From: Gleb Natapov g...@redhat.com
Use groups mechanism to decode 0F BA instructions.
Signed-off-by: Gleb Natapov g...@redhat.com
Cc: sta...@kernel.org
Signed-off-by: Avi Kivity a...@redhat.com
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 645b245..435b1e4 100644
---
From: Gleb Natapov g...@redhat.com
Use groups mechanism to decode 0F C7 instructions.
Signed-off-by: Gleb Natapov g...@redhat.com
Cc: sta...@kernel.org
Signed-off-by: Avi Kivity a...@redhat.com
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 435b1e4..45a4f7c 100644
---
From: Gleb Natapov g...@redhat.com
Make emulator check that vcpu is allowed to execute IN, INS, OUT,
OUTS, CLI, STI.
Signed-off-by: Gleb Natapov g...@redhat.com
Cc: sta...@kernel.org
Signed-off-by: Avi Kivity a...@redhat.com
diff --git a/arch/x86/include/asm/kvm_host.h
From: Gleb Natapov g...@redhat.com
Currently when x86 emulator needs to access memory, page walk is done with
broadest permission possible, so if emulated instruction was executed
by userspace process it can still access kernel memory. Fix that by
providing correct memory access to page walker
From: Gleb Natapov g...@redhat.com
Add CPL checking in case emulator is tricked into emulating
privilege instruction from userspace.
Signed-off-by: Gleb Natapov g...@redhat.com
Cc: sta...@kernel.org
Signed-off-by: Avi Kivity a...@redhat.com
diff --git a/arch/x86/kvm/emulate.c
From: Gleb Natapov g...@redhat.com
Instructions which are not allowed to have LOCK prefix should
generate #UD if one is used.
Signed-off-by: Gleb Natapov g...@redhat.com
Signed-off-by: Avi Kivity a...@redhat.com
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index d632111..31f4526
From: Gleb Natapov g...@redhat.com
POPF behaves differently depending on current CPU mode. Emulate correct
logic to prevent guest from changing flags that it can't change otherwise.
Signed-off-by: Gleb Natapov g...@redhat.com
Cc: sta...@kernel.org
Signed-off-by: Avi Kivity a...@redhat.com
diff
repository: C:/dev/kvm-guest-drivers-windows
branch: master
commit 9a903b9ef2e73c3da6f700aabb86ea38b472a9e5
Author: Yan Vugenfirer yvuge...@redhat.com
Date: Wed Feb 10 17:26:50 2010 +0200
[WIN-GUEST-DRIVERS] VirtIO lib: Adding compilation to Win 7 platform.
Also allowing to specify
hi,
I am hitting this bug as well, using qemu-kvm-0.12.2 with the following command
/usr/bin/kvm -name pfsense,process=pfsense -m 256 -vnc
192.168.0.254:13 -smp 1 -hda /dev/volume01/G-pfsense -vga std
Option 'ipv4': Use 'on' or 'off'
Failed to parse yes for dummy.ipv4
inet_listen_opts:
Linus, please pull from
git://git.kernel.org/pub/scm/virt/kvm/kvm.git kvm-updates/2.6.33
To receive a suspend/resume kvmclock fix, and a memory corruption fix
involving the emulated PIT.
Jason Wang (2):
Export the symbol of getboottime and mmonotonic_to_bootbased
kvmclock:
On Wednesday 10 February 2010 09:30:57 xming wrote:
hi,
Hi,
I am hitting this bug as well, using qemu-kvm-0.12.2 with the following
command
/usr/bin/kvm -name pfsense,process=pfsense -m 256 -vnc
192.168.0.254:13 -smp 1 -hda /dev/volume01/G-pfsense -vga std
Option 'ipv4': Use 'on' or
Hello friends,
I have downloaded package qemu-kvm_0.11.0-0ubuntu6_i386.deb
and when i double click it,it tells 'Error: Dependency is not
satisfiable: bridge-utils'.i tried by downloading the package again
but still problem exists. Any help would be of great use.Thanks .
--
To
Please reuse the changelog when reposing a patch, this makes it easier
for me to apply it.
Thanks. Will follow it from next time.
Should be a host long size, not guest. This will fail when running a
32-bit qemu-system-x86_64 binary.
Sorry. That was our mistake.
Instead of using a
On 02/10/2010 11:55 AM, OHMURA Kei wrote:
Instead of using a nested loop if bitmap_ul[i] != 0, it is possible to
use just a single loop (while (c 0)), and process a long's worth of data.
