https://bugzilla.kernel.org/show_bug.cgi?id=42755
Summary: KVM is being extremely slow on AMD Athlon64 4000+ Dual
Core 2.1GHz Brisbane
Product: Virtualization
Version: unspecified
Kernel Version: 3.2.2
Platform: All
OS/Vers
https://bugzilla.kernel.org/show_bug.cgi?id=42755
--- Comment #1 from Rosen 2012-02-10 08:06:25 ---
in kvm guest this speed are about 2MB/s and is slow like a hell. With it
booting for about 15 minutes.
Same guest machine boot for about 4-7 seconds on virtualbox/vmware
--
Configure bugmai
https://bugzilla.kernel.org/show_bug.cgi?id=42755
--- Comment #2 from Michael Tokarev 2012-02-10 09:18:35 ---
This is very unlikely to do with the kernel. Qemu (which you're using to run
your guest -- kvm is just in-kernel module) will emulate whole machine
including CPU in case it can't ac
This changes are neccesary because people who use
the test want to use this test another way.
Signed-off-by: Jiří Župka
---
client/tests/kvm/tests/cpuflags.py | 890 +---
client/virt/subtests.cfg.sample| 24 +-
2 files changed, 438 insertions(+), 476 deleti
Adds new subtest which tests warning of
-cpu started with parameter "check".
Signed-off-by: Jiří Župka
---
client/tests/kvm/tests/cpuflags.py | 35 +++
client/virt/subtests.cfg.sample|2 ++
client/virt/virt_utils.py |1 +
3 files changed, 3
https://bugzilla.kernel.org/show_bug.cgi?id=42755
--- Comment #3 from Rosen 2012-02-10 09:36:35 ---
cmdline:
sudo kvm -enable-kvm -cpu 'host' -smp '2,cores=2,threads=2,sockets=1' -m 512M
-vga vmware -usbdevice tablet -soundhw es1370 -drive
file=test-amd64.img,if=virtio,cache=none -netdev
ty
If kvm__check_extensions found that some of the required
KVM extention is not supported by OS, we should stop the
init and free all allocated resources.
Signed-off-by: Yang Bai
---
tools/kvm/kvm.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/tools/kvm/kvm.c b/tools/kv
When error occurs, just set the ret to the reason,
then jump to the error handle labels.
This makes the code more readable.
Signed-off-by: Yang Bai
---
tools/kvm/kvm.c | 11 ++-
1 files changed, 6 insertions(+), 5 deletions(-)
diff --git a/tools/kvm/kvm.c b/tools/kvm/kvm.c
index 8e749
If kvm_ipc__start failed, it returns a negative and by checking
this return value, we can ensure that it succeeds.
Signed-off-by: Yang Bai
---
tools/kvm/kvm-ipc.c | 38 --
tools/kvm/kvm.c |7 ++-
2 files changed, 38 insertions(+), 7 deletions(-)
By checking the return value from kvm_ipc__register_handler,
we can ensure that it succeeds.
Signed-off-by: Yang Bai
---
tools/kvm/kvm.c |8 +++-
1 files changed, 7 insertions(+), 1 deletions(-)
diff --git a/tools/kvm/kvm.c b/tools/kvm/kvm.c
index f02d5df..99bcef4 100644
--- a/tools/kvm
On (Thu) 09 Feb 2012 [16:13:29], Igor Mammedov wrote:
> Stalls are probably caused by uninitialized percpu hv_clock, with
> following patch I don't see stalls. Although I might be just lucky.
> http://git.kernel.org/?p=virt/kvm/kvm.git;a=commit;h=e2971ac7e1d186af059e088d305496c5cb47d487
Your comm
And add a new flag definition in kvm_ppc_pvinfo to indicate
whether host support EV_IDLE hcall.
Signed-off-by: Liu Yu
---
v3:
no change
arch/powerpc/include/asm/kvm_para.h | 14 --
arch/powerpc/kvm/powerpc.c |8
include/linux/kvm.h |2 ++
from the kvm guest paravirt init code.
