Zhao Forrest wrote:
http://ols.108.redhat.com/2007/Reprints/kivity-Reprint.pdf
Hi Avi,
I have a question about KVM architecture after reading your paper.
It reads:
..
At the kernel level, the kernel causes the hardware
to enter guest mode. If the processor exits guest
mode due
Glauber Costa wrote:
apic is not acpi, although they are acronyms. Due to a confusion of
mine, those things were mixed, leading to a bug reported at
https://sourceforge.net/tracker/index.php?func=detailaid=1903732group_id=180599atid=893831
This patch fixes it, by assigning smp_cpus
Running kvm 62 on 2.6.24.3 (the extraversion is reported due to a local
patch to vfat) I just got a couple of soft lockups and a hang. I was
installing FreeBSD 7 at the time.
I am trying again with v2.6.25-rc3-369-g976dde0.
Here are the wossnames.
Mar 4 21:20:12 burly kernel: BUG: soft
Thanks for your detailed explanation :). That's quite helpful for me
to understand KVM internals.
If this is the case, see the below example:
1 physical NIC interrupt is received on physical CPU 0 and host kernel
determines that this is a network packet targeted to the emulated NIC
for a
Zhao Forrest wrote:
The incoming packet is processed by the host ethernet stack; it is
forwarded to the bridge, which forwards it to the tap. When the tap
queues the packet, it sends a signal to qemu (since the tap file
descriptor has a signal associated). When the kernel delivers the
Paul Collins wrote:
Running kvm 62 on 2.6.24.3 (the extraversion is reported due to a local
patch to vfat) I just got a couple of soft lockups and a hang. I was
installing FreeBSD 7 at the time.
Mar 4 21:20:13 burly kernel: Call Trace:
Mar 4 21:20:13 burly kernel: [8024b60e]
Marcelo Tosatti wrote:
Hi Avi,
Looks nice.
On Sun, Mar 02, 2008 at 06:31:17PM +0200, Avi Kivity wrote:
+int kvm_pv_mmu_op(struct kvm_vcpu *vcpu, unsigned long bytes,
+ gpa_t addr, unsigned long *ret)
+{
+int r;
+struct kvm_pv_mmu_op_buffer buffer;
Perhaps
Avi Kivity wrote:
As the patchset (less cr3 caching) passed the regression tests I'll
apply it.
Forgot about the unresolved CONFIG_HIGHPTE problem, so not applying for now.
--
error compiling committee.c: too many arguments to function
From f389ff3bc38b1684f15d8073c6d9abd2f2c9 Mon Sep 17 00:00:00 2001
From: Sheng Yang [EMAIL PROTECTED]
Date: Mon, 28 Jan 2008 05:10:22 +0800
Subject: [PATCH] KVM: In kernel pit model
The patch moved PIT from userspace to kernel, and increase the timer accuracy
greatly.
Signed-off-by: Sheng
From 22b56da8fe8f8e7d38075956bef9205f1dab829d Mon Sep 17 00:00:00 2001
From: Sheng Yang [EMAIL PROTECTED]
Date: Tue, 4 Mar 2008 09:45:16 +0800
Subject: [PATCH] kvm: qemu: Add save/restore support for in kernel PIT
Signed-off-by: Sheng Yang [EMAIL PROTECTED]
---
qemu/hw/i8254.c | 73
Hi
Here is the last in-kernel PIT patch for KVM. The mainly change from last
version is the supporting to save/restore. I also tested live migration.
The other modifies including some date structure changed to be better for
supporting the save/restore. I moved the PIT timer to outside of
From 79f06bc838534fc916f9ca90f89b7778b8e64dfe Mon Sep 17 00:00:00 2001
From: Sheng Yang [EMAIL PROTECTED]
Date: Tue, 4 Mar 2008 09:42:27 +0800
Subject: [PATCH] kvm: libkvm: Add interface for PIT save/restore supporting
Signed-off-by: Sheng Yang [EMAIL PROTECTED]
---
libkvm/libkvm-x86.c | 30
From ac2c08ee7a3bcceb8dede74d0d8f1dce0fea8956 Mon Sep 17 00:00:00 2001
From: Sheng Yang [EMAIL PROTECTED]
Date: Tue, 4 Mar 2008 08:50:59 +0800
Subject: [PATCH] KVM: Add save/restore supporting of in kernel PIT
Signed-off-by: Sheng Yang [EMAIL PROTECTED]
---
arch/x86/kvm/i8254.c |7 +++
Avi Kivity [EMAIL PROTECTED] writes:
Paul Collins wrote:
Running kvm 62 on 2.6.24.3 (the extraversion is reported due to a local
patch to vfat) I just got a couple of soft lockups and a hang. I was
installing FreeBSD 7 at the time.
