Checks and operations on the INVPCID feature bit should use EBX of CPUID leaf 7
instead of ECX.
Changes from v1:
- fixed patch format issue.
Signed-off-by: Junjie Mao
Signed-off-by: Yongjie Ren
---
arch/x86/kvm/vmx.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff
updated the head file from kernel for readonly memory
feature.
Signed-off-by: Liu Sheng
---
linux-headers/asm-x86/kvm.h |1 +
linux-headers/linux/kvm.h | 16 +---
2 files changed, 14 insertions(+), 3 deletions(-)
diff --git a/linux-headers/asm-x86/kvm.h b/linux-headers/asm-x
This patch set make the readonly memory in qemu really readonly, using
readonly memory slots feature in kvm, to make qemu-kvm safer. Memory
regions with readonly property would be plug into kvm as readonly memory
slots.
This patch set depends on the last patch "Introduce write_readonly_mem
in mm
make sure the memory regions of rom in pc have readonly property,
prepare for plug these memory to kvm as readonly memory slots.
Signed-off-by: Liu Sheng
---
hw/pc.c |1 +
hw/pci.c |1 +
2 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/hw/pc.c b/hw/pc.c
index c32ee8e..b7fb
as readonly memory is support in kvm, this patch supports this feature
in qemu, mainly pluging the memory region with KVM_MEM_READONLY flag
to kvm.
this can be tested by a kernel module from the author of kvm readonly
memory slot:
static int rom_tester_probe(struct pci_dev *dev, const struct pci_
On 2012-09-06 13:27, Tomoki Sekiyama wrote:
> This RFC patch series provides facility to dedicate CPUs to KVM guests
> and enable the guests to handle interrupts from passed-through PCI devices
> directly (without VM exit and relay by the host).
>
> With this feature, we can improve throughput and
On 2012-09-07 10:26, Liu Sheng wrote:
> updated the head file from kernel for readonly memory
> feature.
>
> Signed-off-by: Liu Sheng
> ---
> linux-headers/asm-x86/kvm.h |1 +
> linux-headers/linux/kvm.h | 16 +---
> 2 files changed, 14 insertions(+), 3 deletions(-)
>
> diff
On 2012-09-07 10:26, Liu Sheng wrote:
> as readonly memory is support in kvm, this patch supports this feature
> in qemu, mainly pluging the memory region with KVM_MEM_READONLY flag
> to kvm.
>
> this can be tested by a kernel module from the author of kvm readonly
> memory slot:
>
> static int r
Il 07/09/2012 08:39, Rusty Russell ha scritto:
>> > So it looks like a bug: we should teach driver to tell host first on leak?
>> > Yan, Vadim, can you comment please?
>> >
>> > Also if true, looks like this bit will be useful to detect a fixed driver
>> > on
>> > the hypervisor side - to avoid un
Hi Jan,
Thanks for your review.
On 09/07/2012 04:49 PM, Jan Kiszka wrote:
>> +#ifdef __KVM_HAVE_READONLY_MEM
>> +__u8 write_readonly_mem;
>> +#endif
>> } mmio;
>> /* KVM_EXIT_HYPERCALL */
>> struct {
>> @@ -618,6 +625,9 @@ struct kvm_ppc
On 09/06/2012 10:09 PM, Avi Kivity wrote:
> On 08/22/2012 03:47 PM, Xiao Guangrong wrote:
>> On 08/22/2012 08:06 PM, Avi Kivity wrote:
>>> On 08/21/2012 06:03 AM, Xiao Guangrong wrote:
Introduce write_readonly_mem in mmio-exit-info to indicate this exit is
caused by write access on readon
On 2012-09-07 11:39, Xiao Guangrong wrote:
> Hi Jan,
>
> Thanks for your review.
>
> On 09/07/2012 04:49 PM, Jan Kiszka wrote:
>
>>> +#ifdef __KVM_HAVE_READONLY_MEM
>>> + __u8 write_readonly_mem;
>>> +#endif
>>> } mmio;
>>> /* KVM_EXIT_HYPERCALL */
>>>
On 2012-08-21 05:02, Xiao Guangrong wrote:
> In current code, if we map a readonly memory space from host to guest
> and the page is not currently mapped in the host, we will get a fault
> pfn and async is not allowed, then the vm will crash
>
> We introduce readonly memory region to map ROM/ROMD
On Fri, Sep 07, 2012 at 04:09:50PM +0930, Rusty Russell wrote:
> > So it looks like a bug: we should teach driver to tell host first on leak?
