On Wed, Jan 09, 2008 at 07:46:04PM +0100, Alexander Graf wrote:
> Darwin requires SSSE3 in 64-bit mode, so KVM needs to reveal the
> appropriate CPUID flag. I could not find a proper define in the kernel
> headers, so I just added the specific bit number.
> Index: kernel/x86.c
> ==
Hi, Marcelo
>+struct balloon_buf *alloc_balloon_buf(struct virtio_device *vdev, gfp_t
>flags)
>+{
>+ struct balloon_buf *buf;
>+
>+ buf = kzalloc(sizeof(struct balloon_buf), flags);
>+ if (!buf)
>+ dev_printk(KERN_ERR, &vdev->dev, "%s: alloc fail\n", __func__);
>+
>+
Hollis Blanchard wrote:
> Add an "is_bigendian" flag to the kvm_run.mmio structure.
>
> This is needed for architectures that can make both little- and
> big-endian memory accesses.
>
> Signed-off-by: Hollis Blanchard <[EMAIL PROTECTED]>
> ---
>
> PowerPC has different instructions for native and b
Darwin requires SSSE3 in 64-bit mode, so KVM needs to reveal the
appropriate CPUID flag. I could not find a proper define in the kernel
headers, so I just added the specific bit number.
Index: kernel/x86.c
===
--- kernel.orig/x86.c
+
On Wednesday 09 January 2008, Dor Laor wrote:
> Some figures: Linux rx 350Mbps, tx 150bps, Windows rx 700mbps, tx 100 mbps.
very nice!
in a related note, the VMWare tools package, which is supposed to 'enhance
performance' by installing 'specially tuned' drivers into a guest, doesn't
include ne
Hollis Blanchard wrote:
> There are a bunch of x86-specific fields in struct kvm_run:
> /* in */
> __u8 request_interrupt_window;
> __u8 padding1[7];
>
> /* out */
> __u8 if_flag;
> __u8 padding2[2];
>
> /* in (pre_kvm_run), out (post_kvm_run) */
>
There are a bunch of x86-specific fields in struct kvm_run:
/* in */
__u8 request_interrupt_window;
__u8 padding1[7];
/* out */
__u8 if_flag;
__u8 padding2[2];
/* in (pre_kvm_run), out (post_kvm_run) */
__u64 cr8;
__u64 apic_
Add an "is_bigendian" flag to the kvm_run.mmio structure.
This is needed for architectures that can make both little- and
big-endian memory accesses.
Signed-off-by: Hollis Blanchard <[EMAIL PROTECTED]>
---
PowerPC has different instructions for native and byte-reversed memory
accesses, and some
On Wed, 9 Jan 2008, Andrea Arcangeli wrote:
> This patch is a first basic implementation of the mmu notifiers. More
> methods can be added in the future.
>
> In short when the linux VM decides to free a page, it will unmap it
> from the linux pagetables. However when a page is mapped not just by
>Windows Vista recognizes it but there is no traffic.
I should note, that there is no traffic only with Microsoft's built-in
drivers. Vista works with e1000 drivers downloaded from www.intel.com
Additionally, tested Linux 2.4 guest kernel (Mandrake Linux 9.2) and Solaris
10 (U3) guest - they work
This patch adds support for the MSR MSR_IA32_PERF_STATUS. This is
required to boot Darwin.
I only implemented dummy support, as the real value would be of no
meaning to any OS I know so far anyway and is only implemented on very
recent Intel CPUs.
Index: kernel/x86.c
==
Darwin requires SSSE3 in 64-bit mode, so KVM needs to reveal the
appropriate CPUID flag. I could not find a proper define in the kernel
headers, so I just added the specific bit number.
Index: kernel/x86.c
===
--- kernel.orig/x86.c
+++
QEMU support for ballooning.
Changes:
- Use int64_t for memory size input, otherwise it wraps around
at 2GB (Daniel Berrange)
- Use virtio_update_config()
- Remove save/load hooks, they will be rewritten after Anthony introduces
proper virtio s/l support
Index: kvm-userspace/libkvm/libkvm.c
=
Hello,
This patch is a first basic implementation of the mmu notifiers. More
methods can be added in the future.
In short when the linux VM decides to free a page, it will unmap it
from the linux pagetables. However when a page is mapped not just by
the regular linux ptes, but also from the shado
Following patch adds virtio_update_config() wrapper to trigger an IRQ on
config space updates.
The write callbacks also need to be hooked.
Index: kvm-userspace/qemu/hw/virtio.c
===
--- kvm-userspace.orig/qemu/hw/virtio.c
+++ kvm-us
Export nukeing of shadow MMU pages through an ioctl, required for
consistency on balloon inflate.
