- Original Message -
> On Wed, Sep 7, 2011 at 3:07 AM, Amos Kong wrote:
> > System services on guest and host take uncertain resource, it
> > effects
> > the perf results. We can use the below two scripts to disable some
> > services of host and guest.
> >
> > stop_serivices_perf.sh is use
This patchset is against https://github.com/avikivity/kvm.git next branch.
In this version, some changes come from Avi's comments:
- fix instruction retried for nested guest
- skip write-flooding for the sp whose level is 1
- rename some functions
--
To unsubscribe from this list: send the line "u
kvm_mmu_pte_write is unsafe since we need to alloc pte_list_desc in the
function when spte is prefetched, unfortunately, we can not know how many
spte need to be prefetched on this path, that means we can use out of the
free pte_list_desc object in the cache, and BUG_ON() is triggered, also some
p
The idea is from Avi:
| tag instructions that are typically used to modify the page tables, and
| drop shadow if any other instruction is used.
| The list would include, I'd guess, and, or, bts, btc, mov, xchg, cmpxchg,
| and cmpxchg8b.
This patch is used to tag the instructions and in the later p
Remove the same code between emulator_pio_in_emulated and
emulator_pio_out_emulated
Signed-off-by: Xiao Guangrong
---
arch/x86/kvm/x86.c | 59 ++-
1 files changed, 26 insertions(+), 33 deletions(-)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/
In current code, the accessed bit is always set when page fault occurred,
do not need to set it on pte write path
Signed-off-by: Xiao Guangrong
---
arch/x86/include/asm/kvm_host.h |1 -
arch/x86/kvm/mmu.c | 22 +-
2 files changed, 1 insertions(+), 22 deleti
Directly Use mmu_page_zap_pte to zap spte in FNAME(invlpg), also remove the
same code between FNAME(invlpg) and FNAME(sync_page)
Signed-off-by: Xiao Guangrong
---
arch/x86/kvm/mmu.c | 16 ++--
arch/x86/kvm/paging_tmpl.h | 44 +---
2
Fast prefetch spte for the unsync shadow page on invlpg path
Signed-off-by: Xiao Guangrong
---
arch/x86/include/asm/kvm_host.h |4 +---
arch/x86/kvm/mmu.c | 38 +++---
arch/x86/kvm/paging_tmpl.h | 30 ++
arch/x
>From cab4eb79efc498abbda19c5b10c7d0858349af5f Mon Sep 17 00:00:00 2001
From: Liu, Jinsong
Date: Thu, 22 Sep 2011 09:49:05 +0800
Subject: [PATCH 1/2] Add some pre-defination
This pre-defination is preparing for KVM tsc deadline timer emulation, but
theirself are no-kvm-specific.
Signed-off-by:
In kvm_mmu_pte_write, we do not need to alloc shadow page, so calling
kvm_mmu_free_some_pages is really unnecessary
Signed-off-by: Xiao Guangrong
---
arch/x86/kvm/mmu.c |1 -
1 files changed, 0 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 4128aba.
kvm_mmu_pte_write is too long, we split it for better readable
Signed-off-by: Xiao Guangrong
---
arch/x86/kvm/mmu.c | 194
1 files changed, 119 insertions(+), 75 deletions(-)
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 7b22f3a
Sometimes, we only modify the last one byte of a pte to update status bit,
for example, clear_bit is used to clear r/w bit in linux kernel and 'andb'
instruction is used in this function, in this case, kvm_mmu_pte_write will
treat it as misaligned access, and the shadow page table is zapped
Signed
>From 4d5b83aba40ce0d421add9a41a6c591a8590a32e Mon Sep 17 00:00:00 2001
From: Liu, Jinsong
Date: Thu, 22 Sep 2011 14:00:08 +0800
Subject: [PATCH 2/2] KVM: emulate lapic tsc deadline timer for guest
This patch emulate lapic tsc deadline timer for guest:
Enumerate tsc deadline timer capability by C
Detecting write-flooding does not work well, when we handle page written, if
the last speculative spte is not accessed, we treat the page is
write-flooding, however, we can speculative spte on many path, such as pte
prefetch, page synced, that means the last speculative spte may be not point
to the
>From 8c39f2ddbf7069342826a83e535c0c7b641d6501 Mon Sep 17 00:00:00 2001
From: Liu, Jinsong
Date: Thu, 22 Sep 2011 16:28:13 +0800
Subject: [PATCH] Qemu co-operation with kvm tsc deadline timer
KVM add emulation of lapic tsc deadline timer for guest.