Signed-off-by: Liu Yu
---
v3:
apply the epapr init for all ppc platform
arch/powerpc/Kconfig|4 +++
arch/powerpc/include/asm/epapr_hcalls.h |8 +
arch/powerpc/kernel/Makefile|1 +
arch/powerpc/kernel/epapr_pa
If the guest hypervisor node contains "has-idle" property.
Signed-off-by: Liu Yu
---
v3:
1. apply the hcall idle for all ppc platform
2. add a loop to prevent spurious wakeups
arch/powerpc/kernel/Makefile |2 +-
arch/powerpc/kernel/epapr.S | 47 +++
Adding Suresh and Peter to Cc.
On 02/10/2012 01:16 AM, Sasha Levin wrote:
> On Wed, Feb 8, 2012 at 7:59 PM, Josh Boyer wrote:
>> On Wed, Feb 8, 2012 at 8:31 PM, Sasha Levin wrote:
>>> Hi all,
>>>
>>> I got the following warning when shutting down a KVM guest with a whole
>>> bunch of cores (2
hello,
Le Sun, Feb 05, 2012 at 11:38:34AM +0200, Avi Kivity ecrivait :
>
> Please post a trace as documented in http://www.linux-kvm.org/page/Tracing.
I made the trace : started just before the slow function launch
and stoped just after. I start only one VM with 2 vcpus/1
Could you send me your .config and commit id of kernel you are using?
- Original Message -
> From: "Amit Shah"
> To: "Igor Mammedov"
> Cc: "Marcelo Tosatti" , kvm@vger.kernel.org,
> t...@linutronix.de, mi...@redhat.com,
> h...@zytor.com, x...@kernel.org, johns...@us.ibm.com, r...@redhat
On (Fri) 10 Feb 2012 [05:11:00], Igor Mammedov wrote:
> Could you send me your .config and commit id of kernel you are using?
Kernel's based on bd3ce7d57c380af110c86d19e256115d0e7053ca plus your
commit + Marcelo's patch.
config is attached below.
#
# Automatically generated file; DO NOT EDIT.
#
This year's Google Summer of Code has been announced:
http://www.google-melange.com/gsoc/events/google/gsoc2012
For those who haven't heard of GSoC before, it funds university
students to work on open source projects during the summer.
Organizations, such as QEMU, can participate to attract stude
On Fri, Feb 10, 2012 at 10:30:24AM +, Stefan Hajnoczi wrote:
> This year's Google Summer of Code has been announced:
>
> http://www.google-melange.com/gsoc/events/google/gsoc2012
>
> For those who haven't heard of GSoC before, it funds university
> students to work on open source projects dur
On Sun, Feb 5, 2012 at 11:15 AM, Paolo Bonzini wrote:
> This is the first implementation of the virtio-scsi driver, a virtual
> HBA that will be supported by KVM. It implements a subset of the spec,
> in particular it does not implement asynchronous notifications for either
> LUN reset/removal/ad
On Fri, Feb 10, 2012 at 10:59 AM, Daniel P. Berrange
wrote:
> On Fri, Feb 10, 2012 at 10:30:24AM +, Stefan Hajnoczi wrote:
>> This year's Google Summer of Code has been announced:
>>
>> http://www.google-melange.com/gsoc/events/google/gsoc2012
>>
>> For those who haven't heard of GSoC before,
On Fri, Feb 10, 2012 at 11:55 AM, Yang Bai wrote:
> If kvm__check_extensions found that some of the required
> KVM extention is not supported by OS, we should stop the
> init and free all allocated resources.
>
> Signed-off-by: Yang Bai
Applied all four patches, thanks!
--
To unsubscribe from th
On Fri, Feb 10, 2012 at 03:32:11PM +0530, Amit Shah wrote:
> On (Thu) 09 Feb 2012 [16:13:29], Igor Mammedov wrote:
>
> > Stalls are probably caused by uninitialized percpu hv_clock, with
> > following patch I don't see stalls. Although I might be just lucky.