Mar 4 21:20:13 burly kernel: Call Trace:
Mar 4 21:20:13
From da231cf5bb50310b4b98cc4b58370723b8e3f65d Mon Sep 17 00:00:00 2001
From: Sheng Yang [EMAIL PROTECTED]
Date: Sun, 27 Jan 2008 11:25:29 +0800
Subject: [PATCH] kvm: libkvm: Add Supporting for in-kernel PIT model
Signed-off-by: Sheng Yang [EMAIL PROTECTED]
---
kernel/Kbuild |2 +-
With paravirt clocksource, reboot and kexec are broken: the clock keeps
updating after the reboot, and the new kernel will have a random memory
location trampled occasionally.
So we need to stop the clock on kexec (in the guest) and reboot (in the
host). On the host side, this can be done
Yunfeng Zhao wrote:
Hi, all,
This is today's KVM test result against kvm.git
daf4de30ec718b16798aba07e9f25fd9e6ba9e53 and kvm-userspace.git
724f8a940ec0e78e607c051e6e82ca2f5055b1e1.
In today's testing , save/restore crashed host once on pae/ia32e hosts.
One new issue has been found:
Avi Kivity wrote:
Dong, Eddie wrote:
I don't know if the patch was still needed now, since it was posted
long ago(I don't know which issue it solved). I'd like to post a
revert patch if necessary.
I believe the patch is still necessary, since we still need to
guarantee that a vcpu's
Avi Kivity wrote:
With paravirt clocksource, reboot and kexec are broken: the clock keeps
updating after the reboot, and the new kernel will have a random memory
location trampled occasionally.
So we need to stop the clock on kexec (in the guest) and reboot (in the
host). On the host
Hello Izik,
On Tue, Mar 04, 2008 at 02:44:07AM +0200, Izik Eidus wrote:
i wrote to you about this before (i didnt get answer for this so i write
Ouch I must have lost your previous comment with a too-fast pgdown in
the full quoting of the patch sorry.
again)
with large pages support i think
Yang, Sheng wrote:
From 40cf60009d1031891a34685b690dad47290a0b04 Mon Sep 17 00:00:00 2001
From: Sheng Yang [EMAIL PROTECTED]
Date: Tue, 4 Mar 2008 09:29:55 +0800
Subject: [PATCH] kvm: qemu: Add option for enable/disable in kernel PIT
diff --git a/libkvm/libkvm-x86.c b/libkvm/libkvm-x86.c
On Mon, Mar 03, 2008 at 11:31:15PM -0800, Christoph Lameter wrote:
@@ -446,6 +450,8 @@ static int page_mkclean_one(struct page
if (address == -EFAULT)
goto out;
+ /* rmap lock held */
+ emm_notify(mm, emm_invalidate_start, address, address + PAGE_SIZE);
On 3/4/08, Avi Kivity [EMAIL PROTECTED] wrote:
apply to kvm guests. With mmu notifiers, the trend will grow even stronger.
could you (or anybody) elaborate on that? the mmu-related threads show
lots of progress, but it's way (way) out of my league.
AFAICT, it's about the infrastructure to
On Tue, Mar 04, 2008 at 11:35:32AM +0100, Peter Zijlstra wrote:
On Mon, 2008-03-03 at 13:15 -0600, Jack Steiner wrote:
I haven't thought about locking requirements for the radix tree. Most
accesses
would be read-only updates infrequent. Any chance of an RCU-based radix
Javier Guerra wrote:
On 3/4/08, Avi Kivity [EMAIL PROTECTED] wrote:
apply to kvm guests. With mmu notifiers, the trend will grow even stronger.
could you (or anybody) elaborate on that? the mmu-related threads show
lots of progress, but it's way (way) out of my league.
