization
for KVM on AMD. There is more to clean up but since I am on vacation the rest
of the week here is what I
It is more efficient to copy only the relevant parts of the vmcb back to
the nested vmcb when we emulate an vmexit.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 68 +--
1 files changed, 33 insertions(+), 35
On Wed, Jul 29, 2009 at 04:13:35PM +0300, Avi Kivity wrote:
I don't see the benefit of this patch. Accessing the cache is just
as expensive as accessing the real vmcb.
The benefit is that we don't have to gup and map the nested vmcb just
for checking who will take the intercept. Another
Joerg Roedel wrote:
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 31 +++
1 files changed, 23 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 27d26c3..9f69b25 100644
--- a/arch/x86/kvm/svm.c
On 07/29/2009 04:13 PM, Joerg Roedel wrote:
On Wed, Jul 29, 2009 at 04:13:35PM +0300, Avi Kivity wrote:
I don't see the benefit of this patch. Accessing the cache is just
as expensive as accessing the real vmcb.
The benefit is that we don't have to gup and map the nested vmcb just
Joerg Roedel wrote:
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
Acked-by: Alexander Graf ag...@suse.de
--
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/majordomo-info.html
Joerg Roedel wrote:
It is more efficient to copy only the relevant parts of the vmcb back to
the nested vmcb when we emulate an vmexit.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
Acked-by: Alexander Graf ag...@suse.de
--
To unsubscribe from this list: send the line unsubscribe kvm
On Wed, Jul 29, 2009 at 09:06:25PM +0800, Yolkfull Chow wrote:
On Wed, Jul 29, 2009 at 03:18:51PM +0300, Avi Kivity wrote:
On 07/16/2009 01:26 PM, Yolkfull Chow wrote:
Signed-off-by: Yolkfull Chowyz...@redhat.com
---
client/tests/kvm/kvm_vm.py | 11 +++
1 files changed, 11
On 07/29/2009 04:09 PM, Joerg Roedel wrote:
kvm/svm: make nested svm debugging runtime configurable
Should be migrated to tracepoints.
Ok, I tried to align debugging to the mmu debugging code. But
tracepoints are better of course.
I plan to switch mmu debugging to
Joerg Roedel wrote:
Only copy the necessary parts of the vmcb save area on vmrun and save
precious time.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
Acked-by: Alexander Graf ag...@suse.de
--
To unsubscribe from this list: send the line unsubscribe kvm in
the body of a message to
On 07/29/2009 04:09 PM, Joerg Roedel wrote:
I seem to recall nested smp used to be broken. Am I mistaken, or do
these patches fix rather than clean up?
Nested SMP was broken in one of the early version of the nested SVM
code. In the current upstream version it works quite well.
In
Avi Kivity wrote:
On 07/29/2009 04:09 PM, Joerg Roedel wrote:
I seem to recall nested smp used to be broken. Am I mistaken, or do
these patches fix rather than clean up?
Nested SMP was broken in one of the early version of the nested SVM
code. In the current upstream version it works
Joerg Roedel wrote:
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
What's the reason behind this?
Acked-by: Alexander Graf ag...@suse.de
--
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
On Wed, Jul 29, 2009 at 04:32:11PM +0300, Avi Kivity wrote:
On 07/29/2009 04:09 PM, Joerg Roedel wrote:
I seem to recall nested smp used to be broken. Am I mistaken, or do
these patches fix rather than clean up?
Nested SMP was broken in one of the early version of the nested SVM
code. In
On Wed, Jul 29, 2009 at 03:01:56PM +0300, Avi Kivity wrote:
On 07/24/2009 08:57 AM, Glauber Costa wrote:
Creation of those devices can be a lot simpler. In particular,
if we start by assuming they will be on, and disabling later if
anything fails (user request counts as failing here), we can
On 07/29/2009 04:38 PM, Joerg Roedel wrote:
In that case, the next interesting target is nested npt. This may
actually make nsvm perform well. Any plans?
Yes. Plan is to implement this when I am through with the cleanups.
Great, this promises to be the most interesting area of
On 07/29/2009 04:50 PM, Glauber Costa wrote:
On Wed, Jul 29, 2009 at 03:01:56PM +0300, Avi Kivity wrote:
On 07/24/2009 08:57 AM, Glauber Costa wrote:
Creation of those devices can be a lot simpler. In particular,
if we start by assuming they will be on, and disabling later if
On Wed, Jul 29, 2009 at 03:15:11PM +0200, Alexander Graf wrote:
Joerg Roedel wrote:
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 31 +++
1 files changed, 23 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kvm/svm.c
Joerg Roedel wrote:
The interrupt completion code must run after nested exits are handled
because not injected interrupts or exceptions may be handled by the l1
guest first.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
If you say so, I believe you :-).
Acked-by: Alexander Graf
On Wed, Jul 29, 2009 at 03:30:01PM +0200, Alexander Graf wrote:
Joerg Roedel wrote:
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
What's the reason behind this?
