Re: Equivalent of vmware SIOC (Storage IO Control) in KVM

2013-10-13 Thread Fam Zheng
On Mon, 10/14 02:18, Wangshen (Peter) wrote: > > > on Sunday, October 13, 2013 9:27 PM, Paolo Bonzini wrote: > > Il 12/10/2013 08:09, Soumendu Satapathy (sosatapa) ha scritto: > > > Do we have an equivalent of vmware SIOC like feature in KVM? > > > > Yes, you have two choices: > > > > 1) use cg

RE: Equivalent of vmware SIOC (Storage IO Control) in KVM

2013-10-13 Thread Wangshen (Peter)
on Sunday, October 13, 2013 9:27 PM, Paolo Bonzini wrote: > Il 12/10/2013 08:09, Soumendu Satapathy (sosatapa) ha scritto: > > Do we have an equivalent of vmware SIOC like feature in KVM? > > Yes, you have two choices: > > 1) use cgroups to throttle I/O at the level of the host disk (i.e. > mul

[RFC PATCH v2 0/1] Workqueue based vhost work scheduling

2013-10-13 Thread Bandan Das
This is a followup to RFC posted by Shirley Ma on 22 March 2012 : NUMA aware scheduling per vhost thread patch [1]. This patch is against 3.12-rc4. This is a step-down from the previous version in the sense that this patch utilizes the workqueue mechanism instead of creating per-cpu vhost threads,

[RFC PATCH v2 1/1] Workqueue based vhost workers

2013-10-13 Thread Bandan Das
Signed-off-by: Bandan Das --- drivers/vhost/net.c | 25 +++ drivers/vhost/vhost.c | 115 +++--- drivers/vhost/vhost.h | 6 +++ 3 files changed, 130 insertions(+), 16 deletions(-) diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c index

Re: Equivalent of vmware SIOC (Storage IO Control) in KVM

2013-10-13 Thread Fam Zheng
On Mon, 10/14 02:13, Andrey Korolyov wrote: > Hello, > > By the way, is there plans to enhance qemu I/O throttling to able to > swallow peaks or to apply various disciplines? Current one-second flat > discipline seemingly is not enough for uneven workloads especially > when there is no alternative

Re: lkvm issue: --console virtio keyboard doesn't work

2013-10-13 Thread Asias He
On Sat, Oct 12, 2013 at 11:50:37AM +0200, Pekka Enberg wrote: > On 10/12/13 11:09 AM, P?ter Szab? wrote: > >I'm using afdf92030c7c43b0f9b32b7edbe07ac3b13780f1 from > >git://github.com/penberg/linux-kvm.git and Linux kernel 3.2.51: > > > >config-3.2.51 (42 KB) > >https://mega.co.nz/#!hgxB1TDJ!SdbX-j

Re: lkvm issue: --network mode=user,trans=mmio doesn't work

2013-10-13 Thread Asias He
On Sat, Oct 12, 2013 at 10:01:42AM -0400, Sasha Levin wrote: > On 10/12/2013 05:49 AM, Pekka Enberg wrote: > >Hi Peter, > > > >(Adding bunch of CCs.) > > > >On 10/12/13 11:05 AM, P?ter Szab? wrote: > >>First, thank you very much for writing lkvm, it's awesome, and very > >>easy to set up. > >> > >>

Re: Equivalent of vmware SIOC (Storage IO Control) in KVM

2013-10-13 Thread Andrey Korolyov
Hello, By the way, is there plans to enhance qemu I/O throttling to able to swallow peaks or to apply various disciplines? Current one-second flat discipline seemingly is not enough for uneven workloads especially when there is no alternative like cgroups for nbd usage. Thanks! On Sun, Oct 13, 2

Re: [PATCH 0/6] Get rid of mandatory huge page defines

2013-10-13 Thread Christoffer Dall
On Sun, Oct 13, 2013 at 08:49:28PM +0300, Gleb Natapov wrote: > On Sun, Oct 13, 2013 at 06:41:22PM +0100, Christoffer Dall wrote: > > On Sun, Oct 13, 2013 at 12:31:29PM +0300, Gleb Natapov wrote: > > > On Sat, Oct 12, 2013 at 10:36:36AM -0700, Christoffer Dall wrote: > > > > On Wed, Oct 02, 2013 at

Re: [PATCH 0/6] Get rid of mandatory huge page defines

2013-10-13 Thread Gleb Natapov
On Sun, Oct 13, 2013 at 06:41:22PM +0100, Christoffer Dall wrote: > On Sun, Oct 13, 2013 at 12:31:29PM +0300, Gleb Natapov wrote: > > On Sat, Oct 12, 2013 at 10:36:36AM -0700, Christoffer Dall wrote: > > > On Wed, Oct 02, 2013 at 02:22:27PM -0700, Christoffer Dall wrote: > > > > The current KVM cod

Re: [PATCH 0/6] Get rid of mandatory huge page defines

2013-10-13 Thread Christoffer Dall
On Sun, Oct 13, 2013 at 12:31:29PM +0300, Gleb Natapov wrote: > On Sat, Oct 12, 2013 at 10:36:36AM -0700, Christoffer Dall wrote: > > On Wed, Oct 02, 2013 at 02:22:27PM -0700, Christoffer Dall wrote: > > > The current KVM code relies on a number of definitions to exist for each > > > architecture:

Re: Problem on enabling kernel module with Ubuntu 12.04 LTS

2013-10-13 Thread Alex Williamson
On Sun, 2013-10-13 at 18:52 +0300, Sami wrote: > Hello! > > > I'm getting this error: > > root@Ubuntu-Server:/# sudo /sbin/modprobe kvm-intel > FATAL: Error inserting kvm_intel > (/lib/modules/3.5.0-41-generic/kernel/arch/x86/kvm/kvm-intel.ko): > Operation not supported > > CPU: Intel Pentium 4

Problem on enabling kernel module with Ubuntu 12.04 LTS

2013-10-13 Thread Sami
Hello! I'm getting this error: root@Ubuntu-Server:/# sudo /sbin/modprobe kvm-intel FATAL: Error inserting kvm_intel (/lib/modules/3.5.0-41-generic/kernel/arch/x86/kvm/kvm-intel.ko): Operation not supported CPU: Intel Pentium 4 2,60GHz (If that matters) Please advice. -- To unsubscribe from th

Re: [PATCH] VFIO: vfio_iommu_type1: fix bug caused by break in nested loop

2013-10-13 Thread Alex Williamson
On Fri, 2013-10-11 at 17:47 +0200, Antonios Motakis wrote: > In vfio_iommu_type1.c there is a bug in vfio_dma_do_map, when checking > that pages are not already mapped. Since the check is being done in a > for loop nested within the main loop, breaking out of it does not create > the intended behav

Re: Equivalent of vmware SIOC (Storage IO Control) in KVM

2013-10-13 Thread Paolo Bonzini
Il 12/10/2013 08:09, Soumendu Satapathy (sosatapa) ha scritto: > Do we have an equivalent of vmware SIOC like feature in KVM? Yes, you have two choices: 1) use cgroups to throttle I/O at the level of the host disk (i.e. multiple virtual disks stored on the same disk share the limit). If you're u

Re: [PATCH 0/6] Get rid of mandatory huge page defines

2013-10-13 Thread Gleb Natapov
On Sat, Oct 12, 2013 at 10:36:36AM -0700, Christoffer Dall wrote: > On Wed, Oct 02, 2013 at 02:22:27PM -0700, Christoffer Dall wrote: > > The current KVM code relies on a number of definitions to exist for each > > architecture: > > KVM_HPAGE_GFN_SHIFT > > KVM_NR_HPAGE_SIZES > > KVM_PAGES_PER

Re: [Patchv5 6/7] KVM: async_pf: Async page fault support on s390

2013-10-13 Thread Gleb Natapov
On Tue, Oct 08, 2013 at 04:54:59PM +0200, Christian Borntraeger wrote: > diff --git a/arch/s390/include/uapi/asm/kvm.h > b/arch/s390/include/uapi/asm/kvm.h > index 33d52b8..1e8fced 100644 > --- a/arch/s390/include/uapi/asm/kvm.h > +++ b/arch/s390/include/uapi/asm/kvm.h > @@ -17,9 +17,12 @@ > #def

Re: [Patchv5 6/7] KVM: async_pf: Async page fault support on s390

2013-10-13 Thread Gleb Natapov
On Tue, Oct 08, 2013 at 04:54:59PM +0200, Christian Borntraeger wrote: > From: Dominik Dingel > > This patch enables async page faults for s390 kvm guests. > It provides the userspace API to enable, disable and disable_wait this > feature. > By providing disable and disable_wait, the userspace c

Re: [Patchv5 5/7] KVM: async_pf: Allow to wait for outstanding work

2013-10-13 Thread Gleb Natapov
On Sun, Oct 13, 2013 at 11:48:03AM +0300, Gleb Natapov wrote: > On Tue, Oct 08, 2013 at 04:54:58PM +0200, Christian Borntraeger wrote: > > From: Dominik Dingel > > > > kvm_clear_async_pf_completion get an additional flag to either cancel > > outstanding > > work or wait for oustanding work to be

Re: [Patchv5 5/7] KVM: async_pf: Allow to wait for outstanding work

2013-10-13 Thread Gleb Natapov
On Tue, Oct 08, 2013 at 04:54:58PM +0200, Christian Borntraeger wrote: > From: Dominik Dingel > > kvm_clear_async_pf_completion get an additional flag to either cancel > outstanding > work or wait for oustanding work to be finished, x86 currentlx cancels all > work. > I do not see why x86 would

Re: [Patchv5 2/7] KVM: s390: add floating irq controller

2013-10-13 Thread Gleb Natapov
On Tue, Oct 08, 2013 at 04:54:55PM +0200, Christian Borntraeger wrote: > From: Jens Freimann > > This patch adds a floating irq controller as a kvm_device. > It will be necessary for migration of floating interrupts as well > as for hardening the reset code by allowing user space to explicitly >

Re: [PATCH] kvm fix: Enable pvspinlock after jump_label_init() to avoid VM hang

2013-10-13 Thread Gleb Natapov
On Sun, Oct 13, 2013 at 01:38:57PM +0530, Raghavendra K T wrote: > On 10/09/2013 02:33 PM, Raghavendra K T wrote: > >We use jump label to enable pv-spinlock. With the changes in > >(442e0973e927 Merge branch 'x86/jumplabel'), the jump label behaviour has > >changed > >that would result in evntual

Re: [PATCH] kvm fix: Enable pvspinlock after jump_label_init() to avoid VM hang

2013-10-13 Thread Raghavendra K T
On 10/09/2013 02:33 PM, Raghavendra K T wrote: We use jump label to enable pv-spinlock. With the changes in (442e0973e927 Merge branch 'x86/jumplabel'), the jump label behaviour has changed that would result in evntual hang of the VM since we would end up in a situation where slowpath locks woul