This patch is co-operation work at qemu side.
S
If the emulation is caused by #PF and it is non-page_table writing instruction,
it means the VM-EXIT is caused by shadow page protected, we can zap the shadow
page and retry this instruction directly
The idea is from Avi
Signed-off-by: Xiao Guangrong
---
arch/x86/include/asm/kvm_emulate.h |
On 2011-09-22 05:12, Alex Williamson wrote:
> We now need to scan PCI capabilities and setup an MSI-X page
> before we walk the device resources since the overlay is now
> setup during init instead of at the first mapping by the guest.
>
> Signed-off-by: Alex Williamson
> ---
>
> hw/device-assi
On 2011-09-22 05:12, Alex Williamson wrote:
> Commit c4525754 added a capability check for KVM_CAP_DEVICE_MSIX,
> which is unfortunately not exposed, resulting in MSIX never
> being listed as a capability.
Oops. Should we fix this nevertheless in the kernel?
> This breaks anything depending on
>
Hi to all,
I'm interested in virtualization support for ARM architectures.
After googling, examining /proc/cpuinfo and reading ARM related posts on
this ml I still don't know
the level of hardware virtualization support of the many ARM CPUs.
Could anyone please point me to some detailed reso
From: KAMEZAWA Hiroyuki
Subject: [PATCH] Fix inject-nmi
Now, inject-nmi sends NMI to all cpus...but this doesn't emulate
pc hardware 'NMI button', which triggers LINT1.
So, now, LINT1 mask is ignored by inject-nmi and NMIs are sent to
all cpus without checking LINT1 mask.
Because Linux masks L
Hi Christian,
On (Wed) 21 Sep 2011 [17:52:23], Christian Borntraeger wrote:
> Amit,
>
> can you have a look at the patch below and give feedback or apply
> if appropriate?
The patch looks good. Just a couple of comments:
> On s390 I have seen some random "Warning: unable to open an initial
> c
The following two patches solve two injection-related nested VMX issues:
1. When we must run L2 next (namely on L1's VMLAUNCH/VMRESUME), injection
into L1 was delayed for an unknown amount of time - until L2 exits.
We now force (using a self IPI) an exit immediately after entry to L2,
This patch adds a new vcpu->requests bit, KVM_REQ_IMMEDIATE_EXIT.
This bit requests that when next entering the guest, we should run it only
for as little as possible, and exit again.
We use this new option in nested VMX: When L1 launches L2, but L0 wishes L1
to continue running so it can inject a
When L0 wishes to inject an interrupt while L2 is running, it emulates an exit
to L1 with EXIT_REASON_EXTERNAL_INTERRUPT. This was explained in the original
nVMX patch 23, titled "Correct handling of interrupt injection".
Unfortunately, it is possible (though rare) that at this point there is vali
On 22/09/11 12:08, Amit Shah wrote:
>> +/* If there was an early virtio console, assume that there are no
>> + * other consoles. We need to wait until the hvc_alloc matches the
>> + * hvc_instantiate, otherwise tty_open will complain, resulting in
>> + * a "Warning: unable to open a
On (Thu) 22 Sep 2011 [13:20:07], Christian Borntraeger wrote:
> On 22/09/11 12:08, Amit Shah wrote:
> >> + /* If there was an early virtio console, assume that there are no
> >> + * other consoles. We need to wait until the hvc_alloc matches the
> >> + * hvc_instantiate, otherwise tty_open wil
This is a second iteration of the patch. The patch has been
significantly reworked to address (offline) comments by Gleb.
I think the infrastructure created is generic enough
to be generally useful beyond the specific bug
that I would like to fix. Specifically it
will be able to find S3 Name to p
Use iasl -l flag to produce a mixed listing, where a
source line is followed by matching AML.