The only trickery is with big endian hosts, where the conversion from
bit number to page number is a bit
dirty-bitmap-traveling is carried out by byte size in qemu-kvm.c.
But We think that dirty-bitmap-traveling by long size is faster than by byte
size especially when most of memory is not dirty.
Signed-off-by: OHMURA Kei ohmura@lab.ntt.co.jp
---
bswap.h|1 -
qemu-kvm.c | 30
The patch let host NIC driver to receive user space skb,
then the driver has chance to directly DMA to guest user
space buffers thru single ethX interface.
Signed-off-by: Xin Xiaohui xiaohui@intel.com
Signed-off-by: Zhao Yu yzha...@gmail.com
Sigend-off-by: Jeff Dike
Add a device to utilize the vhost-net backend driver for
copy-less data transfer between guest FE and host NIC.
It pins the guest user space to the host memory and
provides proto_ops as sendmsg/recvmsg to vhost-net.
Signed-off-by: Xin Xiaohui xiaohui@intel.com
Signed-off-by: Zhao Yu
The vhost-net backend now only supports synchronous send/recv
operations. The patch provides multiple submits and asynchronous
notifications. This is needed for zero-copy case.
Signed-off-by: Xin Xiaohui xiaohui@intel.com
---
drivers/vhost/net.c | 145
X86 emulator fails to do permission/correctness checking like
real CPU does for some instruction. This patch series fixes some
of those discrepancies.
Changelog:
v1-v2
- move IOPL permission checking functions into emulate.c
- rename them to more intuitive names
- fix tr segment limit
For some instructions CPU behaves differently for real-mode and
virtual 8086. Let emulator know which mode cpu is in, so it will
not poke into vcpu state directly.
Signed-off-by: Gleb Natapov g...@redhat.com
---
arch/x86/include/asm/kvm_emulate.h |1 +
arch/x86/kvm/emulate.c |
Use groups mechanism to decode 0F C7 instructions.
Signed-off-by: Gleb Natapov g...@redhat.com
---
arch/x86/kvm/emulate.c |9 +++--
1 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 435b1e4..45a4f7c 100644
---
Instructions which are not allowed to have LOCK prefix should
generate #UD if one is used.
Signed-off-by: Gleb Natapov g...@redhat.com
---
arch/x86/kvm/emulate.c | 85 +++-
1 files changed, 48 insertions(+), 37 deletions(-)
diff --git
Currently when x86 emulator needs to access memory, page walk is done with
broadest permission possible, so if emulated instruction was executed
by userspace process it can still access kernel memory. Fix that by
providing correct memory access to page walker during emulation.
Signed-off-by: Gleb
Use groups mechanism to decode 0F BA instructions.
Signed-off-by: Gleb Natapov g...@redhat.com
---
arch/x86/kvm/emulate.c |7 ++-
1 files changed, 6 insertions(+), 1 deletions(-)
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 645b245..435b1e4 100644
---
POPF behaves differently depending on current CPU mode. Emulate correct
logic to prevent guest from changing flags that it can't change otherwise.
Signed-off-by: Gleb Natapov g...@redhat.com
---
arch/x86/kvm/emulate.c | 55 +++-
1 files changed, 54
Make emulator check that vcpu is allowed to execute IN, INS, OUT,
OUTS, CLI, STI.
Signed-off-by: Gleb Natapov g...@redhat.com
---
arch/x86/include/asm/kvm_host.h |1 +
arch/x86/kvm/emulate.c | 89 +++---
arch/x86/kvm/x86.c | 10 ++---
Add CPL checking in case emulator is tricked into emulating
privilege instruction from userspace.
Signed-off-by: Gleb Natapov g...@redhat.com
---
arch/x86/kvm/emulate.c | 35 ---
1 files changed, 20 insertions(+), 15 deletions(-)
diff --git
Hi Avi, Izik,
today I spent some time investigating the problem and I've found the culprit:
RHEL5 glibc doesn't define MADV_MERGEABLE, so qemu-kvm didn't compile with KSM
support.
After I added -DMADV_MERGEABLE=12 -DMADV_UNMERGEABLE=13 CFLAGS,
things started to work.
Strange it worked for me
On 02/10/2010 02:21 PM, Gleb Natapov wrote:
X86 emulator fails to do permission/correctness checking like
real CPU does for some instruction. This patch series fixes some
of those discrepancies.