> > http://git.kernel.org/?p=virt/kvm/k
On Fri, Feb 10, 2012 at 10:32:16AM -0200, Marcelo Tosatti wrote:
> On Fri, Feb 10, 2012 at 03:32:11PM +0530, Amit Shah wrote:
> > On (Thu) 09 Feb 2012 [16:13:29], Igor Mammedov wrote:
> >
> > > Stalls are probably caused by uninitialized percpu hv_clock, with
> > > following patch I don't see stal
On 02/10/2012 01:33 PM, Marcelo Tosatti wrote:
On Fri, Feb 10, 2012 at 10:32:16AM -0200, Marcelo Tosatti wrote:
On Fri, Feb 10, 2012 at 03:32:11PM +0530, Amit Shah wrote:
On (Thu) 09 Feb 2012 [16:13:29], Igor Mammedov wrote:
Stalls are probably caused by uninitialized percpu hv_clock, with
fo
Avi Kivity wrote:
> On 02/09/2012 04:23 PM, Avi Kivity wrote:
> > > BTW do we really need fast slot creation/destruction?
> >
> > Not really, but it's good to have infrastructure that copes with
> > different workloads. If the patches keep the code simple I think it's a
> > good thing to have.
On (Fri) 10 Feb 2012 [10:32:16], Marcelo Tosatti wrote:
> On Fri, Feb 10, 2012 at 03:32:11PM +0530, Amit Shah wrote:
> > On (Thu) 09 Feb 2012 [16:13:29], Igor Mammedov wrote:
> >
> > > Stalls are probably caused by uninitialized percpu hv_clock, with
> > > following patch I don't see stalls. Altho
Avi Kivity wrote:
> > 2. When we create(and shift?) a memory slot, we call kvm_arch_flush_shadow()
> > to clear all mmio sptes, again not restricted to that slot.
> >
> > /*
> > * If the new memory slot is created, we need to clear all
> > * mmio sptes.
> > */
> > if (npage
On 2/9/12 9:36 AM, "John Fastabend" wrote:
> On 2/8/2012 8:36 PM, Stephen Hemminger wrote:
>> On Wed, 08 Feb 2012 19:22:06 -0800
>> John Fastabend wrote:
>>
>>> Propagate software FDB table into hardware uc, mc lists when
>>> the NETIF_F_HW_FDB is set.
>>>
>>> This resolves the case below w
Hi,
Recently I observed performance regression regarding virtio-blk,
especially different IO bandwidths between qemu-kvm 0.14.1 and 1.0.
So I want to share the benchmark results, and ask you what the reason
would be.
1. Test condition
- On host, ramdisk-backed block device (/dev/ram0)
- qemu-k
Hi John,
I went backwards to summarize at the top after going through your email.
TL;DR version 0.1:
you provide a good use case where it makes sense to do things in the
kernel. IMO, you could make the same arguement if your embedded switch
could do ACLs, IPv4 forwarding etc. And the kernel bloa
On Wed, Feb 08, 2012 at 10:07:44AM -0500, Eric B Munson wrote:
> When a host stops or suspends a VM it will set a flag to show this. The
> watchdog will use these functions to determine if a softlockup is real, or the
> result of a suspended VM.
>
> Signed-off-by: Eric B Munson
> asm-generic cha
On Fri, 10 Feb 2012 10:18:31 -0500
jamal wrote:
> Hi John,
>
> I went backwards to summarize at the top after going through your email.
>
> TL;DR version 0.1:
> you provide a good use case where it makes sense to do things in the
> kernel. IMO, you could make the same arguement if your embedde
On Thu, Feb 09, 2012 at 04:25:36PM +0200, Avi Kivity wrote:
> On 02/08/2012 08:45 PM, Marcelo Tosatti wrote:
> > > BTW do we really need fast slot creation/destruction?
> >
> > At the moment yes. Boot a RHEL/Fedora installation disk (or any other
> > guest which uses SYSLINUX splash screen) and you
On Fri, Feb 10, 2012 at 10:08:12PM +0900, Takuya Yoshikawa wrote:
> Avi Kivity wrote:
>
> > On 02/09/2012 04:23 PM, Avi Kivity wrote:
> > > > BTW do we really need fast slot creation/destruction?
