kvm somehow exempt from frequency scaling?

kvm somehow exempt from frequency scaling?

[David Abrahams]

I'm using the ondemand governor on my core duo laptop.  Even when kvm
is taking 99% of the CPU, somehow instead of the ondemand governor
kicking it up to 2GHz, the CPU is still running at half speed.  What's
up with that?  Anybody know?

TIA,
-- 
Dave Abrahams
Boost Consulting
www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[Avi Kivity]

David Abrahams wrote:
> I'm using the ondemand governor on my core duo laptop.  Even when kvm
> is taking 99% of the CPU, somehow instead of the ondemand governor
> kicking it up to 2GHz, the CPU is still running at half speed.  What's
> up with that?  Anybody know?
>
>   

Strange.  Does a simple bash busy-loop (while :; do :; done) work?

-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Mon Apr 09 2007, Avi Kivity <avi-AT-qumranet.com> wrote:

> David Abrahams wrote:
>> I'm using the ondemand governor on my core duo laptop.  Even when kvm
>> is taking 99% of the CPU, somehow instead of the ondemand governor
>> kicking it up to 2GHz, the CPU is still running at half speed.  What's
>> up with that?  Anybody know?
>>
>>   
>
> Strange.  Does a simple bash busy-loop (while :; do :; done) work?

The frequency goes up to 1.67Ghz.

I've seen it happen that the CPU is pegged at 100% in the WinXP Guest
OS's task manager with my frequency scaling never hitting the max of
2.17 Ghz.  Right now I'm see variable CPU load on the guest averaging
around 75%, and it does cause frequency scaling to hit the max every
once in a while.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[Avi Kivity]

David Abrahams wrote:
> on Mon Apr 09 2007, Avi Kivity <avi-AT-qumranet.com> wrote:
>
>   
>> David Abrahams wrote:
>>     
>>> I'm using the ondemand governor on my core duo laptop.  Even when kvm
>>> is taking 99% of the CPU, somehow instead of the ondemand governor
>>> kicking it up to 2GHz, the CPU is still running at half speed.  What's
>>> up with that?  Anybody know?
>>>
>>>   
>>>       
>> Strange.  Does a simple bash busy-loop (while :; do :; done) work?
>>     
>
> The frequency goes up to 1.67Ghz.
>
>   

Sounds like ondemand is broken.


-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[Michael Riepe]

Hi!

Avi Kivity wrote:

> Sounds like ondemand is broken.

Maybe it only switches when the load rises in user space. With KVM, it
appears in system space. Got to cross-check that on my Core Duo.

Michael.

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Tue Apr 10 2007, Michael Riepe <michael-0QoEqw4nQxo-AT-public.gmane.org> wrote:

> Hi!
>
> Avi Kivity wrote:
>
>> Sounds like ondemand is broken.
>
> Maybe it only switches when the load rises in user space. With KVM, it
> appears in system space. Got to cross-check that on my Core Duo.

More data: it definitely seems to make a difference whether the VM's
desktop is showing.  If I switch Gnome workspaces so it's not visible,
it doesn't seem to drive the frequency up.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Wed Apr 11 2007, David Abrahams <dave-UB3wUj7V41K5azolltMz9laTQe2KTcn/-AT-public.gmane.org> wrote:

> on Tue Apr 10 2007, Michael Riepe <michael-0QoEqw4nQxo-AT-public.gmane.org> wrote:
>
>> Hi!
>>
>> Avi Kivity wrote:
>>
>>> Sounds like ondemand is broken.
>>
>> Maybe it only switches when the load rises in user space. With KVM, it
>> appears in system space. Got to cross-check that on my Core Duo.
>
> More data: it definitely seems to make a difference whether the VM's
> desktop is showing.  If I switch Gnome workspaces so it's not visible,
> it doesn't seem to drive the frequency up.

In fact, kvm doesn't seem to make any progress at all :(.  I'm
currently going through the "microsoft update" dance on a VM there,
and I literally left for an hour during which there was no progress
because the QEMU window wasn't visible.

Even stranger, if I minimize the QEMU window and then switch
workspaces, it gets plenty of cycles.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[Michael Riepe]

Hi!

David Abrahams wrote:
[...]
>>More data: it definitely seems to make a difference whether the VM's
>>desktop is showing.  If I switch Gnome workspaces so it's not visible,
>>it doesn't seem to drive the frequency up.

Because it doesn't have to redraw the window?

What happens if you hide the qemu window below another one, e.g. your
favorite web browser?

> In fact, kvm doesn't seem to make any progress at all :(.  I'm
> currently going through the "microsoft update" dance on a VM there,
> and I literally left for an hour during which there was no progress
> because the QEMU window wasn't visible.
> 
> Even stranger, if I minimize the QEMU window and then switch
> workspaces, it gets plenty of cycles.