Applied all, thanks.
--
error compiling committee.c: too many arguments to function
--
To
Add documentation for indirect descriptors
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
virtio-spec.lyx | 118 +--
1 files changed, 114 insertions(+), 4 deletions(-)
diff --git a/virtio-spec.lyx b/virtio-spec.lyx
index
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 02/10/2010 02:52 AM, OHMURA Kei wrote:
for (i = 0; i len; i++) {
-c = bitmap[i];
-while (c 0) {
-j = ffsl(c) - 1;
-c = ~(1u j);
-page_number = i * 8 + j;
-addr1 =
On 02/10/2010 12:52 PM, OHMURA Kei wrote:
dirty-bitmap-traveling is carried out by byte size in qemu-kvm.c.
But We think that dirty-bitmap-traveling by long size is faster than by byte
size especially when most of memory is not dirty.
--- a/bswap.h
+++ b/bswap.h
@@ -209,7 +209,6 @@ static
On Wednesday 10 February 2010, Xin Xiaohui wrote:
The idea is simple, just to pin the guest VM user space and then
let host NIC driver has the chance to directly DMA to it.
The patches are based on vhost-net backend driver. We add a device
which provides proto_ops as sendmsg/recvmsg to
Le mercredi 10 février 2010 à 19:48 +0800, Xin Xiaohui a écrit :
Add a device to utilize the vhost-net backend driver for
copy-less data transfer between guest FE and host NIC.
It pins the guest user space to the host memory and
provides proto_ops as sendmsg/recvmsg to vhost-net.
On Wed, Feb 10, 2010 at 10:45:41AM +0900, Takuya Yoshikawa wrote:
My motivation: What I want to achive by this
work is to make the basic style of x86 emulator
better for the following developments. Actually
unless we handle the fault properly, our works
implementing each instruction's
Bugs item #2944508, was opened at 2010-02-02 09:46
Message generated for change (Comment added) made by nsnkern
You can respond by visiting:
https://sourceforge.net/tracker/?func=detailatid=893831aid=2944508group_id=180599
Please note that this message will contain a full copy of the comment
On 02/10/2010 07:20 AM, Avi Kivity wrote:
On 02/10/2010 12:52 PM, OHMURA Kei wrote:
dirty-bitmap-traveling is carried out by byte size in qemu-kvm.c.
But We think that dirty-bitmap-traveling by long size is faster than by byte
size especially when most of memory is not dirty.
---
On 02/10/2010 07:20 AM, Avi Kivity wrote:
On 02/10/2010 12:52 PM, OHMURA Kei wrote:
dirty-bitmap-traveling is carried out by byte size in qemu-kvm.c.
But We think that dirty-bitmap-traveling by long size is faster than by byte
size especially when most of memory is not dirty.
---
On 02/10/2010 05:54 PM, Anthony Liguori wrote:
On 02/10/2010 07:20 AM, Avi Kivity wrote:
On 02/10/2010 12:52 PM, OHMURA Kei wrote:
dirty-bitmap-traveling is carried out by byte size in qemu-kvm.c.
But We think that dirty-bitmap-traveling by long size is faster than by byte
size
Anthony Liguori wrote:
On 02/10/2010 07:20 AM, Avi Kivity wrote:
On 02/10/2010 12:52 PM, OHMURA Kei wrote:
dirty-bitmap-traveling is carried out by byte size in qemu-kvm.c.
But We think that dirty-bitmap-traveling by long size is faster than by byte
size especially when most of
FIX:
Patch is based on The previous version [PATCH] KVM test: KSM (kernel shared
memory) test
developed overcommit Lucas Meneghel Rodrigues.
The only fix is change of behavior with large overcommit. Python has a
problem in allocating
memory even when is still enough space in swap
Hello everyone,
I like to submit a set of patches to enable installing multiple guests
in parallel.
To install multiple guests in parallel the role of the host and the
guest need to be changed. In this patch host acts as a client and the
guest acts as server, enabling us to install multiple
On Fri, 2010-01-29 at 18:17 +0800, Feng Yang wrote:
Add check_image.py in scripts folder. It will performance qemu-img info
and qemu-img check.
Add check_image.py as post command of qcow2 variant.
Hi Feng, thanks for your patch! See some comments below:
Signed-off-by: Feng Yang
On 02/10/2010 03:50 AM, Takuya Yoshikawa wrote:
This patch injects page fault when reading descriptor in
load_guest_segment_descriptor() fails with FAULT.
