On Mon, 01 Dec 2014 18:38:34 +0100
Paolo Bonzini wrote:
>
>
> On 01/12/2014 18:29, Igor Mammedov wrote:
> > Series speed-ups GFN to memslot lookup time by:
> > * introducing LRU cache, which improves looukup time for
> >same slot workload (typically boot time of Windows and Linux guest)
>
On 02/12/2014 07:14, Wanpeng Li wrote:
> Expose intel xsaves feature to guest.
>
> Signed-off-by: Wanpeng Li
> ---
> v1 -> v2:
> *auto switch msr ia32_xss if this msr is present
>
> arch/x86/include/asm/kvm_host.h | 1 +
> arch/x86/include/asm/vmx.h | 3 +++
> arch/x86/include/uapi/as
On 02/12/2014 07:14, Wanpeng Li wrote:
> xsaves will be exported to guest in the next patch, so revert the
> mask out xsaves patch.
>
> Signed-off-by: Wanpeng Li
> ---
> arch/x86/kvm/cpuid.c | 10 +-
> 1 file changed, 1 insertion(+), 9 deletions(-)
>
> diff --git a/arch/x86/kvm/cpuid.
Hi Paolo,
On Tue, Dec 02, 2014 at 09:51:22AM +0100, Paolo Bonzini wrote:
>
>
>On 02/12/2014 07:14, Wanpeng Li wrote:
>> Expose intel xsaves feature to guest.
>>
>> Signed-off-by: Wanpeng Li
>> ---
>> v1 -> v2:
>> *auto switch msr ia32_xss if this msr is present
>>
>> arch/x86/include/asm/kvm_h
On Tue, Dec 02, 2014 at 09:55:09AM +0100, Paolo Bonzini wrote:
>
>
>On 02/12/2014 07:14, Wanpeng Li wrote:
>> xsaves will be exported to guest in the next patch, so revert the
>> mask out xsaves patch.
>>
>> Signed-off-by: Wanpeng Li
>> ---
>> arch/x86/kvm/cpuid.c | 10 +-
>> 1 file chan
Hi,
I'm trying to develop a custom virtio driver for myself.
Can I ask here virtio driver implementation details?
I've tried the kvm and qemu irc channel, but had no luck.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.kernel.org
More
On 01/12/14 22:36, Luis R. Rodriguez wrote:
>
> Then I do agree its a fair analogy (and find this obviously odd that how
> widespread cond_resched() is), we just don't have an equivalent for IRQ
> context, why not avoid the special check then and use this all the time in the
> middle of a hypercal
This patchset is to enable xsaves for kvm part, the patch for
qemu part will be sent out later.
The patchset is tested on skylake-client.
v2 -> v3:
* add kvm_get/set for ia32_xss
* fix the type XSS_EXIT_BITMAP
* load host_xss just once in setup_vmcs_config
* add/clear atuo switch ia32_xss m
Add kvm_get/set logic to xsaves.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 20
1 file changed, 20 insertions(+)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 12915f1..66d1e3d 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -99,6 +99,8 @@ m
Intel xsaves vmx and msr handle.
Signed-off-by: Wanpeng Li
---
arch/x86/include/asm/kvm_host.h | 2 ++
arch/x86/include/asm/vmx.h | 3 +++
arch/x86/include/uapi/asm/vmx.h | 6 +-
arch/x86/kvm/vmx.c | 31 ++-
4 files changed, 40 insertions(+),
Exporse intel xsaves feature to guest.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/cpuid.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index a4f5ac4..0d919bc 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -267,
This is the link to the presentation:
http://www.slideshare.net/zenixls2/052-virtio-introduction-17191942
--
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
I'm trying to make a custom virtio driver that mostly interacts with memory.
I've came upon a presentation which shows which files need to be edited
on qemu
and kvm in order to add a new virtio driver.
One of the files in qemu is include/hw/pci/pci.h,
but that was an example specific to virtio p
I got win7 installed, virtio devices working and took forever to trickle
in updates because of a w7 bug update manager bug that take up all cpu
resources. now I got DNS 13 installed but I'm getting no audio.
