vhost-blk is an in-kernel virito-blk device accelerator.
Due to lack of proper in-kernel AIO interface, this version converts
guest's I/O request to bio and use submit_bio() to submit I/O directly.
So this version any supports raw block device as guest's disk image,
e.g. /dev/sda, /dev/ram0. We ca
Hello ,
I wish to run some cpu intensive tasks on a vm , use only one core
specifically #0 of the 6 available ( Xeon 3670 ) and all the while
record basic perf events like cpu-cycles , instructions etc.
I am using taskset to define the cpu-affinity of the qemu process :
taskset -p 0x1 `pgrep
On Mon, Oct 08, 2012 at 12:51:25PM +1030, Rusty Russell wrote:
> (Topic updated, cc's trimmed).
>
> Anthony Liguori writes:
> > Rusty Russell writes:
> >> 4) The only significant change to the spec is that we use PCI
> >>capabilities, so we can have infinite feature bits.
> >>(see
> >>
On Tue, Oct 09, 2012 at 10:26:12AM +1030, Rusty Russell wrote:
> Anthony Liguori writes:
> > Gerd Hoffmann writes:
> >
> >> Hi,
> >>
> So we could have for virtio something like this:
>
> Capabilities: [??] virtio-regs:
> legacy: BAR=0 offset=0
>
Jamie Lokier writes:
> Rusty Russell wrote:
>> I don't think it'll be that bad; reset clears the device to unknown,
>> bar0 moves it from unknown->legacy mode, bar1/2/3 changes it from
>> unknown->modern mode, and anything else is bad (I prefer being strict so
>> we catch bad implementations from
Gerd Hoffmann writes:
> So how about this:
>
> (1) Add a vendor specific pci capability for new-style virtio.
> Specifies the pci bar used for new-style virtio registers.
> Guests can use it to figure whenever new-style virtio is
> supported and to map the correct bar (which will proba
Cornelia Huck writes:
> On Thu, 27 Sep 2012 09:59:33 +0930
> Rusty Russell wrote:
>> 3) Various clarifications, formalizations and cleanups to the spec text,
>>and possibly elimination of old deprecated features.
>>
>> 4) The only significant change to the spec is that we use PCI
>>capab
Anthony Liguori writes:
> Rusty Russell writes:
>
>> Anthony Liguori writes:
>>> We'll never remove legacy so we shouldn't plan on it. There are
>>> literally hundreds of thousands of VMs out there with the current virtio
>>> drivers installed in them. We'll be supporting them for a very, very
On Wed, Oct 10, 2012 at 01:22:22PM +0530, Jaspal wrote:
> Hello ,
>
> I wish to run some cpu intensive tasks on a vm , use only one core
> specifically #0 of the 6 available ( Xeon 3670 ) and all the while
> record basic perf events like cpu-cycles , instructions etc.
>
> I am using taskset to de
On Wed, Oct 10, 2012 at 02:14:11PM +1030, Rusty Russell wrote:
> > See above. A guest could happily just use BAR1/BAR2 and completely
> > ignore BAR0 provided that BAR1/BAR2 are present.
>
> But x86 guests want BAR0 because IO space is faster than MMIO. Right?
Or to be a bit more precise, ATM o
On 10/10/2012 03:38 PM, Gleb Natapov wrote:
On Wed, Oct 10, 2012 at 01:22:22PM +0530, Jaspal wrote:
Hello ,
I wish to run some cpu intensive tasks on a vm , use only one core
specifically #0 of the 6 available ( Xeon 3670 ) and all the while
record basic perf events like cpu-cycles , instructio
On Tue, Oct 09, 2012 at 04:10:30PM +, Auld, Will wrote:
> I am just testing the second version of this patch. It addresses all the
> comments so far except Marcelo's issue with breaking the function
> compute_guest_tsc().
Lets try to merge the missing patch from Zachary first (that'll make
On 10/09/12 15:11, Marcelo Tosatti wrote:
On Tue, Oct 09, 2012 at 03:09:17PM -0400, Don Slutz wrote:
On 10/09/12 14:47, Marcelo Tosatti wrote:
On Tue, Oct 09, 2012 at 03:27:26PM -0300, Marcelo Tosatti wrote:
On Tue, Oct 09, 2012 at 02:18:05PM -0300, Marcelo Tosatti wrote:
On Mon, Sep 24, 2012
On 10/09/12 15:01, Marcelo Tosatti wrote:
On Mon, Sep 24, 2012 at 10:32:08AM -0400, Don Slutz wrote:
Also known as Paravirtualization level.