Effects of this injection: This function is used by
kvm_load_segment_descriptor() which is necessary for the
following instructions.
- mov
On Wed, 2010-02-10 at 11:13 -0500, Jiri Zupka wrote:
FIX:
Patch is based on The previous version [PATCH] KVM test: KSM (kernel shared
memory) test
developed overcommit Lucas Meneghel Rodrigues.
The only fix is change of behavior with large overcommit. Python has a
problem in
On Wed, Feb 10, 2010 at 09:02:00AM +0200, Avi Kivity wrote:
On 02/09/2010 10:58 PM, Marcelo Tosatti wrote:
You're right... this should be enough to avoid a stop with uncomplete
PIO (and this is what happens for MMIO already). The signal will not
be dequeued, so KVM will complete_pio and exit
From b2ddf7e9d434ea33619ced976f61e59d2d20ab71 Mon Sep 17 00:00:00 2001
From: yogananth subramanian anant...@linux.vnet.ibm.com
Date: Wed, 10 Feb 2010 06:15:29 -0500
Subject: [PATCH 1/5] kvm test: unattended install: change host role from
server to client
By changing the role of the host from
From 76aff8427266035867e9c73f8c2c7037a7113d1e Mon Sep 17 00:00:00 2001
From: yogananth subramanian anant...@linux.vnet.ibm.com
Date: Wed, 10 Feb 2010 08:00:06 -0500
Subject: [PATCH 2/5] kvm test: test_base: enabling parallel install
Have added a new port redirection entry
From 0aa8ed40de86eb8ad5b324177b2e0986e1c32c77 Mon Sep 17 00:00:00 2001
From: yogananth subramanian anant...@linux.vnet.ibm.com
Date: Wed, 10 Feb 2010 09:45:58 -0500
Subject: [PATCH 3/5] kvm test: unattended: add support for parallel
install of suse and other guest OS
File extension of the
From 2dbcd5801c31e35a01a671082ba0a461e9d44a5c Mon Sep 17 00:00:00 2001
From: yogananth subramanian anant...@linux.vnet.ibm.com
Date: Wed, 10 Feb 2010 10:02:07 -0500
Subject: [PATCH 4/5] kvm test: kickstart: modified ks.cfg file to suport
parallel install
The guest acts as a server, rather then a
From e9eb6ba4aa481646ac4fddb32c85423e529ff717 Mon Sep 17 00:00:00 2001
From: yogananth subramanian anant...@linux.vnet.ibm.com
Date: Wed, 10 Feb 2010 10:53:45 -0500
Subject: [PATCH 5/5] kvm test: support for sles11 parallel install
Adds new entry SUSE in test_base file for sles and also
contains
On 02/10/2010 10:00 AM, Alexander Graf wrote:
On PPC the bitmap is Little Endian.
Out of curiousity, why? It seems like an odd interface.
Regards,
Anthony Liguori
--
To unsubscribe from this list: send the line unsubscribe kvm in
the body of a message to majord...@vger.kernel.org
More
On 02/10/2010 06:25 PM, Marcelo Tosatti wrote:
On Wed, Feb 10, 2010 at 09:02:00AM +0200, Avi Kivity wrote:
On 02/09/2010 10:58 PM, Marcelo Tosatti wrote:
You're right... this should be enough to avoid a stop with uncomplete
PIO (and this is what happens for MMIO already). The signal
On Wed, Feb 10, 2010 at 06:25:42PM +0200, Avi Kivity wrote:
On 02/10/2010 03:50 AM, Takuya Yoshikawa wrote:
This patch injects page fault when reading descriptor in
load_guest_segment_descriptor() fails with FAULT.
Effects of this injection: This function is used by
Anthony Liguori wrote:
On 02/10/2010 10:00 AM, Alexander Graf wrote:
On PPC the bitmap is Little Endian.
Out of curiousity, why? It seems like an odd interface.
Because on PPC, you usually run PPC32 userspace code on a PPC64 kernel.
Unlike with x86, there's no real benefit
On 02/10/2010 06:35 PM, Anthony Liguori wrote:
On 02/10/2010 10:00 AM, Alexander Graf wrote:
On PPC the bitmap is Little Endian.
Out of curiousity, why? It seems like an odd interface.