> > Yan, Vadim, can you comment please?
> >
> > Also if true, looks like this bit will be useful to detect a fixed driver on
> > the hypervisor side - to avo
On 09/07/2012 06:23 PM, Jan Kiszka wrote:
> On 2012-08-21 05:02, Xiao Guangrong wrote:
>> In current code, if we map a readonly memory space from host to guest
>> and the page is not currently mapped in the host, we will get a fault
>> pfn and async is not allowed, then the vm will crash
>>
>> We i
On Fri, Sep 07, 2012 at 11:27:42AM +0200, Paolo Bonzini wrote:
> Il 07/09/2012 08:39, Rusty Russell ha scritto:
> >> > So it looks like a bug: we should teach driver to tell host first on
> >> > leak?
> >> > Yan, Vadim, can you comment please?
> >> >
> >> > Also if true, looks like this bit will b
On 2012-09-07 12:47, Xiao Guangrong wrote:
> On 09/07/2012 06:23 PM, Jan Kiszka wrote:
>> On 2012-08-21 05:02, Xiao Guangrong wrote:
>>> In current code, if we map a readonly memory space from host to guest
>>> and the page is not currently mapped in the host, we will get a fault
>>> pfn and async
Signed-off-by: Jan Kiszka
---
Documentation/virtual/kvm/api.txt | 22 +++---
1 files changed, 11 insertions(+), 11 deletions(-)
diff --git a/Documentation/virtual/kvm/api.txt
b/Documentation/virtual/kvm/api.txt
index b91bfd4..36befa7 100644
--- a/Documentation/virtual/kvm/api.
Il 07/09/2012 12:53, Michael S. Tsirkin ha scritto:
> Let us start with what is broken currently. Looking at
> it very closely, I think the answer is nothing.
> Even migration in qemu is not broken as you claimed initially.
Correct, migration would be broken as soon as QEMU starts using
MUST_TELL_
Avi Kivity wrote:
>> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
>> index 148ed66..0e64997 100644
>> --- a/arch/x86/kvm/x86.c
>> +++ b/arch/x86/kvm/x86.c
>> @@ -2210,7 +2210,11 @@ int kvm_dev_ioctl_check_extension(long ext)
>> r = kvm_has_tsc_control; break;
>> case KVM_
On Fri, Sep 07, 2012 at 01:20:57PM +0200, Paolo Bonzini wrote:
> Il 07/09/2012 12:53, Michael S. Tsirkin ha scritto:
> > Let us start with what is broken currently. Looking at
> > it very closely, I think the answer is nothing.
> > Even migration in qemu is not broken as you claimed initially.
>
>
Il 07/09/2012 14:17, Michael S. Tsirkin ha scritto:
>>> Next, consider the interface proposed here. You defacto declare
>>> all existing drivers buggy.
>>
>> No, only Windows (and it is buggy, it calls tell_host last).
>
> It is not buggy. It does not ack MUST_TELL_HOST. So it is free to tell
> ho
On Fri, Sep 07, 2012 at 02:22:47PM +0200, Paolo Bonzini wrote:
> Il 07/09/2012 14:17, Michael S. Tsirkin ha scritto:
> >>> Next, consider the interface proposed here. You defacto declare
> >>> all existing drivers buggy.
> >>
> >> No, only Windows (and it is buggy, it calls tell_host last).
> >
>
I have noticed recently that PLE/yield_to() is still not that scalable
for really large guests, sometimes even with no CPU over-commit. I have
a small change that make a very big difference.
First, let me explain what I saw:
Time to boot a 3.6-rc kernel in an 80-way VM on a 4 socket, 40 core, 80
Il 07/09/2012 14:44, Michael S. Tsirkin ha scritto:
>> Well, according to your reading of the spec.
>>
>> In the spec I'm reading "Host must be told before pages from the balloon
>> are used". Doesn't say anything about the guest.
>
> No? How is host told then? By divine force?