This change is based on:
Microsoft Hypervisor CPUID Leaves:
http://msdn.microsoft.com/en-us/library/windows/hardware/ff542428%28v=vs.85%29.aspx
Linu
Rusty Russell writes:
> Gerd Hoffmann writes:
>> So how about this:
>>
>> (1) Add a vendor specific pci capability for new-style virtio.
>> Specifies the pci bar used for new-style virtio registers.
>> Guests can use it to figure whenever new-style virtio is
>> supported and to map t
On Wed, Oct 10, 2012 at 08:36:12AM -0500, Anthony Liguori wrote:
> Rusty Russell writes:
>
> > Gerd Hoffmann writes:
> >> So how about this:
> >>
> >> (1) Add a vendor specific pci capability for new-style virtio.
> >> Specifies the pci bar used for new-style virtio registers.
> >> Guest
On Wed, Oct 10, 2012 at 09:03:20AM -0400, Don Slutz wrote:
> >>Did you mean 3 patch sets (or more)? Or just a different order?
> >>-Don Slutz
> >Different order. Patches should be logically related (think of what
> >information the reviewer needs). Please write changelogs for
> >every patch.
>
I ran 'perf sched map' on the dbench workload for medium and large VMs,
and I thought I would share some of the results. I think it helps to
visualize what's going on regarding the yielding.
These files are png bitmaps, generated from processing output from 'perf
sched map' (and perf data generat
Debugging crash, panics, stack trace WARN_ONs, etc., from both virtual and
bare-metal boots can get difficult very quickly. While there are ways to
decipher the output and determine if the output is from a virtual guest,
the in-kernel hypervisors now have a single registration point and set
x86_hy
On Sun, Oct 07, 2012 at 08:01:34PM +0800, Xiao Guangrong wrote:
> We can not directly call kvm_release_pfn_clean to release the pfn
> since we can meet noslot pfn which is used to cache mmio info into
> spte
>
> Signed-off-by: Xiao Guangrong
> ---
> arch/x86/kvm/mmu.c |3 +--
> virt/kvm/kvm
On 10/09/12 15:13, Don Slutz wrote:
On 10/09/12 12:25, Marcelo Tosatti wrote:
On Mon, Sep 24, 2012 at 10:32:05AM -0400, Don Slutz wrote:
These are modeled after x86_cpuid_get_xlevel and x86_cpuid_set_xlevel.
Signed-off-by: Don Slutz
---
target-i386/cpu.c | 29 +
Add support for saving and restoring VFP register state from the
kernel. This includes a check that the KVM-created CPU has full
VFP support (as the TCG Cortex-A15 model always does), since for
the moment ARM QEMU doesn't have any way to tweak optional features
on created CPUs.
Signed-off-by: Pete
From: Christoffer Dall
Add basic support for KVM on ARM architecture.
Signed-off-by: Christoffer Dall
[PMM: Minor tweaks and code cleanup, switch to ONE_REG]
Signed-off-by: Peter Maydell
---
hw/arm_pic.c | 28
target-arm/Makefile.objs |1 +
target-arm/cpu.h |
Implement support for using the KVM in-kernel GIC for ARM.
Signed-off-by: Peter Maydell
---
hw/a15mpcore.c |8 ++-
hw/arm/Makefile.objs |1 +
hw/kvm/arm_gic.c | 162 ++
3 files changed, 170 insertions(+), 1 deletion(-)
create mo
Add presave/postload hooks to the ARM GIC common base class.
These will be used by the KVM in-kernel GIC subclass to sync
state between kernel and userspace when migrating.
Signed-off-by: Peter Maydell
---
hw/arm_gic_common.c | 10 ++
hw/arm_gic_internal.h |2 ++
2 files changed,
This is a v2 RFC of the QEMU patches to support KVM for
ARM on Cortex-A15 hardware. It's intended for use with
the kernel tree at
git://github.com/virtualopensystems/linux-kvm-arm.git kvm-arm-v12
(ie the 'v2' kernel patches which Christoffer sent out at
the start of the month).