Acked-by: Alexander Graf ag...@suse.de
Reason is to not copy the reserved fields of the control area. These
fields sum
On Wed, Jul 29, 2009 at 04:26:02PM +0300, Avi Kivity wrote:
On 07/29/2009 04:13 PM, Joerg Roedel wrote:
On Wed, Jul 29, 2009 at 04:13:35PM +0300, Avi Kivity wrote:
I don't see the benefit of this patch. Accessing the cache is just
as expensive as accessing the real vmcb.
The benefit is
Joerg Roedel wrote:
On Wed, Jul 29, 2009 at 03:30:01PM +0200, Alexander Graf wrote:
Joerg Roedel wrote:
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
What's the reason behind this?
Acked-by: Alexander Graf ag...@suse.de
Reason is to not copy the reserved
Joerg Roedel wrote:
This makes it more clear for which purpose these members in the vcpu_svm
exist.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
Nice idea!
Acked-by: Alexander Graf ag...@suse.de
--
To unsubscribe from this list: send the line unsubscribe kvm in
the body of a message
Joerg Roedel wrote:
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 30 +++---
1 files changed, 23 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 31467b1..9192c9a 100644
--- a/arch/x86/kvm/svm.c
On Wed, Jul 29, 2009 at 03:50:39PM +0200, Alexander Graf wrote:
Joerg Roedel wrote:
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
arch/x86/kvm/svm.c | 30 +++---
1 files changed, 23 insertions(+), 7 deletions(-)
diff --git a/arch/x86/kvm/svm.c
On Wed, Jul 29, 2009 at 04:48:23PM +0300, Avi Kivity wrote:
On 07/29/2009 04:38 PM, Joerg Roedel wrote:
In that case, the next interesting target is nested npt. This may
actually make nsvm perform well. Any plans?
Yes. Plan is to implement this when I am through with the cleanups.
On 07/29/2009 04:48 PM, Joerg Roedel wrote:
Performance isn't too bad even without this. At least if the host
supports nested paging :) But shadow nested paging should give nested
svm a real boost, thats very true.
Well, any workload that benefits from npt would benefit from nnpt a lot
Joerg Roedel wrote:
When caching guest intercepts there is no need anymore for the
nested_svm_exit_handled_real function. So move its code into
nested_svm_exit_handled.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
Acked-by: Alexander Graf ag...@suse.de
--
To unsubscribe from this
On Wed, Jul 29, 2009 at 04:59:48PM +0300, Avi Kivity wrote:
On 07/29/2009 04:48 PM, Joerg Roedel wrote:
Performance isn't too bad even without this. At least if the host
supports nested paging :) But shadow nested paging should give nested
svm a real boost, thats very true.
Well, any
On 07/29/2009 04:58 PM, Joerg Roedel wrote:
On Wed, Jul 29, 2009 at 04:59:48PM +0300, Avi Kivity wrote:
On 07/29/2009 04:48 PM, Joerg Roedel wrote:
Performance isn't too bad even without this. At least if the host
supports nested paging :) But shadow nested paging should give nested
On Wed, Jul 29, 2009 at 03:44:20PM +0300, Avi Kivity wrote:
On 07/29/2009 03:24 PM, Marcelo Tosatti wrote:
On Thu, Jul 23, 2009 at 06:45:53PM -0300, Marcelo Tosatti wrote:
On Wed, Jul 22, 2009 at 11:53:26PM +0200, Jan Kiszka wrote:
Release and re-acquire preemption and IRQ lock in
Joerg Roedel wrote:
If this function returns true a nested vmexit is required. Move that
vmexit into the nested_svm_exit_handled function. This also simplifies
the handling of nested #pf intercepts in this function.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
Acked-by: Alexander
Joerg Roedel wrote:
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
My poor debug output ... :)
Acked-by: Alexander Graf ag...@suse.de
--
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
On Wed, Jul 29, 2009 at 11:07:16AM -0300, Marcelo Tosatti wrote:
On Wed, Jul 29, 2009 at 03:44:20PM +0300, Avi Kivity wrote:
On 07/29/2009 03:24 PM, Marcelo Tosatti wrote:
On Thu, Jul 23, 2009 at 06:45:53PM -0300, Marcelo Tosatti wrote:
On Wed, Jul 22, 2009 at 11:53:26PM +0200, Jan
On 07/29/2009 05:20 PM, Alexander Graf wrote:
Joerg Roedel wrote:
Signed-off-by: Joerg Roedeljoerg.roe...@amd.com
Acked-by: Alexander Grafag...@suse.de
I'm really starting to miss comments on the commits ...
Me too. Why ack then?
Joerg, not only is this interesting, it is
Avi Kivity wrote:
On 07/29/2009 05:20 PM, Alexander Graf wrote:
Joerg Roedel wrote:
Signed-off-by: Joerg Roedeljoerg.roe...@amd.com
Acked-by: Alexander Grafag...@suse.de
I'm really starting to miss comments on the commits ...
