On Mon, 27 Sep 2010 18:44:04 -0400
Lucas Meneghel Rodrigues l...@redhat.com wrote:
+
+vm = kvm_test_utils.get_living_vm(env, params.get(main_vm))
+session = kvm_test_utils.wait_for_login(vm,
+ timeout=int(params.get(login_timeout, 360)))
+# Let's just error the
On Tue, 5 Oct 2010 11:15:21 pm Dave Young wrote:
Balloon could cause guest memory oom killing and panic.
Add oom notify to leak some memory and retry fill balloon after 5 minutes.
Have you tried registering a shrinker? See mm.h.
Thanks,
Rusty.
--
To unsubscribe from this list: send the line
On Tue, Sep 21, 2010 at 06:05:10PM +0200, Avi Kivity wrote:
On 09/20/2010 07:02 PM, Michael S. Tsirkin wrote:
On Mon, Sep 20, 2010 at 05:06:41PM +0200, Avi Kivity wrote:
This cleans up msix/kvm integration a bit. The really important patch is
the
last one, which allows msix.o to be
On 10/06/2010 11:39 AM, Michael S. Tsirkin wrote:
On Tue, Sep 21, 2010 at 06:05:10PM +0200, Avi Kivity wrote:
On 09/20/2010 07:02 PM, Michael S. Tsirkin wrote:
On Mon, Sep 20, 2010 at 05:06:41PM +0200, Avi Kivity wrote:
This cleans up msix/kvm integration a bit. The really important
On Wed, 6 Oct 2010 14:26:46 +0530
pradeep psuri...@linux.vnet.ibm.com wrote:
On Mon, 27 Sep 2010 18:44:04 -0400
Lucas Meneghel Rodrigues l...@redhat.com wrote:
ion,
vm.get_mac_address(0))
+supported_features = params.get(supported_features).split()
I guess split this expects
Hi again everybody,
One of the admins at the ProxmoxVE project was gracious enough to
quickly release a package including the previously discussed change to
allow up to 32 NICs in qemu.
For future reference the .deb is here:
On Tue, Oct 05, 2010 at 04:51:50PM -0300, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:32PM +0200, Gleb Natapov wrote:
If async page fault is received by idle task or when preemp_count is
not zero guest cannot reschedule, so do sti; hlt and wait for page to be
ready. vcpu can still
On Tue, Oct 05, 2010 at 04:00:51PM -0300, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:31PM +0200, Gleb Natapov wrote:
Send async page fault to a PV guest if it accesses swapped out memory.
Guest will choose another task to run upon receiving the fault.
Allow async page fault
This case can pass with rhel5.5 rhel6.0, not test with fedora.
it would not be the problem of testcase.
I did not touch this problem, can you provide more debug info ? eg,
tcpdump, ...
It seems like RHEL 5.5 issue
it fails only with TCP_CRR
--
To unsubscribe from this list: send the
On 10/05/2010 04:59 PM, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:24PM +0200, Gleb Natapov wrote:
If a guest accesses swapped out memory do not swap it in from vcpu thread
context. Schedule work to do swapping and put vcpu into halted state
instead.
Interrupts will still be
On Wed, Oct 06, 2010 at 12:50:01PM +0200, Avi Kivity wrote:
On 10/05/2010 04:59 PM, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:24PM +0200, Gleb Natapov wrote:
If a guest accesses swapped out memory do not swap it in from vcpu thread
context. Schedule work to do swapping and put
On Tue, Oct 05, 2010 at 03:25:54PM -0300, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:29PM +0200, Gleb Natapov wrote:
Enable async PF in a guest if async PF capability is discovered.