Patch review appr
Enable KVM on ARM hosts, now that all the necessary components
for it exist.
Signed-off-by: Peter Maydell
---
configure |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure b/configure
index c4a7837..f87bcd6 100755
--- a/configure
+++ b/configure
@@ -3886,7 +3886,7 @@
This commit adds the ARM KVM headers. This is not to go to QEMU
upstream -- the correct path there is that the KVM code will be
committed to a mainline upstream kernel, and then upstream QEMU
can do a bulk header update from the upstream kernel, which will
allow us to drop this temporary commit.
T
Am 10.10.2012 17:22, schrieb Don Slutz:
> On 10/09/12 15:13, Don Slutz wrote:
>> On 10/09/12 12:25, Marcelo Tosatti wrote:
>>> On Mon, Sep 24, 2012 at 10:32:05AM -0400, Don Slutz wrote:
+static void x86_cpuid_set_hv_level(Object *obj, Visitor *v, void
*opaque,
+
Hi Linus,
The following changes since commit 2474542f64432398f503373f53bdf620491bcfa8:
Merge tag 'for-3.7-rc1' of git://gitorious.org/linux-pwm/linux-pwm
(2012-10-10 20:15:24 +0900)
are available in the git repository at:
git://github.com/awilliam/linux-vfio.git tags/vfio-for-v3.7-rc1
fo
On Tue, Oct 02, 2012 at 04:25:37PM +0200, Christian Borntraeger wrote:
> From: Cornelia Huck
>
> Signed-off-by: Cornelia Huck
> Signed-off-by: Martin Schwidefsky
> Signed-off-by: Christian Borntraeger
> ---
> Documentation/virtual/kvm/api.txt | 33 +
> 1 file c
On Sun, Oct 07, 2012 at 08:05:11PM +0800, Xiao Guangrong wrote:
> The only difference between FNAME(update_pte) and FNAME(pte_prefetch)
> is that the former is allowed to prefetch gfn from dirty logged slot,
> so introduce a common function to prefetch spte
>
> Signed-off-by: Xiao Guangrong
> ---
On 10.10.2012, at 18:27, Marcelo Tosatti wrote:
> On Tue, Oct 02, 2012 at 04:25:37PM +0200, Christian Borntraeger wrote:
>> From: Cornelia Huck
>>
>> Signed-off-by: Cornelia Huck
>> Signed-off-by: Martin Schwidefsky
>> Signed-off-by: Christian Borntraeger
>> ---
>> Documentation/virtual/kvm/
On Wed, 10 Oct 2012 13:27:09 -0300
Marcelo Tosatti wrote:
> > +Archtectures: s390
>
> typo.
> > +Note that the vcpu ioctl is asynchronous to vpcu execution.
>
> typo
>
>
Would you like a respin with the typos fixed?
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the b
On Wed, 10 Oct 2012 18:32:14 +0200
Alexander Graf wrote:
>
> On 10.10.2012, at 18:27, Marcelo Tosatti wrote:
>
> > On Tue, Oct 02, 2012 at 04:25:37PM +0200, Christian Borntraeger wrote:
> >> From: Cornelia Huck
> >>
> >> Signed-off-by: Cornelia Huck
> >> Signed-off-by: Martin Schwidefsky
>
Am 10.10.2012 17:07, schrieb Peter Maydell:
> Implement support for using the KVM in-kernel GIC for ARM.
>
> Signed-off-by: Peter Maydell
> ---
> hw/a15mpcore.c |8 ++-
> hw/arm/Makefile.objs |1 +
> hw/kvm/arm_gic.c | 162
> ++
Am 10.10.2012 17:07, schrieb Peter Maydell:
> Add presave/postload hooks to the ARM GIC common base class.
> These will be used by the KVM in-kernel GIC subclass to sync
> state between kernel and userspace when migrating.
>
> Signed-off-by: Peter Maydell
Reviewed-by: Andreas Färber
/-F
--
S
On 10 October 2012 18:23, Andreas Färber wrote:
> Am 10.10.2012 17:07, schrieb Peter Maydell:
>> Implement support for using the KVM in-kernel GIC for ARM.