Me too. Why ack then?
Because the patch is
* Lucas Meneghel Rodrigues l...@redhat.com [2009-07-28 22:40]:
This patch adds a small setup script to set up huge memory
pages during the kvm tests execution. Also, added hugepage setup to the
fc8_quick sample.
Signed-off-by: Luká?? Doktor ldok...@redhat.com
Signed-off-by: Lucas Meneghel
Alexander Graf wrote:
Avi Kivity wrote:
On 07/29/2009 01:17 PM, Jan Kiszka wrote:
I just wonder now if/when qemu-kvm will switch over to the
kvm-by-default-off policy of upstream?
That will surely inconvenience/surprise a lot of users.
A migration path could be:
- add -accel
- start
On 07/29/2009 05:41 PM, Alexander Graf wrote:
Me too. Why ack then?
Because the patch is fine. It's just the description that's missing.
Shouldn't I ack when the only thing I don't like is the description?
The description is part of the patch, an important one.
--
error compiling
Avi Kivity wrote:
On 07/29/2009 05:41 PM, Alexander Graf wrote:
Me too. Why ack then?
Because the patch is fine. It's just the description that's missing.
Shouldn't I ack when the only thing I don't like is the description?
The description is part of the patch, an important
On Wed, Jul 29, 2009 at 04:43:11PM +0200, Jan Kiszka wrote:
Alexander Graf wrote:
Avi Kivity wrote:
On 07/29/2009 01:17 PM, Jan Kiszka wrote:
I just wonder now if/when qemu-kvm will switch over to the
kvm-by-default-off policy of upstream?
That will surely inconvenience/surprise a
On 07/29/2009 06:50 PM, Glauber Costa wrote:While I do understand the
value of backwards compatibility, we did change
behaviour of a number of things in the past. Example: qemu would print a help
message,
and now it runs without any disks. It changed people's script already.
There's a
Michael S. Tsirkin wrote:
Here's an untested patch with partial support for level triggered
interrupts in irqfd. What this patch has: support for clearing interrupt
on ack. What this patch does not have: support signalling eventfd on ack
so that userspace can take action and e.g. reenable
Gregory Haskins wrote:
Michael S. Tsirkin wrote:
Here's an untested patch with partial support for level triggered
interrupts in irqfd. What this patch has: support for clearing interrupt
on ack. What this patch does not have: support signalling eventfd on ack
so that userspace can take
On Mon, Jun 8, 2009 at 2:53 PM, Anthony Liguorialigu...@us.ibm.com wrote:
Anthony Liguori wrote:
This is needed for VMware VGA to work properly under KVM.
Signed-off-by: Anthony Liguori aligu...@us.ibm.com
---
hw/vmware_vga.c | 4
1 files changed, 4 insertions(+), 0 deletions(-)
Jordan Justen wrote:
On Mon, Jun 8, 2009 at 2:53 PM, Anthony Liguorialigu...@us.ibm.com wrote:
Anthony Liguori wrote:
This is needed for VMware VGA to work properly under KVM.
Signed-off-by: Anthony Liguori aligu...@us.ibm.com
---
hw/vmware_vga.c |4
1 files changed, 4
Upstream gained fields for irqchip_in_kernel and pit_in_kernel
in KVMState. Remove them from kvm_context, and use them.
Since we are still copying the KVMState structure, add those fields
in our copy too.
Signed-off-by: Glauber Costa glom...@redhat.com
---
kvm-all.c |2 ++
In this patch series, I am folding everything (most things were already
available)
from vcpu context into CPUState. After that is done, I'm able to start using
upstream's kvm_vcpu_ioctl, which needs a few tweaks, whenever we test for -1 as
a
return value.
I, however, did not remove references
remove it from kvm_context too. Besides it all,
helps catching any missing conversion
Signed-off-by: Glauber Costa glom...@redhat.com
---
qemu-kvm.c | 14 ++
qemu-kvm.h |2 --
2 files changed, 6 insertions(+), 10 deletions(-)
diff --git a/qemu-kvm.c b/qemu-kvm.c
index
all they did was to call a qemu function. Call this function instead.
Signed-off-by: Glauber Costa glom...@redhat.com
---
qemu-kvm-x86.c |7 +--
qemu-kvm.c | 34 --
2 files changed, 9 insertions(+), 32 deletions(-)
diff --git a/qemu-kvm-x86.c
this is only to make integration easier.
Signed-off-by: Glauber Costa glom...@redhat.com
---
qemu-kvm.c |1 +
qemu-kvm.h |1 +
2 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/qemu-kvm.c b/qemu-kvm.c
index 65042d2..ba937ca 100644
--- a/qemu-kvm.c
+++ b/qemu-kvm.c
@@ -448,6
1 - 100 of 111 matches
Mail list logo