> > >
> > > Not really, but it's good to have infrastructure that copes with
> > > different workload
PCI 2.3 allows to generically disable IRQ sources at device level. This
enables us to share legacy IRQs of such devices with other host devices
when passing them to a guest.
The new IRQ sharing feature introduced here is optional, user space has
to request it explicitly. Moreover, user space can i
Use OPTION_ROM_START/END from the common header file, add comment to
init code.
Signed-off-by: Jan Kiszka
---
pc-bios/optionrom/kvmvapic.S | 18 --
1 files changed, 8 insertions(+), 10 deletions(-)
diff --git a/pc-bios/optionrom/kvmvapic.S b/pc-bios/optionrom/kvmvapic.S
index
In order to perform critical manipulations on the VM state in the
context of a VCPU, specifically code patching, stopping and resuming of
all VCPUs may be necessary. resume_all_vcpus is already compatible, now
enable pause_all_vcpus for this use case by stopping the calling context
before starting
This enables acceleration for MMIO-based TPR registers accesses of
32-bit Windows guest systems. It is mostly useful with KVM enabled,
either on older Intel CPUs (without flexpriority feature, can also be
manually disabled for testing) or any current AMD processor.
The approach introduced here is
This will allow the APIC core to file a TPR access report. Depending on
the accelerator and kernel irqchip mode, it will either be delivered
right away or queued for later reporting.
In TCG mode, we can restart the triggering instruction and can therefore
forward the event directly. KVM does not a
Here is v2 of the TPR access optimization. Changes:
- plug race between patching and running VCPUs accessing the same TPR
instruction by stopping VCPUs during patch process
- realized forward/backward check in evaluate_tpr_instruction via a
table but kept patch_instruction as is (too much v
The CH registers is only written, never read. So we can remove these
operations and, in case of up_set_tpr, also the ECX push/pop.
Signed-off-by: Jan Kiszka
---
pc-bios/optionrom/kvmvapic.S |6 +-
1 files changed, 1 insertions(+), 5 deletions(-)
diff --git a/pc-bios/optionrom/kvmvapic.S
This imports and builds the original VAPIC option ROM of qemu-kvm.
Its interaction with QEMU is described in the commit that introduces the
corresponding device model.
Signed-off-by: Jan Kiszka
---
.gitignore |1 +
Makefile |2 +-
pc-bios/optionrom/M
As we have thread-local cpu_single_env now and KVM uses exactly one
thread per VCPU, we can drop the cpu_single_env updates from the loop
and initialize this variable only once during setup.
Signed-off-by: Jan Kiszka
---
cpus.c|1 +
kvm-all.c |5 -
2 files changed, 1 insertions(+
Always add a byte before the final 512-bytes alignment to reserve the
space for the ROM checksum.
Signed-off-by: Jan Kiszka
---
pc-bios/optionrom/optionrom.h |3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/pc-bios/optionrom/optionrom.h b/pc-bios/optionrom/optionrom.h
i
On 02/10/2012 04:02 AM, Liu Yu wrote:
> from the kvm guest paravirt init code.
>
> Signed-off-by: Liu Yu
> ---
> v3:
> apply the epapr init for all ppc platform
>
> arch/powerpc/Kconfig|4 +++
> arch/powerpc/include/asm/epapr_hcalls.h |8 +
> arch/powerpc/kernel/
On 02/10/2012 04:02 AM, Liu Yu wrote:
> +_GLOBAL(epapr_ev_idle)
> +epapr_ev_idle:
> +#ifdef CONFIG_E500
> + rlwinm r3,r1,0,0,31-THREAD_SHIFT /* current thread_info */
> + lwz r4,TI_LOCAL_FLAGS(r3) /* set napping bit */
> + ori r4,r4,_TLF_NAPPING /* so when we take
On Fri, 2012-02-10 at 15:36 +0530, Srivatsa S. Bhat wrote:
> >>> [ 32.448626] [ cut here ]
> >>> [ 32.449160] WARNING: at arch/x86/kernel/smp.c:119
> >>> native_smp_send_reschedule+0x25/0x43()
> >>> [ 32.449621] Pid: 1, comm: init_stage2 Not tainted 3.2.0+ #14
> >>> [
On Fri, 2012-02-10 at 19:58 +0100, Peter Zijlstra wrote:
> OK, so a 'modern' kernel does it slightly different and I've no idea
> what exactly goes wrong in your vintage version. But I can see the
> current stuff going at it all wrong.