Signed-off-by: Gleb Natapov g...@redhat.com
---
Documentation/kernel-parameters.txt |3 +
On Tue, Oct 05, 2010 at 12:54:09PM -0300, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:25PM +0200, Gleb Natapov wrote:
When page is swapped in it is mapped into guest memory only after guest
tries to access it again and generate another fault. To save this fault
we can map it
On Tue, Oct 05, 2010 at 01:57:38PM -0300, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:26PM +0200, Gleb Natapov wrote:
Keep track of memslots changes by keeping generation number in memslots
structure. Provide kvm_write_guest_cached() function that skips
gfn_to_hva() translation if
On Tue, Oct 05, 2010 at 11:59:16AM -0300, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:24PM +0200, Gleb Natapov wrote:
If a guest accesses swapped out memory do not swap it in from vcpu thread
context. Schedule work to do swapping and put vcpu into halted state
instead.
On Tuesday 05 October 2010, Krishna Kumar2 wrote:
After testing various combinations of #txqs, #vhosts, #netperf
sessions, I think the drop for 1 stream is due to TX and RX for
a flow being processed on different cpus. I did two more tests:
1. Pin vhosts to same CPU:
- BW drop is
Fix typo in copyright notice.
Signed-off-by: Nicolas Kaiser ni...@nikai.net
---
arch/x86/kvm/emulate.c |2 +-
arch/x86/kvm/i8254.c |2 +-
arch/x86/kvm/i8259.c |2 +-
arch/x86/kvm/irq.c |2 +-
arch/x86/kvm/lapic.c |2 +-
arch/x86/kvm/mmu.c
On Fri, Sep 17, 2010 at 03:33:07PM +0530, Krishna Kumar wrote:
For 1 TCP netperf, I ran 7 iterations and summed it. Explanation
for degradation for 1 stream case:
I thought about possible RX/TX contention reasons, and I realized that
we get/put the mm counter all the time. So I write the
makes it possible to batch use/unuse mm
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
drivers/vhost/vhost.c |9 -
1 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 677d112..8b9d474 100644
---
Move use/unuse mm to vhost.c which makes it possible to batch these
operations.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
drivers/vhost/net.c |7 ---
drivers/vhost/vhost.c |7 ++-
2 files changed, 6 insertions(+), 8 deletions(-)
diff --git a/drivers/vhost/net.c
On Wed, Oct 6, 2010 at 5:05 PM, Rusty Russell ru...@rustcorp.com.au wrote:
On Tue, 5 Oct 2010 11:15:21 pm Dave Young wrote:
Balloon could cause guest memory oom killing and panic.
Add oom notify to leak some memory and retry fill balloon after 5 minutes.
Have you tried registering a
The number of mce_banks needs to be multiplied by 4 in order to actually
reference all of the entries.
Signed-off-by: Dean Nelson dnel...@redhat.com
---
qemu-kvm-x86.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/qemu-kvm-x86.c b/qemu-kvm-x86.c
index
On Wed, Oct 06, 2010 at 01:07:04PM +0200, Gleb Natapov wrote:
Can't you set a bit in vcpu-requests instead, and handle it in out:
at the end of vcpu_enter_guest?
To have a single entry point for pagefaults, after vmexit handling.
Jumping to out: will skip vmexit handling anyway, so we
On Wed, Oct 06, 2010 at 12:55:04PM +0200, Gleb Natapov wrote:
On Tue, Oct 05, 2010 at 03:25:54PM -0300, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:29PM +0200, Gleb Natapov wrote:
Enable async PF in a guest if async PF capability is discovered.
Signed-off-by: Gleb Natapov
On Wed, Oct 06, 2010 at 01:14:17PM +0200, Gleb Natapov wrote:
+int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache
*ghc,
+ gpa_t gpa)
+{
+ struct kvm_memslots *slots = kvm_memslots(kvm);
+ int offset = offset_in_page(gpa);
+ gfn_t gfn
vhost irqfd support: case where many vqs are
mapped to a single msix vector is currently broken.
Fix it up.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
This is on top of the qemu patchset, which is unchanged.