I pass throught the usb audio device (logitech h800 USB 046d:0a29) and
it is seen as
Add xsaves related definition, it also add corresponding part to
kvm_get/put, and vmstate.
Signed-off-by: Wanpeng Li
---
target-i386/cpu.h | 2 ++
target-i386/kvm.c | 15 +++
target-i386/machine.c | 3 ++-
3 files changed, 19 insertions(+), 1 deletion(-)
diff --git a/targ
On 02/12/2014 13:19, Wanpeng Li wrote:
> Add xsaves related definition, it also add corresponding part to
> kvm_get/put, and vmstate.
>
> Signed-off-by: Wanpeng Li
> ---
> target-i386/cpu.h | 2 ++
> target-i386/kvm.c | 15 +++
> target-i386/machine.c | 3 ++-
> 3 files
On 02/12/2014 13:16, Eric S. Johansson wrote:
> I got win7 installed, virtio devices working and took forever to trickle
> in updates because of a w7 bug update manager bug that take up all cpu
> resources. now I got DNS 13 installed but I'm getting no audio.
>
> I pass throught the usb audio d
Another iteration of virtio-1 patches for qemu, as always available on
git://github.com/cohuck/qemu virtio-1
This one seems to work together with the current vhost-next patches
(well, I can ping :)
Changes from v4:
- add helpers for feature bit manipulation and checking
- use 64 bit feature bits
virtio-1 devices always use num_buffers in the header, even if
mergeable rx buffers have not been negotiated.
Signed-off-by: Cornelia Huck
---
hw/net/virtio-net.c | 21 +++--
1 file changed, 15 insertions(+), 6 deletions(-)
diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.
virtio-net (non-vhost) now should have everything in place to support
virtio 1.0: let's enable the feature bit for it.
Note that VIRTIO_F_VERSION_1 is technically a transport feature; once
every device is ready for virtio 1.0, we can move setting this
feature bit out of the individual devices.
Si
Wire up virtio-blk to provide different feature bit sets depending
on whether legacy or v1.0 has been requested.
Note that VERSION_1 is still disabled due to missing ANY_LAYOUT support.
Signed-off-by: Cornelia Huck
---
hw/block/virtio-blk.c | 19 +++
1 file changed, 19 inserti
Devices may support different sets of feature bits depending on which
revision they're operating at. Let's give the transport a way to
re-query the device about its features when the revision has been
changed.
Signed-off-by: Cornelia Huck
---
hw/s390x/virtio-ccw.c | 12 ++--
h
Add code that checks for the VERSION_1 feature bit in order to make
decisions about the device's endianness. This allows us to support
transitional devices.
Signed-off-by: Cornelia Huck
---
hw/virtio/virtio.c|6 +-
include/hw/virtio/virtio-access.h |4
include/hw
For virtio-1 devices, we allow a more complex queue layout that doesn't
require descriptor table and rings on a physically-contigous memory area:
add virtio_queue_set_rings() to allow transports to set this up.
Signed-off-by: Cornelia Huck
---
hw/virtio/virtio.c | 16
For virtio-1 devices, the driver must not attempt to set feature bits
after it set FEATURES_OK in the device status. Simply reject it in
that case.
Signed-off-by: Cornelia Huck
---
hw/virtio/virtio.c | 16 ++--
include/hw/virtio/virtio.h |2 ++
2 files changed, 16 inser
From: Thomas Huth
We need a possibility to run code when a subchannel gets disabled.
This patch adds the necessary infrastructure.
Signed-off-by: Thomas Huth
Signed-off-by: Cornelia Huck
---
hw/s390x/css.c | 12
hw/s390x/css.h |1 +
2 files changed, 13 insertions(+)
diff -
virtio-1 allow setting of the FEATURES_OK status bit to fail if
the negotiated feature bits are inconsistent: let's fail
virtio_set_status() in that case and update virtio-ccw to post an
error to the guest.