AFAICT,
On Mon, 2008-03-03 at 09:27 -0600, Anthony Liguori wrote:
Hi Dor,
Dor Laor wrote:
void *virtio_net_init(PCIBus *bus, NICInfo *nd, int devfn)
{
VirtIONet *n;
+const char *info_str = virtio-net;
-n = (VirtIONet *)virtio_init_pci(bus, virtio-net, 6900, 0x1000,
+n
repository: /home/dor/src/merge/kvm
branch: trunk
commit acdb6d2aa7332179eead306ead17494ae23edbd6
Author: Dor Laor [EMAIL PROTECTED]
Date: Tue Mar 4 08:16:06 2008 -0500
Check device status in tx timer.
The device can be down when the timer pops.
If it is down it will cause
and ran with your patch, and is rather faster than KVM-37 for
networking...
Hi Michael !
If you're looking for blazing fast networking, then KVM (61+) offers VirtIO -
PV networking, capable of breaking the 1 Gbps barrier...
But to achieve such speeds you'll need to compile custom kernels
repository: /home/dor/src/merge/kvm
branch: trunk
commit 27541e5166e4bb6d053b77f7fd7b3ed75db58138
Author: Dor Laor [EMAIL PROTECTED]
Date: Tue Mar 4 08:59:03 2008 -0500
Add virtio_net_id for every instance, as suggested by Anthony Liguori
Signed-off-by: Dor Laor [EMAIL PROTECTED]
repository: /home/dor/src/merge/kvm
branch: trunk
commit db89e8403773a68d4c0ac3d6813f19ec83cb7011
Author: Dor Laor [EMAIL PROTECTED]
Date: Tue Mar 4 08:14:37 2008 -0500
Fix virtio migration when the device is down
If the guest did not initalize the pfn, don't use it.
repository: /home/dor/src/merge/kvm
branch: trunk
commit bf1e6da21722a4699ea3739e587f282b9a5db418
Author: Dor Laor [EMAIL PROTECTED]
Date: Tue Mar 4 08:57:52 2008 -0500
Use specific avail/used ring structures
This is more accurate than using page_size,
as suggested by Anthony
Alexey Eremenko napsal(a):
and ran with your patch, and is rather faster than KVM-37 for
networking...
Hi Michael !
If you're looking for blazing fast networking, then KVM (61+) offers
VirtIO - PV networking, capable of breaking the 1 Gbps barrier...
But to achieve such speeds you'll
Yang, Sheng wrote:
Hi
Here is the last in-kernel PIT patch for KVM. The mainly change from last
version is the supporting to save/restore. I also tested live migration.
The other modifies including some date structure changed to be better for
supporting the save/restore. I moved the PIT
On Tue, Mar 04, 2008 at 12:47:58PM +0200, Avi Kivity wrote:
With paravirt clocksource, reboot and kexec are broken: the clock keeps
updating after the reboot, and the new kernel will have a random memory
location trampled occasionally.
So we need to stop the clock on kexec (in the guest)
Hello
I'm having weird problem and being a bit puzzled about where to look
for this bug.
I'm using T61 - C2D 2GB
So I'll describe symptoms:
When I run inside my 0.5G smp qemu-kvm guest with Debian these two
loops in parallel:
'while : ; do dmsetup status ; done'
and
'while : ; do cat
Linus, please pull the kvm fixes in the repo and branch
git://git.kernel.org/pub/scm/linux/kernel/git/avi/kvm.git for-linus
comprising an ABI fix, a few host crash fixes, AMD specific fixes,
a Kbuild fix for the randconfig addicts, fallout from the scaling
work, and other miscellany.
Avi
Zdenek Kabelac wrote:
Hello
I'm having weird problem and being a bit puzzled about where to look
for this bug.
I'm using T61 - C2D 2GB
So I'll describe symptoms:
When I run inside my 0.5G smp qemu-kvm guest with Debian these two
loops in parallel:
'while : ; do dmsetup status ;
this patch introduces a -cpu-map option.