I guess that's a qemu and/or SDL issue.

Ideally, the screen should only be updated when the window is mapped
(via XMapWindow) *and* visible (that is, neither fully obscured nor
moved outside the visible part of the screen). Since "fully obscured" is
hard to test for, this usually translates to "mapped and not off-screen".

When a window is minimized ("iconified"), it usually is unmapped and
another one (the "icon window") is mapped instead. When switching
workspaces, however, some window managers just move the window away (say
to x=10000,y=10000) but keep it mapped. If there is no test for
visibility (or if the test doesn't work correctly), SDL may happily
continue updating the window's contents - and that will burn a lot of
cycles.

IMO, there's a lot of room for improvement here. But it has nothing to
do with kvm in the first place (unless you consider qemu an integral
part of kvm, which I don't).

-- 
Michael "Tired" Riepe <michael-0QoEqw4nQxo@public.gmane.org>
X-Tired: Each morning I get up I die a little

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Wed Apr 11 2007, Michael Riepe <michael-AT-mr511.de> wrote:

> Hi!
>
> David Abrahams wrote:
> [...]
>>>More data: it definitely seems to make a difference whether the VM's
>>>desktop is showing.  If I switch Gnome workspaces so it's not visible,
>>>it doesn't seem to drive the frequency up.
>
> Because it doesn't have to redraw the window?
>
> What happens if you hide the qemu window below another one, e.g. your
> favorite web browser?

It seems to make progress.  I can play internet radio in the VM and I
keep hearing it (glitchy, but it doesn't stop).

>> In fact, kvm doesn't seem to make any progress at all :(.  I'm
>> currently going through the "microsoft update" dance on a VM there,
>> and I literally left for an hour during which there was no progress
>> because the QEMU window wasn't visible.
>> 
>> Even stranger, if I minimize the QEMU window and then switch
>> workspaces, it gets plenty of cycles.
>
> I guess that's a qemu and/or SDL issue.

SDL?

> Ideally, the screen should only be updated when the window is mapped
> (via XMapWindow) *and* visible (that is, neither fully obscured nor
> moved outside the visible part of the screen). Since "fully obscured" is
> hard to test for, this usually translates to "mapped and not off-screen".

Yeah, but I don't care about the window being updated.  I want the
virtual CPU to make progress.  Switching workspaces is literally like
suspending the VM!

> When a window is minimized ("iconified"), it usually is unmapped and
> another one (the "icon window") is mapped instead. When switching
> workspaces, however, some window managers just move the window away (say
> to x=10000,y=10000) but keep it mapped. If there is no test for
> visibility (or if the test doesn't work correctly), SDL may happily
> continue updating the window's contents - and that will burn a lot of
> cycles.
>
> IMO, there's a lot of room for improvement here. But it has nothing to
> do with kvm in the first place (unless you consider qemu an integral
> part of kvm, which I don't).

I don't either, but it's not 100% clear to me yet that this is a qemu
problem.  Maybe I'll report it in a QEMU forum as well.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[Michael Riepe]

Hi!

David Abrahams wrote:
[...]
>>What happens if you hide the qemu window below another one, e.g. your
>>favorite web browser?
> 
> 
> It seems to make progress.  I can play internet radio in the VM and I
> keep hearing it (glitchy, but it doesn't stop).

What about the load?

By the way, how is your DISPLAY variable set? It should be ":0", not
"localhost:0" or "hostname:0". Otherwise video output will slow down
(and you'll probably see a lot of network load on the lo device).

>>>In fact, kvm doesn't seem to make any progress at all :(.  I'm
>>>currently going through the "microsoft update" dance on a VM there,
>>>and I literally left for an hour during which there was no progress
>>>because the QEMU window wasn't visible.
>>>
>>>Even stranger, if I minimize the QEMU window and then switch
>>>workspaces, it gets plenty of cycles.
>>
>>I guess that's a qemu and/or SDL issue.
> 
> 
> SDL?

Simple Directmedia Layer (see www.libsdl.org).
It's a library that Qemu uses for video output.

>>Ideally, the screen should only be updated when the window is mapped
>>(via XMapWindow) *and* visible (that is, neither fully obscured nor
>>moved outside the visible part of the screen). Since "fully obscured" is
>>hard to test for, this usually translates to "mapped and not off-screen".
> 
> 
> Yeah, but I don't care about the window being updated.  I want the
> virtual CPU to make progress.  Switching workspaces is literally like
> suspending the VM!

Maybe the qemu process is waiting for something (like an expose event
from the X server - which will never happen as long as the window is
invisible). What does ps/top report?