Signed-off-by: Cornelia Huck
---
hw/s390x/virtio-ccw.c | 20
From: Thomas Huth
Add the new VIRTIO_F_VERSION_1 definition to the virtio_config.h
linux header.
Signed-off-by: Thomas Huth
Signed-off-by: Cornelia Huck
---
linux-headers/linux/virtio_config.h |3 +++
1 file changed, 3 insertions(+)
diff --git a/linux-headers/linux/virtio_config.h
b/lin
Devices operating as virtio 1.0 may not allow writes to the mac
address in config space.
Signed-off-by: Cornelia Huck
---
hw/net/virtio-net.c |1 +
1 file changed, 1 insertion(+)
diff --git a/hw/net/virtio-net.c b/hw/net/virtio-net.c
index d6d1b98..ebbea60 100644
--- a/hw/net/virtio-net.c
+
From: Thomas Huth
Handle the virtio-ccw revision according to what the guest sets.
When revision 1 is selected, we have a virtio-1 standard device
with byteswapping for the virtio rings.
When a channel gets disabled, we have to revert to the legacy behavior
in case the next user of the device do
Support the new CCW_CMD_SET_VQ format for virtio-1 devices.
While we're at it, refactor the code a bit and enforce big endian
fields (which had always been required, even for legacy).
Reviewed-by: Thomas Huth
Signed-off-by: Cornelia Huck
---
hw/s390x/virtio-ccw.c | 114 +++
Add virtio_{add,clear}_feature helper functions for manipulating a
feature bits variable. This has some benefits over open coding:
- add check that the bit is in a sane range
- make it obvious at a glance what is going on
- have a central point to change when we want to extend feature bits
Convert
Add a helper function for checking whether a bit is set in the guest
features for a vdev as well as one that works on a feature bit set.
Convert code that open-coded this: It cleans up the code and makes it
easier to extend the guest feature bits.
Signed-off-by: Cornelia Huck
---
hw/block/virti
virtio-ccw should now have everything in place to operate virtio 1.0
devices, so let's enable revision 1.
Signed-off-by: Cornelia Huck
---
hw/s390x/virtio-ccw.h |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/s390x/virtio-ccw.h b/hw/s390x/virtio-ccw.h
index fe5c782..d40
The only user of this function was virtio-ccw, and it should use
virtio_set_features() like everybody else: We need to make sure
that bad features are masked out properly, which this function did
not do.
Reviewed-by: Thomas Huth
Signed-off-by: Cornelia Huck
---
hw/s390x/virtio-ccw.c |
With virtio-1, we support more than 32 feature bits. Let's extend both
host and guest features to 64, which should suffice for a while.
vhost and migration have been ignored for now.
Signed-off-by: Cornelia Huck
---
hw/9pfs/virtio-9p-device.c |2 +-
hw/block/virtio-blk.c |
Handle endianness conversion for virtio-1 virtqueues correctly.
Note that dataplane now needs to be built per-target.
Signed-off-by: Cornelia Huck
---
hw/block/dataplane/virtio-blk.c |4 +-
hw/scsi/virtio-scsi-dataplane.c |2 +-
hw/virtio/Makefile.objs
On 02/12/2014 12:14, Wanpeng Li wrote:
> This patchset is to enable xsaves for kvm part, the patch for
> qemu part will be sent out later.
>
> The patchset is tested on skylake-client.
>
> v2 -> v3:
> * add kvm_get/set for ia32_xss
> * fix the type XSS_EXIT_BITMAP
> * load host_xss just on
- EAX=0Dh, ECX=1: output registers EBX/ECX/EDX are reserved.
- EAX=0Dh, ECX>1: output register ECX is zero for all the CPUID leaves
we support, because variable "supported" comes from XCR0 and not XSS.
However, only bits above 0 are reserved. Output register EDX is reserved.