Fixes from v1:
correct error handling
hw/msix.c | 68
* pradeep psuri...@linux.vnet.ibm.com [2010-10-06 03:57]:
On Mon, 27 Sep 2010 18:44:04 -0400
Lucas Meneghel Rodrigues l...@redhat.com wrote:
+
+vm = kvm_test_utils.get_living_vm(env, params.get(main_vm))
+session = kvm_test_utils.wait_for_login(vm,
+
On Wed, Oct 06, 2010 at 10:10:51AM +0900, Hidetoshi Seto wrote:
(snip)
Index: qemu/kvm.h
===
--- qemu.orig/kvm.h
+++ qemu/kvm.h
@@ -110,6 +110,9 @@ int kvm_arch_init_vcpu(CPUState *env);
void
On Wed, Oct 06, 2010 at 10:58:36AM +0900, Hidetoshi Seto wrote:
I got some more question:
(2010/10/05 3:54), Marcelo Tosatti wrote:
Index: qemu/target-i386/cpu.h
===
--- qemu.orig/target-i386/cpu.h
+++
On Wed, 2010-10-06 at 16:56 +0200, Michael S. Tsirkin wrote:
vhost irqfd support: case where many vqs are
mapped to a single msix vector is currently broken.
Fix it up.
Signed-off-by: Michael S. Tsirkin m...@redhat.com
---
This is on top of the qemu patchset, which is unchanged.
Fixes
On Wed, 2010-10-06 at 10:55 -0500, Ryan Harper wrote:
* pradeep psuri...@linux.vnet.ibm.com [2010-10-06 03:57]:
On Mon, 27 Sep 2010 18:44:04 -0400
Lucas Meneghel Rodrigues l...@redhat.com wrote:
+
+vm = kvm_test_utils.get_living_vm(env, params.get(main_vm))
+session =
Michael S. Tsirkin m...@redhat.com wrote on 10/06/2010 07:04:31 PM:
Michael S. Tsirkin m...@redhat.com
10/06/2010 07:04 PM
To
Krishna Kumar2/India/i...@ibmin
cc
ru...@rustcorp.com.au, da...@davemloft.net, kvm@vger.kernel.org,
a...@arndb.de, net...@vger.kernel.org, a...@redhat.com,
On Wed, Oct 06, 2010 at 10:48:44AM -0600, Alex Williamson wrote:
-int msix_unset_mask_notifier(PCIDevice *dev, unsigned vector)
+static int msix_unset_mask_notifier_for_vector(PCIDevice *dev, unsigned
vector)
{
int r = 0;
-void *opaque;
if (vector =
It's 8 otherwise- and after the patch is applied, it still only goes to
28 for some reason.
28's acceptable for my needs, so I'll step aside from here leave it to
the experts.
As for the new -device method, that's all fine good but AFAIK it's not
implemented on my platform, so this was the
Arnd Bergmann a...@arndb.de wrote on 10/06/2010 05:49:00 PM:
I don't see any reasons mentioned above. However, for higher
number of netperf sessions, I see a big increase in retransmissions:
___
#netperf ORG NEW
BW (#retr)
* Chris Wright (chr...@sous-sol.org) wrote:
* Pete Ashdown (pashd...@xmission.com) wrote:
ProxMox guest:
/usr/bin/kvm -monitor unix:/var/run/qemu-server/104.mon,server,nowait -vnc
unix:/var/run/qemu-server/104.vnc,password -pidfile
/var/run/qemu-server/104.pid -daemonize -usbdevice
* linux_...@proinbox.com (linux_...@proinbox.com) wrote:
Hi again everybody,
One of the admins at the ProxmoxVE project was gracious enough to
quickly release a package including the previously discussed change to
allow up to 32 NICs in qemu.
You mean they patched qemu to increase the
On Wed, 2010-10-06 at 19:02 +0200, Michael S. Tsirkin wrote:
On Wed, Oct 06, 2010 at 10:48:44AM -0600, Alex Williamson wrote:
-int msix_unset_mask_notifier(PCIDevice *dev, unsigned vector)
+static int msix_unset_mask_notifier_for_vector(PCIDevice *dev, unsigned
vector)
{
int
On Wed, Oct 06, 2010 at 11:24:24AM -0600, Alex Williamson wrote:
On Wed, 2010-10-06 at 19:02 +0200, Michael S. Tsirkin wrote:
On Wed, Oct 06, 2010 at 10:48:44AM -0600, Alex Williamson wrote:
-int msix_unset_mask_notifier(PCIDevice *dev, unsigned vector)
+static int
commit ce6325ff1af34dbaee91c8d28e792277e43f1227
Author: Glauber Costa gco...@redhat.com
Date: Wed Mar 5 17:01:10 2008 -0300
Augment info cpus
This patch exposes the thread id associated with each
cpu through the already well known 'info cpus' interface.