Add a tool tools/acpi_extract.py to process this
listing. The tool looks for ACPI_EXTRACT tags
in the ASL source and outputs matching AML offsets
in an array.
To make these directives pass through ASL wi
Modify ACPI to only supply _EJ0 methods for PCI
slots that support hotplug.
This is done by runtime patching:
- Instrument ASL code with ACPI_EXTRACT directives
tagging _EJ0 and _ADR fields.
- At compile time, tools/acpi_extract.py looks for these methods
in ASL source finds the matching AML,
The macro gen_pci_device is used to add _RMV
method to a slot device so it is no longer needed:
presence of _EJ0 now indicates that the slot is ejectable.
It is also placing two devices with the same _ADR
on the same bus, which isn't defined by the ACPI spec.
So let's remove it.
Signed-off-by: Mic
Recent (3.0+) Linux guests check for the fast string bit in
MSR_IA32_MISC_ENABLE before enabling rep/movs based memcpy and
related on fam 6/model 13+ processors.
Enable the bit by default, as required by the specification.
Signed-off-by: Avi Kivity
---
arch/x86/kvm/x86.c |2 ++
1 files chan
Hi,
What is the fastest way to exchange data between host and guest?
Thanks
Anjali
--
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
On s390 I have seen some random "Warning: unable to open an initial
console" boot failure. Turns out that tty_open fails, because the
hvc_alloc was not yet done. In former times this could not happen,
since the probe function automatically called hvc_alloc. With newer
versions (multiport) some host
On Thu, 2011-09-22 at 05:04 -0700, Anjali Kulkarni wrote:
> Hi,
>
> What is the fastest way to exchange data between host and guest?
>
> Thanks
> Anjali
With regards to speed, I guess it's the ivshmem device.
--
Sasha.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the
On Thu, Sep 22, 2011 at 09:09:49AM +0300, Michael S. Tsirkin wrote:
> On Thu, Sep 22, 2011 at 12:35:13AM -0400, Kevin O'Connor wrote:
> > On Wed, Sep 21, 2011 at 03:44:13PM +0300, Michael S. Tsirkin wrote:
> > > The correct way to suppress hotplug is not to have _EJ0,
> > > so this is what this pat
On Thu, Sep 22, 2011 at 04:55:52PM +0800, Liu, Jinsong wrote:
> >From 4d5b83aba40ce0d421add9a41a6c591a8590a32e Mon Sep 17 00:00:00 2001
> From: Liu, Jinsong
> Date: Thu, 22 Sep 2011 14:00:08 +0800
> Subject: [PATCH 2/2] KVM: emulate lapic tsc deadline timer for guest
>
> This patch emulate lapic
On 2011-09-22 11:50, Lai Jiangshan wrote:
>
> From: KAMEZAWA Hiroyuki
> Subject: [PATCH] Fix inject-nmi
>
> Now, inject-nmi sends NMI to all cpus...but this doesn't emulate
> pc hardware 'NMI button', which triggers LINT1.
>
> So, now, LINT1 mask is ignored by inject-nmi and NMIs are sent to
>
Marcelo Tosatti wrote:
> On Thu, Sep 22, 2011 at 04:55:52PM +0800, Liu, Jinsong wrote:
>>> From 4d5b83aba40ce0d421add9a41a6c591a8590a32e Mon Sep 17 00:00:00
>>> 2001
>> From: Liu, Jinsong
>> Date: Thu, 22 Sep 2011 14:00:08 +0800
>> Subject: [PATCH 2/2] KVM: emulate lapic tsc deadline timer for gu
All,
Sorry if I am not keeping up on the subject but wanted to know whether
there is any effort going on for inter VM communication / PF-VF
communication (in case of SR-IOV)
I see that most of SR-IOV capable NIC supports mailboxes for that
purpose to avoid the security hole.