Source: Intel Archit
On Mon, Dec 01, 2014 at 11:50:14AM +, Alex Bennée wrote:
>
> Christoffer Dall writes:
>
> > On Tue, Nov 25, 2014 at 04:10:03PM +, Alex Bennée wrote:
> >> This adds support for single-stepping the guest. As userspace can and
> >> will manipulate guest registers before restarting any tweak
On Mon, Dec 01, 2014 at 11:52:44AM +, Alex Bennée wrote:
>
> Christoffer Dall writes:
>
> > On Tue, Nov 25, 2014 at 04:10:04PM +, Alex Bennée wrote:
> >> This is a pre-cursor to sharing the code with the guest debug support.
> >> This replaces the big macro that fishes data out of a fixe
On 02/12/2014 08:57, Igor Mammedov wrote:
>> On 01/12/2014 18:29, Igor Mammedov wrote:
>>> Series speed-ups GFN to memslot lookup time by:
>>> * introducing LRU cache, which improves looukup time for
>>>same slot workload (typically boot time of Windows and Linux guest)
>>> * switching to b
On Mon, Dec 01, 2014 at 05:57:53PM +, Peter Maydell wrote:
> On 27 November 2014 at 23:10, Peter Maydell wrote:
> > It seems odd to have this unmap happen on attempted system reset/powerdown,
> > not on cpu init/start.
>
> Here's a concrete case that I think requires the unmap to be
> done on
On 27/11/2014 13:16, mad Engineer wrote:
> a random thought can we set qemu user/group rss to a particular hard
> limit in limits.conf
Those apply per process, not per user/group.
Paolo
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majord...@vger.
On Thu, Nov 27, 2014 at 10:44:29PM +, Peter Maydell wrote:
> On 27 November 2014 at 18:40, Christoffer Dall
> wrote:
> > The implementation of KVM_ARM_VCPU_INIT is currently not doing what
> > userspace expects, namely making sure that a vcpu which may have been
> > turned off using PSCI is re
On 12/01/2014 05:37 PM, Borislav Petkov wrote:
On Mon, Dec 01, 2014 at 05:31:56PM -0500, Boris Ostrovsky wrote:
I think so. The problem we have now is __pa() macro that we only use
on 32-bit. I'll queue this for overnight tests to make sure and if it
indeed works then 3.19 should be fine.
Cool,
On Tue, Dec 02, 2014 at 02:00:15PM +0100, Cornelia Huck wrote:
> For virtio-1 devices, we allow a more complex queue layout that doesn't
> require descriptor table and rings on a physically-contigous memory area:
> add virtio_queue_set_rings() to allow transports to set this up.
>
> Signed-off-by:
On Thu, Nov 27, 2014 at 10:53:50PM +, Peter Maydell wrote:
> On 27 November 2014 at 18:40, Christoffer Dall
> wrote:
> > It is not clear that this ioctl can be called multiple times for a given
> > vcpu. Userspace already does this, so clarify the ABI.
> >
> > Signed-off-by: Christoffer Dall
On Mon, Dec 01, 2014 at 02:34:12PM +0100, Andrew Jones wrote:
> On Thu, Nov 27, 2014 at 07:40:55PM +0100, Christoffer Dall wrote:
> > Several people have reported problems with rebooting ARM VMs, especially
> > on 32-bit ARM. This is mainly due to the same reason we were seeing
> > boot errors in
On Tue, 2 Dec 2014 16:46:28 +0200
"Michael S. Tsirkin" wrote:
> On Tue, Dec 02, 2014 at 02:00:15PM +0100, Cornelia Huck wrote:
> > For virtio-1 devices, we allow a more complex queue layout that doesn't
> > require descriptor table and rings on a physically-contigous memory area:
> > add virtio_q
On Tue, Dec 02, 2014 at 09:36:40AM -0500, Boris Ostrovsky wrote:
> All tests passed.
Thanks!
> I wonder whether we should prevent all guests (not just paravirt) from
> loading microcode driver (and from doing early microcode loading).