Exactly this issue. If you specify it as unsigned long native endian,
there is ambiguity
On 02/10/2010 06:43 PM, Alexander Graf wrote:
Out of curiousity, why? It seems like an odd interface.
Because on PPC, you usually run PPC32 userspace code on a PPC64 kernel.
Unlike with x86, there's no real benefit in using 64 bit userspace.
btw, does 32-bit ppc qemu support
Avi Kivity wrote:
On 02/10/2010 06:43 PM, Alexander Graf wrote:
Out of curiousity, why? It seems like an odd interface.
Because on PPC, you usually run PPC32 userspace code on a PPC64 kernel.
Unlike with x86, there's no real benefit in using 64 bit userspace.
On Wed, Feb 10, 2010 at 06:25:42PM +0200, Avi Kivity wrote:
On 02/10/2010 03:50 AM, Takuya Yoshikawa wrote:
This patch injects page fault when reading descriptor in
load_guest_segment_descriptor() fails with FAULT.
Effects of this injection: This function is used by
Avi Kivity wrote:
On 02/10/2010 06:25 PM, Marcelo Tosatti wrote:
On Wed, Feb 10, 2010 at 09:02:00AM +0200, Avi Kivity wrote:
On 02/09/2010 10:58 PM, Marcelo Tosatti wrote:
You're right... this should be enough to avoid a stop with uncomplete
PIO (and this is what happens for MMIO
On 02/10/2010 06:47 PM, Alexander Graf wrote:
Because on PPC, you usually run PPC32 userspace code on a PPC64 kernel.
Unlike with x86, there's no real benefit in using 64 bit userspace.
btw, does 32-bit ppc qemu support large memory guests? It doesn't on
x86, and I don't
Avi Kivity wrote:
On 02/10/2010 06:47 PM, Alexander Graf wrote:
Because on PPC, you usually run PPC32 userspace code on a PPC64 kernel.
Unlike with x86, there's no real benefit in using 64 bit userspace.
btw, does 32-bit ppc qemu support large memory guests? It
netperf results showed about 20% transmission gain with this patch for
virtio_net.
Thanks
Shirley
--
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 02/10/2010 03:45 AM, Takuya Yoshikawa wrote:
This patch set consists of macro replacements
and some fixes of fault handling in the x86
emulator.
Suggested by Marcelo, I separated these two
works and tried to make it clear what effects
each patch will produce: if you think that
reordering
On Wed, Feb 10, 2010 at 06:57:05PM +0200, Avi Kivity wrote:
On 02/10/2010 03:45 AM, Takuya Yoshikawa wrote:
This patch set consists of macro replacements
and some fixes of fault handling in the x86
emulator.
Suggested by Marcelo, I separated these two
works and tried to make it clear
On 02/10/2010 06:52 PM, Alexander Graf wrote:
Hrm, trying to read the thread I'm still somewhat lost. What exactly do
you want to document?
The problem: if KVM_RUN exits with KVM_EXIT_MMIO or KVM_EXIT_IO, then
the internal state is inconsistent. The instruction is only half
completed,
Avi Kivity wrote:
On 02/10/2010 06:52 PM, Alexander Graf wrote:
Hrm, trying to read the thread I'm still somewhat lost. What exactly do
you want to document?
The problem: if KVM_RUN exits with KVM_EXIT_MMIO or KVM_EXIT_IO, then
the internal state is inconsistent. The instruction is
On Wed, Feb 10, 2010 at 07:01:46PM +0200, Avi Kivity wrote:
On 02/10/2010 06:52 PM, Alexander Graf wrote:
Hrm, trying to read the thread I'm still somewhat lost. What exactly do
you want to document?
The problem: if KVM_RUN exits with KVM_EXIT_MMIO or KVM_EXIT_IO,
then the internal
On 02/10/2010 07:03 PM, Alexander Graf wrote:
I see. Yes, that works for PPC Book3S too. We check for signals on the
beginning of vcpu_run. I'm not sure about BookE though.
Either way - wouldn't it make more sense to just move the check to
generic code?
Well, the check happens very deep
On 02/10/2010 07:07 PM, Gleb Natapov wrote:
On Wed, Feb 10, 2010 at 07:01:46PM +0200, Avi Kivity wrote:
On 02/10/2010 06:52 PM, Alexander Graf wrote:
Hrm, trying to read the thread I'm still somewhat lost. What exactly do
you want to document?