Signed-off-by: Marcelo
From: Huang Ying ying.hu...@intel.com
In QEMU-KVM, physical address != RAM address. While MCE simulation
needs physical address instead of RAM address. So
kvm_physical_memory_addr_from_ram() is implemented to do the
conversion, and it is invoked before being filled in the IA32_MCi_ADDR
MSR.
Port qemu-kvm's
commit 1bab5d11545d8de5facf46c28630085a2f9651ae
Author: Huang Ying ying.hu...@intel.com
Date: Wed Mar 3 16:52:46 2010 +0800
Add savevm/loadvm support for MCE
MCE registers are saved/load into/from CPUState in
kvm_arch_save/load_regs. To simulate the MCG_STATUS
To be used by next patches.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
Index: qemu/cpu-common.h
===
--- qemu.orig/cpu-common.h
+++ qemu/cpu-common.h
@@ -47,7 +47,8 @@ void qemu_ram_free(ram_addr_t addr);
/* This should only
On Wed, Oct 06, 2010 at 10:08:19AM -0400, Dean Nelson wrote:
The number of mce_banks needs to be multiplied by 4 in order to actually
reference all of the entries.
Signed-off-by: Dean Nelson dnel...@redhat.com
Applied, thanks.
--
To unsubscribe from this list: send the line unsubscribe kvm
Block SIGALRM, SIGIO and consume them via signalfd.
Signed-off-by: Marcelo Tosatti mtosa...@redhat.com
Index: qemu/cpus.c
===
--- qemu.orig/cpus.c
+++ qemu/cpus.c
@@ -33,6 +33,7 @@
#include exec-all.h
#include cpus.h
+#include
Port qemu-kvm's signalfd compat code.
commit 5a7fdd0abd7cd24dac205317a4195446ab8748b5
Author: Anthony Liguori aligu...@us.ibm.com
Date: Wed May 7 11:55:47 2008 -0500
Use signalfd() in io-thread
This patch reworks the IO thread to use signalfd() instead of sigtimedwait()
This
Port qemu-kvm's KVM MCE (Machine Check Exception) handling to qemu. It
allows qemu to propagate MCEs to the guest.
v2:
- rename do_qemu_ram_addr_from_host.
- fix kvm_on_sigbus/kvm_on_sigbus_vcpu naming.
- fix bank register restoration (Dean Nelson).
--
To unsubscribe from this list: send the
Port qemu-kvm's MCE support
commit c68b2374c9048812f488e00ffb95db66c0bc07a7
Author: Huang Ying ying.hu...@intel.com
Date: Mon Jul 20 10:00:53 2009 +0800
Add MCE simulation support to qemu/kvm
KVM ioctls are used to initialize MCE simulation and inject MCE. The
real MCE
Port qemu-kvm's
commit 4b62fff1101a7ad77553147717a8bd3bf79df7ef
Author: Huang Ying ying.hu...@intel.com
Date: Mon Sep 21 10:43:25 2009 +0800
MCE: Relay UCR MCE to guest
UCR (uncorrected recovery) MCE is supported in recent Intel CPUs,
where some hardware error such as some
Michael S. Tsirkin m...@redhat.com wrote on 10/05/2010 11:53:23 PM:
Any idea where does this come from?
Do you see more TX interrupts? RX interrupts? Exits?
Do interrupts bounce more between guest CPUs?