>>
>> Signed-off-by: Peter Maydell
>> ---
>> hw/a15mpcore.c |8 ++-
>> hw/arm/Makefile.objs |1 +
>> hw/kvm/arm_gic.c | 1
On 10/10/2012 07:54 PM, Andrew Theurer wrote:
I ran 'perf sched map' on the dbench workload for medium and large VMs,
and I thought I would share some of the results. I think it helps to
visualize what's going on regarding the yielding.
These files are png bitmaps, generated from processing out
On 10/10/2012 08:29 AM, Andrew Theurer wrote:
On Wed, 2012-10-10 at 00:21 +0530, Raghavendra K T wrote:
* Avi Kivity [2012-10-04 17:00:28]:
On 10/04/2012 03:07 PM, Peter Zijlstra wrote:
On Thu, 2012-10-04 at 14:41 +0200, Avi Kivity wrote:
Again the numbers are ridiculously high for arch_lo
On 10/10/12 11:54 AM, Raghavendra K T wrote:
No, I did something like this
perf kvm --guestvmlinux ./vmlinux.guest top -g -U -d 3. Yes that is a
good idea.
(I am getting some segfaults with perf top, I think it is already fixed
but yet to see the patch that fixes)
What version of perf: perf
On Mon, Oct 08, 2012 at 06:36:11PM -0600, Shuah Khan wrote:
> Change existing kernel error message to include return value from
> iommu_attach_device() when it fails. This will help debug device
> assignment failures more effectively.
>
> Signed-off-by: Shuah Khan
> ---
> virt/kvm/iommu.c |6
On Wed, Sep 05, 2012 at 05:28:46PM +0930, Rusty Russell wrote:
> This is a generic interface to find out what you can use
> KVM_GET_ONE_REG/KVM_SET_ONE_REG on. Archs need to define
> KVM_HAVE_REG_LIST and then kvm_arch_num_regs() and
> kvm_arch_copy_reg_indices() functions.
>
> It's inspired by K
On 10/10/2012 11:33 PM, David Ahern wrote:
On 10/10/12 11:54 AM, Raghavendra K T wrote:
No, I did something like this
perf kvm --guestvmlinux ./vmlinux.guest top -g -U -d 3. Yes that is a
good idea.
(I am getting some segfaults with perf top, I think it is already fixed
but yet to see the pat
On Wed, Oct 10, 2012 at 10:51 AM, Marcelo Tosatti wrote:
> On Mon, Oct 08, 2012 at 06:36:11PM -0600, Shuah Khan wrote:
>> Change existing kernel error message to include return value from
>> iommu_attach_device() when it fails. This will help debug device
>> assignment failures more effectively.
>
On Wed, Oct 10, 2012 at 06:37:57PM +0200, Cornelia Huck wrote:
> On Wed, 10 Oct 2012 13:27:09 -0300
> Marcelo Tosatti wrote:
>
> > > +Archtectures: s390
> >
> > typo.
>
> > > +Note that the vcpu ioctl is asynchronous to vpcu execution.
> >
> > typo
> >
> >
>
> Would you like a respin with t
On Mon, Oct 08, 2012 at 06:36:11PM -0600, Shuah Khan wrote:
> Change existing kernel error message to include return value from
> iommu_attach_device() when it fails. This will help debug device
> assignment failures more effectively.
>
> Signed-off-by: Shuah Khan
Applied, thanks.
--
To unsubsc
On Mon, Oct 01, 2012 at 05:09:59AM -0400, Christoffer Dall wrote:
> The following series implements KVM support for ARM processors,
> specifically on the Cortex A-15 platform. We feel this is ready to be
> merged.
>
> Work is done in collaboration between Columbia University, Virtual Open
> Syste
On Wed, Oct 10, 2012 at 03:43:10PM -0300, Marcelo Tosatti wrote:
> On Wed, Oct 10, 2012 at 06:37:57PM +0200, Cornelia Huck wrote:
> > On Wed, 10 Oct 2012 13:27:09 -0300
> > Marcelo Tosatti wrote:
> >
> > > > +Archtectures: s390
> > >
> > > typo.