It has the form C:x,y,z..., and have the effect
of setting the affinity mask of vcpu C to processors
x,y,z...
Signed-off-by: Glauber Costa [EMAIL PROTECTED]
---
qemu/vl.c | 13 +
1 files changed, 13 insertions(+), 0 deletions(-)
diff --git
store the thread id through our new call as soon as
the process is created.
Signed-off-by: Glauber Costa [EMAIL PROTECTED]
---
qemu/qemu-kvm.c |2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/qemu/qemu-kvm.c b/qemu/qemu-kvm.c
index 8ee3bf8..1aeb97b 100644
---
store default process affinities before we get the chance to process
any options
Signed-off-by: Glauber Costa [EMAIL PROTECTED]
---
qemu/vl.c |3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/qemu/vl.c b/qemu/vl.c
index f10fbd8..6a94724 100644
--- a/qemu/vl.c
+++
This allow us to track which thread is currently running the cpu
Signed-off-by: Glauber Costa [EMAIL PROTECTED]
---
qemu/qemu-kvm.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/qemu/qemu-kvm.c b/qemu/qemu-kvm.c
index 051946e..43e0f2e 100644
--- a/qemu/qemu-kvm.c
+++
Hi guys,
Here's a first series of patch aiming at vcpu pinning support in qemu.
Ideally, as vcpu as just normal threads, the usual userspace tools can be used
to set cpu affinities mask.
However, It makes it very difficult to _start_ a vm with vcpus pinned, since
we don't know the thread ids
Since glibc does not provide a gettid call directly, only through
syscall, we wrap one for kvm
Signed-off-by: Glauber Costa [EMAIL PROTECTED]
---
qemu/qemu-kvm.c |6 ++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/qemu/qemu-kvm.c b/qemu/qemu-kvm.c
index 43e0f2e..8ee3bf8
This patch provides a cpu_set variable to vcpu_info structure.
It stores the current cpu mask for a thread.
We also provide a wrapper for storing a provided affinity, and a
function to set default affinities: The default affinities are current
process'
Signed-off-by: Glauber Costa [EMAIL
Signed-off-by: Glauber Costa [EMAIL PROTECTED]
---
qemu/qemu-kvm.c |8
1 files changed, 8 insertions(+), 0 deletions(-)
diff --git a/qemu/qemu-kvm.c b/qemu/qemu-kvm.c
index a36fbf6..80fe8e5 100644
--- a/qemu/qemu-kvm.c
+++ b/qemu/qemu-kvm.c
@@ -68,6 +68,12 @@ void
2008/3/4, Avi Kivity [EMAIL PROTECTED]:
Zdenek Kabelac wrote:
Hello
I'm having weird problem and being a bit puzzled about where to look
for this bug.
I'm using T61 - C2D 2GB
So I'll describe symptoms:
When I run inside my 0.5G smp qemu-kvm guest with Debian
Zdenek Kabelac wrote:
Is it emulated ? I've thought it's running natively with vmx?
In some cases (memory mapped I/O, writes to page tables) some
instructions are emulated. Usually they run natively.
Please post the output of 'kvm_stat -1' to ensure the problem is with
the emulator.
--
Avi Kivity [EMAIL PROTECTED] writes:
Most likely movs emulation is broken for long counts. Please post a
disassembly of copy_user_generic_string to make sure we're looking at
the same code.
Be careful -- this code is patched at runtime and what you
see in the vmlinux is not necessarily
In certain circumstances, the calculated CHS can result in a total number of
sectors that is less than the actual number of sectors. I'm not entirely
sure why this upsets grub, but it seems to be the source of the Grub Error 18
that sometimes occurs when using extboot.
The solution is to
Glauber Costa wrote:
Hi guys,
Here's a first series of patch aiming at vcpu pinning support in qemu.
Ideally, as vcpu as just normal threads, the usual userspace tools can be used
to set cpu affinities mask.
However, It makes it very difficult to _start_ a vm with vcpus pinned, since
we
david ahern daahern at cisco.com writes:
I know this issue has been discussed on this list before, but I am still
experiencing network freezes in a guest that requires a restart to clear. When
the network freezes in the guest I no longer see the network interrupts
counter
incrementing
Anthony Liguori wrote:
Glauber Costa wrote:
Hi guys,
Here's a first series of patch aiming at vcpu pinning support in qemu.