4. Identify reasons for single netperf BW regression.
After testing various
On Wednesday 06 October 2010 19:14:42 Krishna Kumar2 wrote:
Arnd Bergmann a...@arndb.de wrote on 10/06/2010 05:49:00 PM:
I don't see any reasons mentioned above. However, for higher
number of netperf sessions, I see a big increase in retransmissions:
On 10/06/2010 11:05 AM, Marcelo Tosatti wrote:
On Wed, Oct 06, 2010 at 10:58:36AM +0900, Hidetoshi Seto wrote:
I got some more question:
(2010/10/05 3:54), Marcelo Tosatti wrote:
Index: qemu/target-i386/cpu.h
===
---
On Tue, Sep 28, 2010 at 12:05:20PM +0200, Roedel, Joerg wrote:
On Tue, Sep 28, 2010 at 05:37:58AM -0400, Avi Kivity wrote:
On 09/28/2010 11:28 AM, Roedel, Joerg wrote:
Weird, it worked here as I tested it. I had it on qemu/master and with
all three patches. But patch 1 should not make
On Mon, Sep 27, 2010 at 03:16:15PM +0200, Joerg Roedel wrote:
As requested by Alex this patch makes kvm64 the default CPU
model when qemu is started with -enable-kvm. This takes only
effect for qemu-versions newer or equal to 0.14.0.
Signed-off-by: Joerg Roedel joerg.roe...@amd.com
---
On Wed, Oct 06, 2010 at 11:13:31PM +0530, Krishna Kumar2 wrote:
Michael S. Tsirkin m...@redhat.com wrote on 10/05/2010 11:53:23 PM:
Any idea where does this come from?
Do you see more TX interrupts? RX interrupts? Exits?
Do interrupts bounce more between guest CPUs?
4. Identify
On 10/06/2010 01:53 PM, Marcelo Tosatti wrote:
On Mon, Sep 27, 2010 at 03:16:15PM +0200, Joerg Roedel wrote:
As requested by Alex this patch makes kvm64 the default CPU
model when qemu is started with -enable-kvm. This takes only
effect for qemu-versions newer or equal to 0.14.0.
On 10/06/2010 12:34 PM, Marcelo Tosatti wrote:
Port qemu-kvm's MCE support
commit c68b2374c9048812f488e00ffb95db66c0bc07a7
Author: Huang Yingying.hu...@intel.com
Date: Mon Jul 20 10:00:53 2009 +0800
Add MCE simulation support to qemu/kvm
KVM ioctls are used to initialize MCE
On Wed, Oct 06, 2010 at 11:45:12AM -0300, Marcelo Tosatti wrote:
On Wed, Oct 06, 2010 at 12:55:04PM +0200, Gleb Natapov wrote:
On Tue, Oct 05, 2010 at 03:25:54PM -0300, Marcelo Tosatti wrote:
On Mon, Oct 04, 2010 at 05:56:29PM +0200, Gleb Natapov wrote:
Enable async PF in a guest if
On Wed, Oct 06, 2010 at 11:38:47AM -0300, Marcelo Tosatti wrote:
On Wed, Oct 06, 2010 at 01:14:17PM +0200, Gleb Natapov wrote:
+int kvm_gfn_to_hva_cache_init(struct kvm *kvm, struct gfn_to_hva_cache
*ghc,
+ gpa_t gpa)
+{
+ struct kvm_memslots
From: Michael S. Tsirkin m...@redhat.com
Date: Tue, 5 Oct 2010 20:27:32 +0200
It looks like it was a quiet cycle for vhost-net:
probably because most of energy was spent on bugfixes
that went in for 2.6.36.
People are working on multiqueue, tracing but I'm not
sure it'll get done in time for
On Mon, Oct 04, 2010 at 03:26:18PM -0600, Alex Williamson wrote:
This cleans up device assignment option ROM support and allows
us to use romfile and rombar default PCI options. Thanks,
Alex
---
Alex Williamson (2):
device-assignment: Allow PCI to manage the option ROM
Our code paths for saving or migrating a VM are full of functions that
return void, leaving no opportunity for a device to cancel a migration,
either from error or incompatibility. The ivshmem driver attempted to
solve this with a no_migrate flag on the save state entry. I think the
more generic
Some devices may not always able to save their state, allow
the save handler to return an error.