>>When a window is minimized ("iconified"), it usually is unmapped and
>>another one (the "icon window") is mapped instead. When switching
>>workspaces, however, some window managers just move the window away (say
>>to x=10000,y=10000) but keep it mapped. If there is no test for
>>visibility (or if the test doesn't work correctly), SDL may happily
>>continue updating the window's contents - and that will burn a lot of
>>cycles.
>>
>>IMO, there's a lot of room for improvement here. But it has nothing to
>>do with kvm in the first place (unless you consider qemu an integral
>>part of kvm, which I don't).
> 
> 
> I don't either, but it's not 100% clear to me yet that this is a qemu
> problem.  Maybe I'll report it in a QEMU forum as well.

Since it's related to screen output, it's likely to be a qemu/SDL
problem. The kvm module itself doesn't do any screen i/o.

Does qemu behave differently if you use -no-kvm?

-- 
Michael "Tired" Riepe <michael-0QoEqw4nQxo@public.gmane.org>
X-Tired: Each morning I get up I die a little

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Wed Apr 11 2007, Michael Riepe <michael-AT-mr511.de> wrote:

> Hi!
>
> David Abrahams wrote:
> [...]
>>>What happens if you hide the qemu window below another one, e.g. your
>>>favorite web browser?
>> 
>> 
>> It seems to make progress.  I can play internet radio in the VM and I
>> keep hearing it (glitchy, but it doesn't stop).
>
> What about the load?

kvm's load is independent of whether the window is obscured or
minimized, but when it's not on the visible Gnome workspace, the load
drops to near zero (and kvm gets nothing, or not much, done).

> By the way, how is your DISPLAY variable set? It should be ":0", not
> "localhost:0" or "hostname:0". Otherwise video output will slow down
> (and you'll probably see a lot of network load on the lo device).

It's ":0.0"

>>>>In fact, kvm doesn't seem to make any progress at all :(.  I'm
>>>>currently going through the "microsoft update" dance on a VM there,
>>>>and I literally left for an hour during which there was no progress
>>>>because the QEMU window wasn't visible.
>>>>
>>>>Even stranger, if I minimize the QEMU window and then switch
>>>>workspaces, it gets plenty of cycles.
>>>
>>>I guess that's a qemu and/or SDL issue.
>> 
>> 
>> SDL?
>
> Simple Directmedia Layer (see www.libsdl.org).
> It's a library that Qemu uses for video output.

Hm.

>>>Ideally, the screen should only be updated when the window is mapped
>>>(via XMapWindow) *and* visible (that is, neither fully obscured nor
>>>moved outside the visible part of the screen). Since "fully obscured" is
>>>hard to test for, this usually translates to "mapped and not off-screen".
>> 
>> 
>> Yeah, but I don't care about the window being updated.  I want the
>> virtual CPU to make progress.  Switching workspaces is literally like
>> suspending the VM!
>
> Maybe the qemu process is waiting for something (like an expose event
> from the X server - which will never happen as long as the window is
> invisible). What does ps/top report?

I've been using top to observe the load.  The state as reported by "ps
waux" is either Rl+ or Sl+.  Most of the time when qemu is on the
visible workspace, it's Rl+.  When not on the visible workspace, the
%CPU goes way down and the state is usually "Sl+".  Unless, of course,
the window is minimized, in which case it gets plenty of time to run.

>>>When a window is minimized ("iconified"), it usually is unmapped and
>>>another one (the "icon window") is mapped instead. When switching
>>>workspaces, however, some window managers just move the window away (say
>>>to x=10000,y=10000) but keep it mapped. If there is no test for
>>>visibility (or if the test doesn't work correctly), SDL may happily
>>>continue updating the window's contents - and that will burn a lot of
>>>cycles.

That doesn't explain why the "invisible workspace effect" doesn't
apply to iconified windows.

>>>IMO, there's a lot of room for improvement here. But it has nothing to
>>>do with kvm in the first place (unless you consider qemu an integral
>>>part of kvm, which I don't).
>> 
>> 
>> I don't either, but it's not 100% clear to me yet that this is a qemu
>> problem.  Maybe I'll report it in a QEMU forum as well.
>
> Since it's related to screen output, it's likely to be a qemu/SDL
> problem. The kvm module itself doesn't do any screen i/o.
>
> Does qemu behave differently if you use -no-kvm?

Haven't tried that one yet.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Wed Apr 11 2007, David Abrahams <dave-UB3wUj7V41K5azolltMz9laTQe2KTcn/-AT-public.gmane.org> wrote:

>> Does qemu behave differently if you use -no-kvm?
>
> Haven't tried that one yet.

Newsflash: yes, it does.  The VM keeps using cycles at a normal rate
even when not on a visible workspace.  This problem appears to be
KVM-specific.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[Avi Kivity]

David Abrahams wrote:
> on Wed Apr 11 2007, David Abrahams <dave-UB3wUj7V41K5azolltMz9laTQe2KTcn/-AT-public.gmane.org> wrote:
>
>   
>>> Does qemu behave differently if you use -no-kvm?
>>>       
>> Haven't tried that one yet.
>>     
>
> Newsflash: yes, it does.  The VM keeps using cycles at a normal rate
> even when not on a visible workspace.  This problem appears to be
> KVM-specific.
>   

Amazing.