Changes:
- Move to x86-specific kvm_arch_vm_ioctl()
- Grab mmap_sem
Index: kvm.quilt/arch/x86/kvm/x86.c
===
--- kvm.quilt.orig/arch/x
Updated balloon driver.
Changes from last version:
- Get rid of global variables/structure
- Use page->lru to link ballooned pages
- Use dev_dbg/dev_printk
- Proper kthread_should_stop handling
- Move shared definitions to separate header
- Use ->config_changed method for notification
This depen
On Tue, Jan 08, 2008 at 10:50:00AM +0200, Avi Kivity wrote:
> Yes, you can press the VM's power button (system_powerdown in the
> monitor) and it will invoke the OS shutdown sequence.
>
> Only worked on Linux in my testing though.
It works for me _when_ I pass --no-kvm-irqchip. So it seems Window
On Wed, 2008-01-09 at 10:39 +0200, Avi Kivity wrote:
> Zhao, Yunfeng wrote:
> > Today's tip fails to compile with following error messages:
> >
>
> [...]
>
> Fixed now. Hollis, I had to revert your asm symlink patch to fix this,
> so we need to find an alternative solution to that problem.
Subject: [PATCH 1/3] Add OBJCOPY to kvm-userspace/user/configure output
From: Hollis Blanchard <[EMAIL PROTECTED]>
From: Christian Ehrhardt <[EMAIL PROTECTED]>
Adds objectcopy to the configure script because it is needed for some kvmctl
implementations for a simple binary loader.
Signed-off-by: H
tired of people mocking your tiny dick? Change that forever with V-P-X-L!
http://whatyouaree.com/
aditehca adlf adasah
adnolb addersgr adamsepl
adonisse adisoyz adnimaap
adoiling a"digtes adartson
-
Check out the new SourceF
Avi Kivity wrote:
> Dong, Eddie wrote:
> If migration happens while rip is in the hypercall page, and if
> the
>
>
I didn't quit catch here. The source VM vCPU is in Qemu migration
part, The target VM VCPU is always waiting for migration
data/command. If you mean S
Christian Ehrhardt wrote:
> Subject: [PATCH 3/3] Add dcr read/write to the PowerPC kvmctl implementation
> From: Christian Ehrhardt <[EMAIL PROTECTED]>
>
> Adds simple functions to catch dcr access while executing testcases printing
> information about that access when triggered.
>
>
Applied all
Dong, Eddie wrote:
If migration happens while rip is in the hypercall page, and if the
>>> I didn't quit catch here. The source VM vCPU is in Qemu migration
>>> part, The target VM VCPU is always waiting for migration
>>> data/command. If you mean SMP case, all target VCPUs
Subject: [PATCH 2/3] Add PowerPC kvmctl harness and fix building the testcases
From: Hollis Blanchard <[EMAIL PROTECTED]>
From: Christian Ehrhardt <[EMAIL PROTECTED]>
These patches get the kvm-userspace/user/ test harness building and somewhat
working for PowerPC. For the moment it's just a simply
Subject: [PATCH 3/3] Add dcr read/write to the PowerPC kvmctl implementation
From: Christian Ehrhardt <[EMAIL PROTECTED]>
Adds simple functions to catch dcr access while executing testcases printing
information about that access when triggered.
Signed-off-by: Christian Ehrhardt <[EMAIL PROTECTED]
>>> If migration happens while rip is in the hypercall page, and if the
>>>
>>
>> I didn't quit catch here. The source VM vCPU is in Qemu migration
>> part, The target VM VCPU is always waiting for migration
>> data/command. If you mean SMP case, all target VCPUs are in waiting
>> for data/cmd,
Dor Laor wrote:
> On Wed, 2008-01-09 at 08:29 -0600, Anthony Liguori wrote:
>> Dor Laor wrote:
>>> Now that we have a host timer based tx wakeup it waits for 64
>>> packets or timeout before processing them.
>>> This might cause the guest to run out of tx buffers while the host
>>> holds them up.
>
On Wed, 2008-01-09 at 08:29 -0600, Anthony Liguori wrote:
> Dor Laor wrote:
> > Now that we have a host timer based tx wakeup it waits for 64
> > packets or timeout before processing them.
> > This might cause the guest to run out of tx buffers while the host
> > holds them up.
> >
>
> There's
Glauber de Oliveira Costa wrote:
>
> That said, if acpi is really the preference here, and this patches
> have chance, no problem. But it will take me a little more time to
> implement them ;-)
The power button support that was recently added at least proves that
the host->guest notification pa
Dor Laor wrote:
> Now that we have a host timer based tx wakeup it waits for 64
> packets or timeout before processing them.