I don't think the unmodified ones need to. At least I haven't
On Thu, Nov 27, 2014 at 11:10:14PM +, Peter Maydell wrote:
> On 27 November 2014 at 18:41, Christoffer Dall
> wrote:
> > When a vcpu calls SYSTEM_OFF or SYSTEM_RESET with PSCI v0.2, the vcpus
> > should really be turned off for the VM adhering to the suggestions in
> > the PSCI spec, and it's
On 2 December 2014 at 14:47, Christoffer Dall
wrote:
> On Thu, Nov 27, 2014 at 10:53:50PM +, Peter Maydell wrote:
>> On 27 November 2014 at 18:40, Christoffer Dall
>> wrote:
>> > Possible features:
>> > - KVM_ARM_VCPU_POWER_OFF: Starts the CPU in a power-off state.
>> > Dep
On Tue, 2 Dec 2014 15:54:44 +0100
Cornelia Huck wrote:
> On Tue, 2 Dec 2014 16:46:28 +0200
> "Michael S. Tsirkin" wrote:
> > pa == -1ULL tricks look quite ugly.
> > Can't we set desc/avail/used unconditionally, and drop
> > the pa value?
>
> And have virtio_queue_get_addr() return desc? Let me
On 2 December 2014 at 15:01, Christoffer Dall
wrote:
> On Thu, Nov 27, 2014 at 11:10:14PM +, Peter Maydell wrote:
>> Also, the clarification we want here should not I think be architecture
>> specific -- the handling of the exit system event in QEMU is in common
>> code. What you want to say i
On Tue, 2014-12-02 at 08:52 +0100, Eric Auger wrote:
> On 12/02/2014 05:48 AM, Alex Williamson wrote:
> > On Tue, 2014-12-02 at 02:08 +, Wu, Feng wrote:
> >>
> >>> -Original Message-
> >>> From: Eric Auger [mailto:eric.au...@linaro.org]
> >>> Sent: Monday, December 01, 2014 6:10 PM
> >>
(tl;dr version at the bottom)
2014-12-01 15:43-0800, Phil White:
> On Mon, Dec 1, 2014 at 5:47 AM, Radim Krčmář wrote:
> > 2014-11-28 17:29-0800, Phil White:
> >> Good questions.
> >>
> >> One thing that prompted this code is the presence and proliferation of
> >> architecture specific hypercalls
Since the advent of dynamic initialization of VGIC, this latter is
initialized very late, on the first vcpu run. This initialization
could be initiated much earlier by the user, as soon as it has
provided the requested dimensioning parameters:
- number of IRQs and number of vCPUs,
- DIST and CPU in
2014-12-01 17:29+, Igor Mammedov:
> Current linear search doesn't scale well when
> large amount of memslots is used and looked up slot
> is not in the beginning memslots array.
> Taking in account that memslots don't overlap, it's
> possible to switch sorting order of memslots array from
> 'np
On 11/11/2014 12:12 PM, Ben Hutchings wrote:
> This reverts commit 88e0e0e5aa722b193c8758c8b45d041de5316924 for
> the tap drivers, but leaves UFO disabled in virtio_net.