The problem: if KVM_RUN exits
* Yogananth Subramanian anant...@linux.vnet.ibm.com [2010-02-10 10:30]:
From 0aa8ed40de86eb8ad5b324177b2e0986e1c32c77 Mon Sep 17 00:00:00 2001
From: yogananth subramanian anant...@linux.vnet.ibm.com
Date: Wed, 10 Feb 2010 09:45:58 -0500
Subject: [PATCH 3/5] kvm test: unattended: add support
Bugs item #2944508, was opened at 2010-02-02 03:46
Message generated for change (Comment added) made by iggy_cav
You can respond by visiting:
https://sourceforge.net/tracker/?func=detailatid=893831aid=2944508group_id=180599
Please note that this message will contain a full copy of the comment
On Wed, 2010-02-10 at 21:53 +0530, yogi wrote:
Hello everyone,
I like to submit a set of patches to enable installing multiple guests
in parallel.
Hi Yogi, great to receive your patches!
To install multiple guests in parallel the role of the host and the
guest need to be changed. In this
On Wed, 2010-02-10 at 21:59 +0530, yogi wrote:
From 0aa8ed40de86eb8ad5b324177b2e0986e1c32c77 Mon Sep 17 00:00:00 2001
From: yogananth subramanian anant...@linux.vnet.ibm.com
Date: Wed, 10 Feb 2010 09:45:58 -0500
Subject: [PATCH 3/5] kvm test: unattended: add support for parallel
install of
From: Eddie Dong eddie.d...@intel.com
Move Double-Fault generation logic out of page fault
exception generating function to cover more generic case.
Signed-off-by: Eddie Dong eddie.d...@intel.com
Signed-off-by: Gleb Natapov g...@redhat.com
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
In the past we've had errors of single-bit in the other two cases; the
printk() may confirm it for the third case (many-many).
Signed-off-by: Avi Kivity a...@redhat.com
---
arch/x86/kvm/mmu.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/arch/x86/kvm/mmu.c
Instead of specifying the bits which we want to trap on, specify the bits
which we allow the guest to change transparently. This is safer wrt future
changes to cr4.
Signed-off-by: Avi Kivity a...@redhat.com
---
arch/x86/kvm/vmx.c | 10 ++
1 files changed, 6 insertions(+), 4
From: Sheng Yang sh...@linux.intel.com
We don't support these instructions, but guest can execute them even if the
feature('monitor') haven't been exposed in CPUID. So we would trap and inject
a #UD if guest try this way.
Cc: sta...@kernel.org
Signed-off-by: Sheng Yang sh...@linux.intel.com
This is the first of four batches of patches for the 2.6.34 merge window. KVM
changes for this cycle include:
- rdtscp support
- powerpc server-class updates
- much improved large-guest scaling (now up to 64 vcpus)
- improved guest fpu handling
- initial Hyper-V emulation
- better swapping
From: Marcelo Tosatti mtosa...@redhat.com
Windows 2003 uses task switch to triple fault and reboot (the other
exception being reserved pdptrs bits).
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
arch/x86/kvm/x86.c | 11 +++
1 files changed, 11 insertions(+), 0 deletions(-)
From: Sheng Yang sh...@linux.intel.com
KVM need vsyscall_init() to initialize MSR_TSC_AUX before it read the value.
Per Avi's suggestion, this patch raised vsyscall priority on hotplug notifier
chain, to 30.
CC: Ingo Molnar mi...@elte.hu
CC: linux-ker...@vger.kernel.org
Signed-off-by: Sheng Yang
From: Marcelo Tosatti mtosa...@redhat.com
Using a similar two-step procedure as for memslots.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
arch/x86/include/asm/kvm_host.h |4 ++
arch/x86/kvm/x86.c | 60 +--
From: Marcelo Tosatti mtosa...@redhat.com
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
arch/ia64/kvm/kvm-ia64.c |4 +-
arch/x86/kvm/i8254.c |6 +-
arch/x86/kvm/i8259.c |4 +-
arch/x86/kvm/x86.c| 13 +++---
include/linux/kvm_host.h | 27 ++--
From: Marcelo Tosatti mtosa...@redhat.com
So its possible to iommu map a memslot before making it visible to
kvm.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
include/linux/kvm_host.h |3 +--
virt/kvm/iommu.c | 13 ++---
virt/kvm/kvm_main.c |2 +-
3
From: Marcelo Tosatti mtosa...@redhat.com
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
arch/ia64/kvm/kvm-ia64.c |4 ++--
arch/powerpc/kvm/book3s.c |4 ++--
arch/x86/kvm/i8254.c |2 +-
arch/x86/kvm/i8259.c |4 ++--
arch/x86/kvm/vmx.c|8
From: Marcelo Tosatti mtosa...@redhat.com
Have a pointer to an allocated region inside struct kvm.