> This might cause the guest to run out of tx buffers while the host
> holds them up.
>
There's a proper fix that Rusty added last night.
> This is a temporal solution t
Christian Borntraeger wrote:
> Am Mittwoch, 9. Januar 2008 schrieb Glauber de Oliveira Costa:
>> I'm sending a first draft of my proposed cpu hotplug driver for kvm/virtio
>> The first patch is the kernel module, while the second, the userspace pci
> device.
>
> I personally prefer to use non par
Avi Kivity wrote:
> Glauber de Oliveira Costa wrote:
>> I'm sending a first draft of my proposed cpu hotplug driver for
>> kvm/virtio
>> The first patch is the kernel module, while the second, the userspace
>> pci device.
>>
>> The host boots with the maximum cpus it should ever use, through the
Now that we have a host timer based tx wakeup it waits for 64
packets or timeout before processing them.
This might cause the guest to run out of tx buffers while the host
holds them up.
This is a temporal solution to quickly bring back performance to 800mbps.
But a better fix will soon be sent (i
This includes the comments from Jerone. Moving the targets back into the
Makefile e.g. easen things for people looking at the Makefile for the first
time and looking for targets. So here v2 of the patch.
changes to v1:
- defining per arch targets via variables in config-$arch
- removed that user
Am Mittwoch, 9. Januar 2008 schrieb Glauber de Oliveira Costa:
> I'm sending a first draft of my proposed cpu hotplug driver for kvm/virtio
> The first patch is the kernel module, while the second, the userspace pci
device.
I personally prefer to use non paravirtualized cpu hotplug and implement
On Wed, Jan 09, 2008 at 11:06:21AM +0100, Andrea Arcangeli wrote:
> On Tue, Jan 08, 2008 at 09:42:13AM -0600, Anthony Liguori wrote:
> > Instead of allocating a node for each page, you could use page->private
>
> page->lru is probably better for this so splice still works
> etc... (the struct pag
Glauber de Oliveira Costa wrote:
> I'm sending a first draft of my proposed cpu hotplug driver for kvm/virtio
> The first patch is the kernel module, while the second, the userspace pci
> device.
>
> The host boots with the maximum cpus it should ever use, through the -smp
> parameter.
> Due to r
Signed-off-by: Glauber de Oliveira Costa <[EMAIL PROTECTED]>
---
qemu/Makefile.target |2 +-
qemu/hw/pc.c |4 +-
qemu/hw/pc.h |3 +
qemu/hw/virtio-hotplug.c | 111 ++
qemu/monitor.c | 11 +
qemu/qe
Signed-off-by: Glauber de Oliveira Costa <[EMAIL PROTECTED]>
---
drivers/virtio/Kconfig |6 +
drivers/virtio/Makefile |1 +
drivers/virtio/virtio_cpu.c | 226 +++
drivers/virtio/virtio_pci.c |1 +
kernel/cpu.c|2 +
5
I'm sending a first draft of my proposed cpu hotplug driver for kvm/virtio
The first patch is the kernel module, while the second, the userspace pci
device.
The host boots with the maximum cpus it should ever use, through the -smp
parameter.
Due to real machine constraints (which qemu copies), i
Jerone Young wrote:
> # HG changeset patch
> # User Jerone Young <[EMAIL PROTECTED]>
> # Date 1199827424 21600
> # Node ID bff590d13f32071b50d79f511424821847ed5ba7
> # Parent a6e6b30d5f4154fd91dbf68e7e3986b2af02ee0f
> Add function kvm_create_kernel_phys_mem to libkvm-powerpc.c
>
> Function kvm_cre
On Tue, Jan 08, 2008 at 09:42:13AM -0600, Anthony Liguori wrote:
> Instead of allocating a node for each page, you could use page->private
page->lru is probably better for this so splice still works
etc... (the struct page isn't visible to the guest VM so it's free to
use)
--
Hollis Blanchard wrote:
> +
> +static int misc_init(void)
> +{
> + int err;
> +
> + err = io_table_register(&pio_table, 0xff, 1, misc_io, NULL);
> + if (err < 0)
> + return err;
> +
> + err = io_table_register(&pio_table, 0xf1, 1, misc_io, NULL);
> + if (err < 0)
> +
Zhao, Yunfeng wrote:
> Today's tip fails to compile with following error messages:
>
[...]
Fixed now. Hollis, I had to revert your asm symlink patch to fix this,
so we need to find an alternative solution to that problem.
--
error compiling committee.c: too many arguments to function
---
45 matches
Mail list logo