Ideally, as vcpu as just normal threads, the usual userspace tools can
be used
to set cpu affinities mask.
However, It makes it very difficult to _start_ a vm with
Joerg Roedel wrote:
On Tue, Mar 04, 2008 at 01:21:53PM -0300, Glauber Costa wrote:
Hi guys,
Here's a first series of patch aiming at vcpu pinning support in qemu.
Ideally, as vcpu as just normal threads, the usual userspace tools can be
used
to set cpu affinities mask.
However, It makes
On Tue, Mar 04, 2008 at 03:11:27PM -0300, Glauber Costa wrote:
Joerg Roedel wrote:
On Tue, Mar 04, 2008 at 01:21:53PM -0300, Glauber Costa wrote:
Hi guys,
Here's a first series of patch aiming at vcpu pinning support in qemu.
Ideally, as vcpu as just normal threads, the usual userspace
Change the PCI network drivers init functions to return the PCIDev, to
inform which slot has been hot-plugged.
Also record devfn on the NICInfo structure to locate for release
on hot-removal.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/hw/e1000.c
On Tue, 4 Mar 2008, Nick Piggin wrote:
Then put it into the arch code for TLB invalidation. Paravirt ops gives
good examples on how to do that.
Put what into arch code?
The mmu notifier code.
What about a completely different approach... XPmem runs over NUMAlink,
right? Why not
On Tue, 4 Mar 2008, Andrea Arcangeli wrote:
When working with single pages it's more efficient and preferable to
call invalidate_page and only later release the VM reference on the
page.
But as you pointed out before that path is a slow path anyways. Its rarely
taken. Having a single
Return PCIDevice from bus number and slot.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/hw/pci.c
===
--- kvm-userspace.hotplug.orig/qemu/hw/pci.c
+++ kvm-userspace.hotplug/qemu/hw/pci.c
@@
Use GPE _L01 to notify OSPM.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/bios/acpi-dsdt.dsl
===
--- kvm-userspace.hotplug.orig/bios/acpi-dsdt.dsl
+++ kvm-userspace.hotplug/bios/acpi-dsdt.dsl
@@
Handle the _EJ0 notifications.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/hw/acpi.c
===
--- kvm-userspace.hotplug.orig/qemu/hw/acpi.c
+++ kvm-userspace.hotplug/qemu/hw/acpi.c
@@ -26,6 +26,7 @@
Add monitor command to hot-remove devices.
Remove device data on _EJ0 notification.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/monitor.c
===
--- kvm-userspace.hotplug.orig/qemu/monitor.c
+++
Dynamically allocate drive options and drive table index, so to
reused indexes when devices are removed.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/sysemu.h
===
---
Presence of _EJ0 method indicates that slots are hot-pluggable.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/bios/acpi-dsdt.dsl
===
--- kvm-userspace.hotplug.orig/bios/acpi-dsdt.dsl
+++
Add monitor command to hot-add PCI devices (nic and drive).
Save QEMUMachine necessary for drive_init.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/Makefile.target
===
---
Device hotplug will use that structure from a separate
file.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/sysemu.h
===
--- kvm-userspace.hotplug.orig/qemu/sysemu.h
+++
To be used by hot-remove.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/vl.c
===
--- kvm-userspace.hotplug.orig/qemu/vl.c
+++ kvm-userspace.hotplug/qemu/vl.c
@@ -4972,6 +4972,12 @@ static int
Record devfn on the BlockDriverState structure to locate for release
on hot-removal.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/block_int.h
===
--- kvm-userspace.hotplug.orig/qemu/block_int.h
Enable the corresponding bit on the PCIST region and trigger the SCI.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/hw/acpi.c
===
--- kvm-userspace.hotplug.orig/qemu/hw/acpi.c
+++
Return PCIBus pointer from bus number integer.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/hw/pci.c
===
--- kvm-userspace.hotplug.orig/qemu/hw/pci.c
+++ kvm-userspace.hotplug/qemu/hw/pci.c
@@
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/net.h
===
--- kvm-userspace.hotplug.orig/qemu/net.h
+++ kvm-userspace.hotplug/qemu/net.h
@@ -38,6 +38,7 @@ void do_info_network(void);
int
If the io port is unassigned, the previous private pointer is
meaningless.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/vl.c
===
--- kvm-userspace.hotplug.orig/qemu/vl.c
+++
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/hw/lsi53c895a.c
===
--- kvm-userspace.hotplug.orig/qemu/hw/lsi53c895a.c
+++ kvm-userspace.hotplug/qemu/hw/lsi53c895a.c
@@ -1849,6 +1849,18 @@ void
So drivers can clear their mem io table entries on exit back to unassigned
state.