[alex: fix ppc book 3s]
Signed-off-by: Alexander Graf ag...@suse.de
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
arch/ia64/kvm/kvm-ia64.c | 10 ++
arch/powerpc/kvm/book3s.c |
Some exit reasons missed their strings; fill out the table.
Signed-off-by: Avi Kivity a...@redhat.com
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
arch/x86/kvm/vmx.c | 58 +++-
1 files changed, 39 insertions(+), 19 deletions(-)
diff
From: Alexander Graf ag...@suse.de
We treated the DEC interrupt like an edge based one. This is not true for
Book3s. The DEC keeps firing until mtdec is issued again and thus clears
the interrupt line.
So let's implement this logic in KVM too. This patch moves the line clearing
from the firing
From: Marcelo Tosatti mtosa...@redhat.com
Required for SRCU convertion later.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
arch/ia64/kvm/kvm-ia64.c | 16 ++---
arch/powerpc/kvm/powerpc.c | 18 ---
arch/s390/kvm/kvm-s390.c | 25 +---
When the guest acknowledges an interrupt, it sends an EOI message to the local
apic, which broadcasts it to the ioapic. To handle the EOI, we need to take
the ioapic mutex.
On large guests, this causes a lot of contention on this mutex. Since large
guests usually don't route interrupts via the
From: Marcelo Tosatti mtosa...@redhat.com
Use two steps for memslot deletion: mark the slot invalid (which stops
instantiation of new shadow pages for that slot, but allows destruction),
then instantiate the new empty slot.
Also simplifies kvm_handle_hva locking.
Signed-off-by: Marcelo Tosatti
From: Marcelo Tosatti mtosa...@redhat.com
Which takes a memslot pointer instead of using kvm-memslots.
To be used by SRCU convertion later.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
include/linux/kvm_host.h |2 ++
virt/kvm/kvm_main.c | 33
From: Alexander Graf ag...@suse.de
Progress on KVM for Embedded PowerPC has stalled, but for Book3S there's quite
a lot of work to do and going on.
So in agreement with Hollis and Avi, we should switch maintainers for PowerPC.
Signed-off-by: Alexander Graf ag...@suse.de
Acked-by: Hollis
With slots_lock converted to rcu, the entire kvm hotpath on modern processors
(with npt or ept) now scales beautifully. Increase the maximum vcpu count to
64 to reflect this.
Signed-off-by: Avi Kivity a...@redhat.com
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
From: Marcelo Tosatti mtosa...@redhat.com
Have a pointer to an allocated region inside x86's kvm_arch.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
arch/x86/include/asm/kvm_host.h |8 ++--
arch/x86/kvm/x86.c | 21 -
2 files changed, 22
Signed-off-by: Avi Kivity a...@redhat.com
---
virt/kvm/kvm_main.c | 20 +++-
1 files changed, 15 insertions(+), 5 deletions(-)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 92f5c98..c10d117 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -341,6
- add destructor function
- move related allocation into constructor
- add stubs for !CONFIG_KVM_MMIO
Signed-off-by: Avi Kivity a...@redhat.com
---
virt/kvm/coalesced_mmio.c | 25 +++--
virt/kvm/coalesced_mmio.h | 10 ++
virt/kvm/kvm_main.c |7 +--
3
From: Marcelo Tosatti mtosa...@redhat.com
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
arch/x86/kvm/x86.c | 49 +
1 files changed, 41 insertions(+), 8 deletions(-)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index
From: Wu Fengguang fengguang...@intel.com
Signed-off-by: Wu Fengguang fengguang...@intel.com
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
---
Documentation/kvm/api.txt | 12 ++--
1 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/Documentation/kvm/api.txt
From: Alexander Graf ag...@suse.de
Because we now emulate the DEC interrupt according to real life behavior,
there's no need to keep the AGGRESSIVE_DEC hack around.
Let's just remove it.
Signed-off-by: Alexander Graf ag...@suse.de
Acked-by: Acked-by: Hollis Blanchard hol...@penguinppc.org
1 - 100 of 139 matches
Mail list logo