>
> libvirt at least assumes that tap features will never be dropped
> in new kernel versions, and doing so prevents migration o
On 2 December 2014 at 17:27, Eric Auger wrote:
> Since the advent of dynamic initialization of VGIC, this latter is
> initialized very late, on the first vcpu run. This initialization
> could be initiated much earlier by the user, as soon as it has
> provided the requested dimensioning parameters:
On 12/02/2014 06:50 PM, Peter Maydell wrote:
> On 2 December 2014 at 17:27, Eric Auger wrote:
>> Since the advent of dynamic initialization of VGIC, this latter is
>> initialized very late, on the first vcpu run. This initialization
>> could be initiated much earlier by the user, as soon as it has
On 12/02/2014 05:02 PM, Alex Williamson wrote:
> On Tue, 2014-12-02 at 08:52 +0100, Eric Auger wrote:
>> On 12/02/2014 05:48 AM, Alex Williamson wrote:
>>> On Tue, 2014-12-02 at 02:08 +, Wu, Feng wrote:
> -Original Message-
> From: Eric Auger [mailto:eric.au...@linaro.org]
On 02/12/2014 18:33, Radim Krčmář wrote:
>> > + while (start < end) {
>> > + slot = start + (end - start) / 2;
>> > +
>> > + if (gfn >= memslots[slot].base_gfn)
> (Even thought division is costly, I think that checking here if 'slot'
> is the one we want wouldn't help very muc
On Tue, Dec 02, 2014 at 03:39:05PM +, Peter Maydell wrote:
> On 2 December 2014 at 14:47, Christoffer Dall
> wrote:
> > On Thu, Nov 27, 2014 at 10:53:50PM +, Peter Maydell wrote:
> >> On 27 November 2014 at 18:40, Christoffer Dall
> >> wrote:
> >> > Possible features:
> >> > - KV
On Tue, Dec 02, 2014 at 04:41:36PM +0100, Cornelia Huck wrote:
> void virtio_queue_set_num(VirtIODevice *vdev, int n, int num)
> {
> +/*
> + * For virtio-1 devices, the number of buffers may only be
> + * updated if the ring addresses have not yet been set up.
Where does it say that?
On Wed, Oct 29, 2014 at 04:19:10PM -0400, Waiman Long wrote:
> This patch adds the necessary KVM specific code to allow KVM to
> support the CPU halting and kicking operations needed by the queue
> spinlock PV code.
>
> Two KVM guests of 20 CPU cores (2 nodes) were created for performance
> testin
Hi
Please, send any topic that you are interested in covering.
hanks, Juan.
By popular demand, a google calendar public entry with it
https://www.google.com/calendar/embed?src=dG9iMXRqcXAzN3Y4ZXZwNzRoMHE4a3BqcXNAZ3JvdXAuY2FsZW5kYXIuZ29vZ2xlLmNvbQ
(Let me know if you have any problems with
2014-12-02 19:45+0100, Paolo Bonzini:
> On 02/12/2014 18:33, Radim Krčmář wrote:
> >> > +while (start < end) {
> >> > +slot = start + (end - start) / 2;
> >> > +
> >> > +if (gfn >= memslots[slot].base_gfn)
> > (Even thought division is costly, I think that ch
In emulator.c/tests_smsw, smsw (3) fails because h_mem isn't being set correctly
before smsw is called. By declaring the h_mem function parameter as volatile,
the compiler no longer optimizes out the assignment before smsw.
Signed-off-by: Chris J Arges
---
x86/emulator.c | 2 +-
1 file changed,
2014-12-02 14:09+0100, Paolo Bonzini:
> - EAX=0Dh, ECX=1: output registers EBX/ECX/EDX are reserved.
(As good as reserved without XSAVES/IA32_XSS.)
> - EAX=0Dh, ECX>1: output register ECX is zero for all the CPUID leaves
> we support, because variable "supported" comes from XCR0 and not XSS.
> Ho
On Tue, Dec 02, 2014 at 02:00:24PM +0100, Paolo Bonzini wrote:
>
>
>On 02/12/2014 12:14, Wanpeng Li wrote:
>> This patchset is to enable xsaves for kvm part, the patch for
>> qemu part will be sent out later.
>>
>> The patchset is tested on skylake-client.
>>
>> v2 -> v3:
>> * add kvm_get/set
On Wed, 29 Oct 2014, Waiman Long wrote:
> AIM7 XFS Disk Test (no overcommit)
> kernel JPMReal Time Sys TimeUsr Time
> - ----
> PV ticketlock 25423737.08 98.95 5.44
> PV
Expose intel xsaves feature to guest.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/cpuid.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c
index a4f5ac4..0d919bc 100644
--- a/arch/x86/kvm/cpuid.c
+++ b/arch/x86/kvm/cpuid.c
@@ -267,6
Add kvm_get/set logic to xsaves.