Also make the io mem index allocation dynamic.
Perhaps freeing the state created during cpu_register_physical_memory()
is also necessary.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index:
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/bios/acpi-dsdt.dsl
===
--- kvm-userspace.hotplug.orig/bios/acpi-dsdt.dsl
+++ kvm-userspace.hotplug/bios/acpi-dsdt.dsl
@@ -249,6 +249,162 @@ DefinitionBlock
The _EJ0 method is executed by the OS once it has successfully finished
device removal. Inform that event through IO port space so QEMU
can free the associated data.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/bios/acpi-dsdt.dsl
Unregister the pci device, unassign its IO and memory regions, and free
associated data.
Add a callback so drivers can free device state.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/hw/pci.c
The following patchset adds ACPI PCI hotplug support for QEMU.
It extends the number of slots with IRQ routing information from 6 to 32.
The only PCI driver which the unregister method has been added is LSI SCSI,
would
like more comments to implement that for the remaining drivers.
--
Export net/drive add/remove functions for device hotplug usage.
Return the table index on add.
Return failure instead of exiting if limit has been reached
on drive_add.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/vl.c
The same, but for nics.
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/net.h
===
--- kvm-userspace.hotplug.orig/qemu/net.h
+++ kvm-userspace.hotplug/qemu/net.h
@@ -46,6 +46,7 @@ struct NICInfo {
Signed-off-by: Marcelo Tosatti [EMAIL PROTECTED]
Index: kvm-userspace.hotplug/qemu/hw/irq.c
===
--- kvm-userspace.hotplug.orig/qemu/hw/irq.c
+++ kvm-userspace.hotplug/qemu/hw/irq.c
@@ -56,6 +56,12 @@ qemu_irq
Glauber Costa wrote:
My main interest is in management tools being able to specify pinning
set ups at VM creation time.
As I said, it can be done through tools like taskset, but then you'd
have to know:
* when are the threads created
* which thread ids corresponds to each cpu
And of
Daniel P. Berrange wrote:
On Tue, Mar 04, 2008 at 03:34:33PM -0300, Marcelo Tosatti wrote:
Add monitor command to hot-add PCI devices (nic and drive).
[snip]
Index: kvm-userspace.hotplug/qemu/monitor.c
===
---
Daniel P. Berrange wrote:
Removing based on pci device number is very un-pleasant, since its not
something
the user of the monitor cares about. Nor do they even know what the PCI
device number
assigned by 'pci_add' is.
As with addition, I'd like separate commands for NIC vs Drive, and for
On Tue, Mar 04, 2008 at 01:28:24PM -0600, Anthony Liguori wrote:
Glauber Costa wrote:
My main interest is in management tools being able to specify pinning
set ups at VM creation time.
As I said, it can be done through tools like taskset, but then you'd
have to know:
* when are the
On Tue, Mar 04, 2008 at 01:30:42PM -0600, Anthony Liguori wrote:
Daniel P. Berrange wrote:
On Tue, Mar 04, 2008 at 03:34:33PM -0300, Marcelo Tosatti wrote:
Add monitor command to hot-add PCI devices (nic and drive).
[snip]
Index: kvm-userspace.hotplug/qemu/monitor.c
On Tue, Mar 04, 2008 at 01:32:22PM -0600, Anthony Liguori wrote:
Daniel P. Berrange wrote:
Removing based on pci device number is very un-pleasant, since its not
something
the user of the monitor cares about. Nor do they even know what the PCI
device number
assigned by 'pci_add' is.