>
> What seems to happen is that native_nmi_stop_other_cpus()
On Fri, Feb 10, 2012 at 08:03:53PM +0100, Peter Zijlstra wrote:
> On Fri, 2012-02-10 at 19:58 +0100, Peter Zijlstra wrote:
> > OK, so a 'modern' kernel does it slightly different and I've no idea
> > what exactly goes wrong in your vintage version. But I can see the
> > current stuff going at it al
On Fri, 2012-02-10 at 15:02 -0500, Don Zickus wrote:
> I also ran into the same problem you did and hacked up another patch that
> checked a global atomic variable that let the system know we were shutting
> down and not to do the WARN_ON (the global is already created for the NMI
> case now).
sy
On Fri, Feb 10, 2012 at 09:18:41PM +0100, Peter Zijlstra wrote:
> On Fri, 2012-02-10 at 15:02 -0500, Don Zickus wrote:
> > I also ran into the same problem you did and hacked up another patch that
> > checked a global atomic variable that let the system know we were shutting
> > down and not to do
On Fri, 2012-02-10 at 15:31 -0500, Don Zickus wrote:
> So my second patch which I will eventually post will just skip the WARN_ON
> if the system is going down. Not sure if that is the proper way to address
> this problem or change all of the stop_this_cpu code to use a different
> bitmask than th
I've to revoke my ack and say NAK to this patch. Patch itself is in right
direction but clock restore happens too late.
With patch I've used to hunt down overflow for cpu hotplug
(maybe it will be better for it to be in kernel, which will help to detect
overflow problem without spending a lot of
On Fri, Feb 10, 2012 at 09:36:03PM +0100, Peter Zijlstra wrote:
> On Fri, 2012-02-10 at 15:31 -0500, Don Zickus wrote:
> > So my second patch which I will eventually post will just skip the WARN_ON
> > if the system is going down. Not sure if that is the proper way to address
> > this problem or c
In order to avoid compilation failure when KVM is not compiled in,
guard the mmu_notifier specific sections with both CONFIG_MMU_NOTIFIER
and KVM_ARCH_WANT_MMU_NOTIFIER, like it is being done in the rest of
the KVM code.
Signed-off-by: Marc Zyngier
---
include/linux/kvm_host.h |4 ++--
1 fil
Use __kvm_hyp_code_end to mark the end of the main HYP code instead of
__kvm_vcpu_run_end. It's a bit cleaner as we're about to add more code
to that section.
Signed-off-by: Marc Zyngier
---
arch/arm/include/asm/kvm_asm.h |3 ++-
arch/arm/kvm/arm.c |4 ++--
arch/arm/kvm/inter
Add the necessary infrastructure to handle MMU notifiers on KVM/ARM.
As we don't have shadow page tables, the implementation is actually very
simple. The only supported operation is kvm_unmap_hva(), where we remove
the HVA from the 2nd stage translation. All other hooks are NOPs.
Signed-off-by: Ma
On Fri, Feb 10, 2012 at 2:22 PM, Marc Zyngier wrote:
> Use __kvm_hyp_code_end to mark the end of the main HYP code instead of
> __kvm_vcpu_run_end. It's a bit cleaner as we're about to add more code
> to that section.
this is good, but should we not rename the beginning of the section as
well (so
On Fri, Feb 10, 2012 at 2:22 PM, Marc Zyngier wrote:
> Add the necessary infrastructure to handle MMU notifiers on KVM/ARM.
> As we don't have shadow page tables, the implementation is actually very
> simple. The only supported operation is kvm_unmap_hva(), where we remove
> the HVA from the 2nd s
61 matches
Mail list logo