> >
> > > > +Note that the vcpu ioctl is asynchro
On Wed, 2012-10-10 at 23:13 +0530, Raghavendra K T wrote:
> On 10/10/2012 07:54 PM, Andrew Theurer wrote:
> > I ran 'perf sched map' on the dbench workload for medium and large VMs,
> > and I thought I would share some of the results. I think it helps to
> > visualize what's going on regarding the
On Wed, 2012-10-10 at 23:24 +0530, Raghavendra K T wrote:
> On 10/10/2012 08:29 AM, Andrew Theurer wrote:
> > On Wed, 2012-10-10 at 00:21 +0530, Raghavendra K T wrote:
> >> * Avi Kivity [2012-10-04 17:00:28]:
> >>
> >>> On 10/04/2012 03:07 PM, Peter Zijlstra wrote:
> On Thu, 2012-10-04 at 14:
Hey,
I just installed Windows 8 developer preview on our libvirt/kvm setup
here at work, for IE10 testing.
Host:
- Debian 6.0.4
- 2.6.32-5-amd64
- AMD FX(tm)-6100 Six-Core Processor (3.3 Ghz)
- 4 gig ram
Guest is set up to use 1 gig of ram, it's the 32-bit version installed.
Everything appe
Marcelo,
You are suggesting that I:
Kvm_host.h:
Struct kvm_x86_ops {
...
change
Int (*set_msr)(struct kvm_vcpu * vcpu, u32 mrs_index, u64 data);
to
Int (*set_msr)(struct kvm_vcpu * vcpu, u32 mrs_index, u64 data, bool
from_guest);
...
};
and so on down the line to set_msr_common
Rusty Russell writes:
> Jamie Lokier writes:
>
>> Rusty Russell wrote:
>>> I don't think it'll be that bad; reset clears the device to unknown,
>>> bar0 moves it from unknown->legacy mode, bar1/2/3 changes it from
>>> unknown->modern mode, and anything else is bad (I prefer being strict so
>>> we
"Michael S. Tsirkin" writes:
> On Wed, Oct 10, 2012 at 08:36:12AM -0500, Anthony Liguori wrote:
>> Rusty Russell writes:
>>
>> > Gerd Hoffmann writes:
>> >> So how about this:
>> >>
>> >> (1) Add a vendor specific pci capability for new-style virtio.
>> >> Specifies the pci bar used for new
Paolo Bonzini writes:
> Il 09/10/2012 06:59, Rusty Russell ha scritto:
>> Paolo Bonzini writes:
>>> Il 05/10/2012 07:43, Rusty Russell ha scritto:
That's good. But virtio_blk's scsi command is insoluble AFAICT. As I
said to Anthony, the best rules are "always" and "never", so I'd real
Needed for changing the hypercall handlers' argument type to PowerPCCPU.
Signed-off-by: Andreas Färber
---
hw/spapr.c |2 +-
hw/spapr.h |2 +-
hw/spapr_hcall.c |4 +++-
target-ppc/kvm.c |3 ++-
4 Dateien geändert, 7 Zeilen hinzugefügt(+), 4 Zeilen entfernt(-)
diff --
Needed for changing qemu_cpu_kick() argument type to CPUState.
Signed-off-by: Andreas Färber
---
target-ppc/kvm.c |8 ++--
1 Datei geändert, 6 Zeilen hinzugefügt(+), 2 Zeilen entfernt(-)
diff --git a/target-ppc/kvm.c b/target-ppc/kvm.c
index 5cbe98a..d7d8e8f 100644
--- a/target-ppc/kvm.
"Michael S. Tsirkin" writes:
> On Mon, Oct 08, 2012 at 12:51:25PM +1030, Rusty Russell wrote:
>> Note before anyone gets confused; we were talking about using the PCI
>> config space to indicate what BAR(s) the virtio stuff is in. An
>> alternative would be to simply specify a new layout format i
Christoffer Dall writes:
> On Tue, Oct 9, 2012 at 2:11 PM, Peter Maydell
> wrote:
>> On 1 October 2012 10:11, Christoffer Dall
>> wrote:
>>> From: Rusty Russell
...
>>> +ARM 32-bit VFP control registers have the following id bit patterns:
>>> + 0x4002 0012 1
>>> +
>>> +ARM 64-bit FP re
This is analogous to commit a1b383870a made by Rusty Russell to all
the VirtIO headers at the time. This eases the use of the header as
is by other OSes.
Signed-off-by: Bryan Venteicher
---
include/linux/virtio_scsi.h | 28 ++--
1 file changed, 26 insertions(+), 2 deletio
61 matches
Mail list logo