Signed-off-by: Wanpeng Li
---
arch/x86/kvm/vmx.c | 26 ++
1 file changed, 26 insertions(+)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 12915f1..ad1153a 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -99,6 +99,8
This patchset is to enable xsaves for kvm part and tested on skylake-client.
v3 -> v4:
* return false from an svm_xsaves_supported function
* prevent setting MSR_IA32_XSS to any non-zero value
v2 -> v3:
* add kvm_get/set for ia32_xss
* fix the type XSS_EXIT_BITMAP
* load host_xss just once i
Intel xsaves vmx and msr handle.
Signed-off-by: Wanpeng Li
---
arch/x86/include/asm/kvm_host.h | 2 ++
arch/x86/include/asm/vmx.h | 3 +++
arch/x86/include/uapi/asm/vmx.h | 6 +-
arch/x86/kvm/svm.c | 6 ++
arch/x86/kvm/vmx.c | 31 +++
On Tue, Dec 02, 2014 at 09:55:09AM +0100, Paolo Bonzini wrote:
>
>
>On 02/12/2014 07:14, Wanpeng Li wrote:
>> xsaves will be exported to guest in the next patch, so revert the
>> mask out xsaves patch.
>>
>> Signed-off-by: Wanpeng Li
>> ---
>> arch/x86/kvm/cpuid.c | 10 +-
>> 1 file chan
Add xsaves related definition, it also adds corresponding part
to kvm_get/put, and vmstate.
Signed-off-by: Wanpeng Li
---
v1 -> v2:
* use a subsection instead of bumping the version number.
target-i386/cpu.h | 2 ++
target-i386/kvm.c | 15 +++
target-i386/machine.c | 21 +
On Tue, Dec 02, 2014 at 11:11:18AM +, David Vrabel wrote:
> On 01/12/14 22:36, Luis R. Rodriguez wrote:
> >
> > Then I do agree its a fair analogy (and find this obviously odd that how
> > widespread cond_resched() is), we just don't have an equivalent for IRQ
> > context, why not avoid the sp
This series of patches is based on Alex Graf's kvm-ppc-queue branch
and is intended for the 3.19 merge window. It starts by removing the
code to support HV KVM on PPC970 processors. This code is hardly used
now since there are not many HV-capable PPC970 machines (Apple G5
machines are not HV-capa
From: Sam Bobroff
Currently the H_CONFER hcall is implemented in kernel virtual mode,
meaning that whenever a guest thread does an H_CONFER, all the threads
in that virtual core have to exit the guest. This is bad for
performance because it interrupts the other threads even if they
are doing use
This removes the code that was added to enable HV KVM to work
on PPC970 processors. The PPC970 is an old CPU that doesn't
support virtualizing guest memory. Removing PPC970 support also
lets us remove the code for allocating and managing contiguous
real-mode areas, the code for the !kvm->arch.usi
From: Anton Blanchard
The in-kernel XICS emulation is faster than doing it all in QEMU
and it has got a lot of testing, so enable it by default.
Signed-off-by: Anton Blanchard
Signed-off-by: Paul Mackerras
---
arch/powerpc/kvm/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/p
From: Michael Ellerman
Some PowerNV systems include a hardware random-number generator.
This HWRNG is present on POWER7+ and POWER8 chips and is capable of
generating one 64-bit random number every microsecond. The random
numbers are produced by sampling a set of 64 unstable high-frequency
oscil
There are two ways in which a guest instruction can be obtained from
the guest in the guest exit code in book3s_hv_rmhandlers.S. If the
exit was caused by a Hypervisor Emulation interrupt (i.e. an illegal
instruction), the offending instruction is in the HEIR register
(Hypervisor Emulation Instruc
Add xsaves related definition, it also adds corresponding part
to kvm_get/put, and vmstate.