Xen has virtual device
Hi guys,
Do you remember the discussion about cgroup testing in autotest vs. LTP? I hope
there won't be any doubts about this one as ground_test (+ first 2 subtests)
are strictly focused on cgroups features enforced on KVM guest systems. Also
more subtests will follow if you approve the test st
[FIX] incorrect prop/dir variable usage
[MOD] Use __del__() instead of cleanup - Simplifies the code with small
drawback (failures can't be handled. Anyway, they are not critical and were
never handled before...)
Signed-off-by: Lukas Doktor
---
client/tests/cgroup/cgroup_common.py | 41 +
basic structure:
* similar to general client/tests/cgroup/ test (imports from the
cgroup_common.py)
* uses classes for better handling
* improved logging and error handling
* checks/repair the guests after each subtest
* subtest mapping is specified in test dictionary in cgroup.py
* allows
Assigned device MSI-X support hasn't been working, this fixes
it. I believe this should also fix:
https://bugs.launchpad.net/qemu/+bug/830558
v2:
Incorporate comments from Jan
WRT exposing KVM_CAP_DEVICE_MSIX, I'll send a patch with a big
comment noting that we can't rely on it for older ker
We now need to scan PCI capabilities and setup an MSI-X page
before we walk the device resources since the overlay is now
setup during init instead of at the first mapping by the guest.
Signed-off-by: Alex Williamson
---
hw/device-assignment.c | 19 +++
1 files changed, 11 ins
Commit c4525754 added a capability check for KVM_CAP_DEVICE_MSIX,
which is unfortunately not exposed, resulting in MSIX never
being listed as a capability. This breaks anything depending on
MSIX, such as igbvf. Instead let's use a dummy call to
KVM_ASSIGN_SET_MSIX_NR which will return -EFAULT if
Hi Rusty,
This is a fix from Christian for early console handling with multiport
support. Please apply.
Christian, I've made some changes to the patch as noted in the commit
message. Nothing major, but an ACK would be nice.
Thanks.
Christian Borntraeger (1):
virtio: console: wait for first
From: Christian Borntraeger
On s390 I have seen some random
"Warning: unable to open an initial console"
boot failure. Turns out that tty_open fails, because the
hvc_alloc was not yet done. In former times this could not happen,
since the probe function automatically called hvc_alloc. With newe
Hi all,
is there an update regarding simultaneous use of VNC and VGA output?
Thanks.
Best regards,
Erik
Bitman Zhou wrote:
We tried before with both spice(QXL) and VNC enabled at the same time
for the same VM. It works a little bit, I mean VNC session can hold some
time. I use gtk-vnc and
On 22/09/11 20:14, Amit Shah wrote:
> Hi Rusty,
>
> This is a fix from Christian for early console handling with multiport
> support. Please apply.
>
> Christian, I've made some changes to the patch as noted in the commit
> message. Nothing major, but an ACK would be nice.
The changes look fin
On Thu, Sep 22, 2011 at 11:22:02PM +0800, Liu, Jinsong wrote:
> Marcelo Tosatti wrote:
> > On Thu, Sep 22, 2011 at 04:55:52PM +0800, Liu, Jinsong wrote:
> >>> From 4d5b83aba40ce0d421add9a41a6c591a8590a32e Mon Sep 17 00:00:00
> >>> 2001
> >> From: Liu, Jinsong
> >> Date: Thu, 22 Sep 2011 14:00:08
On 09/22/2011 10:23 AM, Sagar Borikar wrote:
All,
Sorry if I am not keeping up on the subject but wanted to know whether
there is any effort going on for inter VM communication / PF-VF
communication (in case of SR-IOV)
I see that most of SR-IOV capable NIC supports mailboxes for that
purpose to
> I'm not aware of any vendor these days that actually requires a PV driver
> for PF-VF communications. I know some toyed with the idea years ago but I
> thought malboxes have become defacto standard.
May be because Intel started the mailbox implementation ;). But just
wondering from hypervisor po
On Thu, 22 Sep 2011 20:59:13 +0200, Christian Borntraeger
wrote:
> On 22/09/11 20:14, Amit Shah wrote:
> > Hi Rusty,
> >
> > This is a fix from Christian for early console handling with multiport
> > support. Please apply.
> >
> > Christian, I've made some changes to the patch as noted in the
53 matches
Mail list logo