As
Daniel P. Berrange wrote:
On Tue, Mar 04, 2008 at 01:28:24PM -0600, Anthony Liguori wrote:
Glauber Costa wrote:
My main interest is in management tools being able to specify pinning
set ups at VM creation time.
As I said, it can be done through tools like taskset, but then you'd
have to
Anthony Liguori wrote:
Glauber Costa wrote:
My main interest is in management tools being able to specify pinning
set ups at VM creation time.
As I said, it can be done through tools like taskset, but then you'd
have to know:
* when are the threads created
* which thread ids corresponds
On Tue, Mar 04, 2008 at 05:00:09PM -0300, Marcelo Tosatti wrote:
On Tue, Mar 04, 2008 at 01:32:22PM -0600, Anthony Liguori wrote:
Daniel P. Berrange wrote:
nic_remove [vlan=n][,macaddr=addr][,model=type]
drive_remove [[file=file][,if=type][,bus=n][,unit=m][,media=d][index=i]]
Actually, the following (old) patch by Yuval Kashtan was sent to
qemu-devel back in October (but not merged AFAIK) to facilitate usb_del
using the same string used during usb_add.
Hello,
This proposed patch add support for deleting usb devices by providing
the (exact) same string they were
On Tue, Mar 04, 2008 at 11:00:31AM -0800, Christoph Lameter wrote:
But as you pointed out before that path is a slow path anyways. Its rarely
It's a slow path but I don't see why you think two hooks are better
than one, when only one is necessary.
I once ripped invalidate_page while working on
On Tue, 4 Mar 2008, Andrea Arcangeli wrote:
I once ripped invalidate_page while working on #v8 but then I
reintroduced it because I thought reducing the total number of hooks
was beneficial to the core linux VM (even if only a
microoptimization, I sure agree about that, but it's trivial to
On Tue, 2008-03-04 at 09:42 -0600, Anthony Liguori wrote:
diff --git a/qemu/hw/virtio-net.c b/qemu/hw/virtio-net.c
index 612cf6b..a05eafb 100644
--- a/qemu/hw/virtio-net.c
+++ b/qemu/hw/virtio-net.c
@@ -280,6 +280,10 @@ static void virtio_net_tx_timer(void *opaque)
{
VirtIONet
Hello
Sorry for my intervention into your communication, but are you sure,
when you want to use virtio, you must have 2.6.25 kernel on both - guest
and host system?
I test this case right now, and I have running 2.6.25-rc3 only on guest
system with kvm-62.
It's enough running the
On Tue, 4 Mar 2008, Peter Zijlstra wrote:
On Tue, 2008-03-04 at 14:35 -0800, Christoph Lameter wrote:
RCU means that the callbacks occur in an atomic context.
Not really, if it requires moving the VM locks to sleepable locks under
a .config option, I think its also fair to require
On Tue, 2008-03-04 at 09:52 -0600, Anthony Liguori wrote:
Yang, Sheng wrote:
Hi
Here is the last in-kernel PIT patch for KVM. The mainly change from last
version is the supporting to save/restore. I also tested live migration.
The other modifies including some date structure changed
FWIW, I'll cut the kvm and openfabrics lists from any future posts.
I'm getting tired of the bounces.
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
On Mon, Mar 03, 2008 at 11:01:22AM -0800, Christoph Lameter wrote:
On Mon, 3 Mar 2008, Nick Piggin wrote:
I'm still not completely happy with this. I had a very quick look
at the GRU driver, but I don't see why it can't be implemented
more like the regular TLB model, and have TLB
Dor Laor wrote:
On Tue, 2008-03-04 at 09:52 -0600, Anthony Liguori wrote:
Yang, Sheng wrote:
Hi
Here is the last in-kernel PIT patch for KVM. The mainly change from last
version is the supporting to save/restore. I also tested live migration.
The other modifies including some
On Wednesday 05 March 2008 05:58, Christoph Lameter wrote:
On Tue, 4 Mar 2008, Nick Piggin wrote:
Then put it into the arch code for TLB invalidation. Paravirt ops gives
good examples on how to do that.
Put what into arch code?
The mmu notifier code.
It isn't arch specific.
1 - 100 of 126 matches
Mail list logo