Signed-off-by: Wanpeng Li
---
v1 -> v2:
* use a subsection instead of bumping the version number.
target-i386/cpu.h | 2 ++
target-i386/kvm.c | 15 +++
target-i386/machine.c | 21 +
When a secondary hardware thread has finished running a KVM guest, we
currently put that thread into nap mode using a nap instruction in
the KVM code. This changes the code so that instead of doing a nap
instruction directly, we instead cause the call to power7_nap() that
put the thread into nap m
On 12/03/2014 03:28 AM, Luis R. Rodriguez wrote:
On Tue, Dec 02, 2014 at 11:11:18AM +, David Vrabel wrote:
On 01/12/14 22:36, Luis R. Rodriguez wrote:
Then I do agree its a fair analogy (and find this obviously odd that how
widespread cond_resched() is), we just don't have an equivalent fo
Make kvm_set_msi_irq() public, we can use this function outside.
Signed-off-by: Feng Wu
---
include/linux/kvm_host.h |2 ++
virt/kvm/irq_comm.c |2 +-
2 files changed, 3 insertions(+), 1 deletions(-)
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index cfa85ac..5c
Enable VT-d Posted-Interrtups and add a command line
parameter for it.
Signed-off-by: Feng Wu
---
Documentation/kernel-parameters.txt |1 +
drivers/iommu/irq_remapping.c | 12
2 files changed, 9 insertions(+), 4 deletions(-)
diff --git a/Documentation/kernel-parameters.
When vCPU is blocked and an external interrupts from assigned
devices is delivered to it, VT-d Posted-Interrupts mechanism
will deliver an interrupt to the associated physical CPU with
Wake-up Vector. In its handler, we find the destination vCPU
and wake up it.
Signed-off-by: Feng Wu
---
arch/x8
Currently, we don't support urgent interrupt, all interrupts
are recognized as non-urgent interrupt, so we cannot send
posted-interrupt when 'SN' is set.
Signed-off-by: Feng Wu
---
arch/x86/kvm/vmx.c | 11 +--
1 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kvm/
Update Posted-Interrupts descriptor according to the
following rules:
- Before vCPU block, set 'NV' to POSTED_INTR_WAKEUP_VECTOR
- After vCPU block, set 'NV' back to POSTED_INTR_VECTOR
Signed-off-by: Feng Wu
---
arch/x86/include/asm/kvm_host.h |5 ++
arch/x86/kvm/vmx.c | 83 ++
This patch changes the NDST filed of Posted-Interrupts
Descriptor after vCPU is scheduled to another physical
CPU.
Signed-off-by: Feng Wu
---
arch/x86/kvm/vmx.c | 25 +
1 files changed, 25 insertions(+), 0 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx
This patch defines macro __KVM_HAVE_ARCH_KVM_VFIO_POSTING and
implement kvm_arch_vfio_update_pi_irte for x86 architecture.
Signed-off-by: Feng Wu
---
arch/x86/include/asm/kvm_host.h |2 +
arch/x86/kvm/Makefile |2 +-
arch/x86/kvm/kvm_vfio_x86.c | 68 ++
This patch adds and documents a new attribute
KVM_DEV_VFIO_DEVICE_POSTING_IRQ in KVM_DEV_VFIO_DEVICE group.
This new attribute is used for VT-d Posted-Interrupts.
When guest OS changes the interrupt configuration for an
assigned device, such as, MSI/MSIx data/address fields,
QEMU will use this IRQ
Currently, we use a global vector as the Posted-Interrupts
Notification Event for all the vCPUs in the system. We need
to introduce another global vector for VT-d Posted-Interrtups,
which will be used to wakeup the sleep vCPU when an external
interrupt from a direct-assigned device happens for that
This patch adds the kvm-vfio interface for VT-d Posted-Interrrupts.
When guests updates MSI/MSI-x information for an assigned-device,
QEMU will use KVM_DEV_VFIO_DEVICE_POSTING_IRQ attribute to setup
IRTE for VT-d PI. This patch implement this IRQ attribute.
Signed-off-by: Feng Wu
---
include/lin
This patch adds some helper functions to manipulate the
Posted-Interrupts Descriptor.
Signed-off-by: Feng Wu
---
arch/x86/kvm/vmx.c | 26 ++
1 files changed, 26 insertions(+), 0 deletions(-)
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index abdb84f..0b1383e 10
1 - 100 of 114 matches
Mail list logo