Signed-off-by: Alex Williamson alex.william...@redhat.com
---
hw/adb.c|8 ++--
hw/ads7846.c|4 +++-
hw/arm_gic.c|4 +++-
This allows vmsd based saves to also have a way to signal that
they can't be saved or migrated.
Signed-off-by: Alex Williamson alex.william...@redhat.com
---
hw/fdc.c |3 ++-
hw/hpet.c |3 ++-
hw/hw.h |6 +++---
hw/i2c.c |3 ++-
Carry the vmsd pre_save error reporting through pci_device_save().
Signed-off-by: Alex Williamson alex.william...@redhat.com
---
hw/grackle_pci.c |4 +---
hw/gt64xxx.c |3 +--
hw/ivshmem.c |7 ++-
hw/openpic.c |4 +---
hw/pci.c |9 +++--
hw/pci.h
Carry pci_device_save() error through to virtio_save().
Signed-off-by: Alex Williamson alex.william...@redhat.com
---
hw/virtio-balloon.c|6 +-
hw/virtio-blk.c|6 +-
hw/virtio-net.c|7 ++-
hw/virtio-pci.c| 10 --
This lets a save state handler NAK a migration or cancel if
it runs into problems.
Signed-off-by: Alex Williamson alex.william...@redhat.com
---
block-migration.c |4 +++-
hw/hw.h |2 +-
savevm.c | 18 +++---
3 files changed, 19 insertions(+), 5
Now that the save state handlers can return error, individual drivers
can cancel a migration if they hit an error or don't support it. This
makes the unmigratable callback redundant. Remove it and change the
only user to cancel the migration in a set_params callback, which
actually happens much
On Wed, Oct 06, 2010 at 11:24:24AM -0600, Alex Williamson wrote:
You could always keep the functions as separate wrapper callers of the
common function so you only need to keep true = unset, false = set
straight in one place. Thanks,
Just to show why it does not work, I did exactly this: as
On Wed, 2010-10-06 at 23:44 +0200, Michael S. Tsirkin wrote:
On Wed, Oct 06, 2010 at 11:24:24AM -0600, Alex Williamson wrote:
You could always keep the functions as separate wrapper callers of the
common function so you only need to keep true = unset, false = set
straight in one place.
From: Amos Kong ak...@redhat.com
Framework could not totalise default extra_params and extra_params_vm1 in the
following condition, it's difficult to realise when parsing config file or
calling get_sub_dict*().
extra_params += ' str1'
- case:
extra_params_vm1 += str2
Signed-off-by: Amos Kong
Repeatedly load/unload nic driver, try to transfer file between guest and host
by threads at the same time, and check the md5sum.
Changes from v4:
- Give some time for the interface to be present after
modprobe is executed.
Changes from v1:
- Use a new method to get nic driver name
- Use
From: Amos Kong ak...@redhat.com
This is an enhancement of existed vlan test. Rename the vlan_tag.py to vlan.py,
it is more reasonable.
. Setup arp from /proc/sys/net/ipv4/conf/all/arp_ignore
. Multiple vlans exist simultaneously
. Test ping between same and different vlans
. Test by TCP data
* Anthony Liguori (anth...@codemonkey.ws) wrote:
BTW, using -device, it should be possible to add a very high number of
nics because you can specify the PCI address including a function. If
this doesn't Just Work today, we should make it work.
Should work...test...mostly[1], but I don't
On Mon, 2010-09-27 at 18:43 -0400, Lucas Meneghel Rodrigues wrote:
We are close to the end of this journey. Several little problems
were fixed and we are down to some little problems:
Ok, all patches applied. Thanks to everyone that helped on this effort!
1 - jumbo test - tap interface name
(2010/10/07 3:10), Dean Nelson wrote:
On 10/06/2010 11:05 AM, Marcelo Tosatti wrote:
On Wed, Oct 06, 2010 at 10:58:36AM +0900, Hidetoshi Seto wrote:
I got some more question:
(2010/10/05 3:54), Marcelo Tosatti wrote:
Index: qemu/target-i386/cpu.h
76 matches
Mail list logo