For a simple madv
On Fri, Sep 07, 2012 at 04:04:13PM +0200, Paolo Bonzini wrote:
> Il 07/09/2012 14:44, Michael S. Tsirkin ha scritto:
> >> Well, according to your reading of the spec.
> >>
> >> In the spec I'm reading "Host must be told before pages from the balloon
> >> are used". Doesn't say anything about the g
Il 07/09/2012 16:25, Michael S. Tsirkin ha scritto:
>> > Silent deflate is deflating just by using the page, without using the
>> > deflateq at all. So it can be done even from GFP_ATOMIC context.
> BTW I don't see a need to avoid deflateq.
> Without MUST_TELL_HOST you can just deflate and then
>
Il 07/09/2012 08:48, Nicholas A. Bellinger ha scritto:
> Cc: Stefan Hajnoczi
> Cc: Zhi Yong Wu
> Cc: Michael S. Tsirkin
> Cc: Paolo Bonzini
> Signed-off-by: Nicholas Bellinger
> ---
> hw/virtio-pci.c |2 ++
> hw/virtio-scsi.c | 49 +
> hw
CCing PeterZ also.
On 09/07/2012 06:41 PM, Andrew Theurer wrote:
I have noticed recently that PLE/yield_to() is still not that scalable
for really large guests, sometimes even with no CPU over-commit. I have
a small change that make a very big difference.
First, let me explain what I saw:
Tim
On Fri, 2012-09-07 at 18:00 +0200, Paolo Bonzini wrote:
> Il 07/09/2012 08:48, Nicholas A. Bellinger ha scritto:
> > Cc: Stefan Hajnoczi
> > Cc: Zhi Yong Wu
> > Cc: Michael S. Tsirkin
> > Cc: Paolo Bonzini
> > Signed-off-by: Nicholas Bellinger
> > ---
> > hw/virtio-pci.c |2 ++
> > hw/vi
On Fri, 2012-09-07 at 23:36 +0530, Raghavendra K T wrote:
> CCing PeterZ also.
>
> On 09/07/2012 06:41 PM, Andrew Theurer wrote:
> > I have noticed recently that PLE/yield_to() is still not that scalable
> > for really large guests, sometimes even with no CPU over-commit. I have
> > a small chang
Il 07/09/2012 21:23, Nicholas A. Bellinger ha scritto:
>> > Please create a completely separate device vhost-scsi-pci instead (or
>> > virtio-scsi-tcm-pci, or something like that). It is used completely
>> > differently from virtio-scsi-pci, it does not make sense to conflate the
>> > two.
>> >
>
On Thu, Aug 23, 2012 at 12:45 AM, Feng Tang wrote:
> From 57a28ee5e7662ca28ba4c793aa037d64bd082dee Mon Sep 17 00:00:00 2001
> From: Feng Tang
> Date: Wed, 22 Aug 2012 15:41:51 +0800
> Subject: [PATCH 1/2] PCI: Use local parameter pci_device_id for
> pci_get_subsys/class()
>
> This fixes a kernel
On Fri, Sep 7, 2012 at 6:00 PM, Yinghai Lu wrote:
> On Thu, Aug 23, 2012 at 12:45 AM, Feng Tang wrote:
>> From 57a28ee5e7662ca28ba4c793aa037d64bd082dee Mon Sep 17 00:00:00 2001
>> From: Feng Tang
>> Date: Wed, 22 Aug 2012 15:41:51 +0800
>> Subject: [PATCH 1/2] PCI: Use local parameter pci_device
On Fri, Sep 07, 2012 at 06:32:48PM -0700, Yinghai Lu wrote:
> Greg,
>
> Any reason for using kmalloc instead of local variable during your
> rewriting pci search code?
>
> commit 95247b57ed844511a212265b45cf9a919753aea1
> Author: Greg Kroah-Hartman
> Date: Wed Feb 13 11:03:58 2008 -0800
Serio
"Michael S. Tsirkin" writes:
> On Fri, Sep 07, 2012 at 04:09:50PM +0930, Rusty Russell wrote:
>> > So it looks like a bug: we should teach driver to tell host first on leak?
>> > Yan, Vadim, can you comment please?
>> >
>> > Also if true, looks like this bit will be useful to detect a fixed driver
36 matches
Mail list logo