Can you strace a hung vm and post the log?


-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[Anthony Liguori]

Avi Kivity wrote:
> David Abrahams wrote:
>   
>> on Wed Apr 11 2007, David Abrahams <dave-UB3wUj7V41K5azolltMz9laTQe2KTcn/-AT-public.gmane.org> wrote:
>>
>>   
>>     
>>>> Does qemu behave differently if you use -no-kvm?
>>>>       
>>>>         
>>> Haven't tried that one yet.
>>>     
>>>       
>> Newsflash: yes, it does.  The VM keeps using cycles at a normal rate
>> even when not on a visible workspace.  This problem appears to be
>> KVM-specific.
>>   
>>     
>
> Amazing.
>
> Can you strace a hung vm and post the log?
>   

I've heard of this occurring without KVM (although I don't recall if 
kqemu was involved).  I've never been able to reproduce myself.

An strace or core dump (via kill -SEGV) would be terribly useful in 
seeing what's going on.

Regards,

Anthony Liguori



-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Wed Apr 11 2007, Avi Kivity <avi-AT-qumranet.com> wrote:

> David Abrahams wrote:
>> on Wed Apr 11 2007, David Abrahams <dave-UB3wUj7V41K5azolltMz9laTQe2KTcn/-AT-public.gmane.org> wrote:
>>
>>   
>>>> Does qemu behave differently if you use -no-kvm?
>>>>       
>>> Haven't tried that one yet.
>>>     
>>
>> Newsflash: yes, it does.  The VM keeps using cycles at a normal rate
>> even when not on a visible workspace.  This problem appears to be
>> KVM-specific.
>>   
>
> Amazing.
>
> Can you strace a hung vm and post the log?

If you can tell me how, I'll be happy to try anything you like.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[Avi Kivity]

David Abrahams wrote:

    

>>> Newsflash: yes, it does.  The VM keeps using cycles at a normal rate
>>> even when not on a visible workspace.  This problem appears to be
>>> KVM-specific.
>>>   
>>>       
>> Amazing.
>>
>> Can you strace a hung vm and post the log?
>>     
>
> If you can tell me how, I'll be happy to try anything you like.
>
>   

do a

    strace -ttT -o /tmp/log -p pid-of-qemu-process

and post /tmp/log.


-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

[-- Attachment #1: Type: text/plain, Size: 584 bytes --]


on Thu Apr 12 2007, Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w-AT-public.gmane.org> wrote:

> David Abrahams wrote:
>
>     
>
>>>> Newsflash: yes, it does.  The VM keeps using cycles at a normal rate
>>>> even when not on a visible workspace.  This problem appears to be
>>>> KVM-specific.
>>>>   
>>>>       
>>> Amazing.
>>>
>>> Can you strace a hung vm and post the log?
>>>     
>>
>> If you can tell me how, I'll be happy to try anything you like.
>>
>>   
>
> do a
>
>     strace -ttT -o /tmp/log -p pid-of-qemu-process
>
> and post /tmp/log.


Here's about 5 seconds of strace:


[-- Attachment #2: log.bz2 --]
[-- Type: application/octet-stream, Size: 99214 bytes --]

[-- Attachment #3: Type: text/plain, Size: 125 bytes --]




-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com

[-- Attachment #4: Type: text/plain, Size: 345 bytes --]

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

[-- Attachment #5: Type: text/plain, Size: 186 bytes --]

_______________________________________________
kvm-devel mailing list
kvm-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/kvm-devel

Re: kvm somehow exempt from frequency scaling?

[Avi Kivity]

David Abrahams wrote:
> on Thu Apr 12 2007, Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w-AT-public.gmane.org> wrote:
>
>   
>> David Abrahams wrote:
>>
>>     
>>
>>     
>>>>> Newsflash: yes, it does.  The VM keeps using cycles at a normal rate
>>>>> even when not on a visible workspace.  This problem appears to be
>>>>> KVM-specific.
>>>>>   
>>>>>       
>>>>>           
>>>> Amazing.
>>>>
>>>> Can you strace a hung vm and post the log?
>>>>     
>>>>         
>>> If you can tell me how, I'll be happy to try anything you like.
>>>
>>>   
>>>       
>> do a
>>
>>     strace -ttT -o /tmp/log -p pid-of-qemu-process
>>
>> and post /tmp/log.
>>     
>
>
> Here's about 5 seconds of strace:
>   

qemu is getting cpu time, but not calling into kvm.

I'll try to reproduce this locally, right now I'm completely bewildered 
by this.

-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Thu Apr 12 2007, Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w-AT-public.gmane.org> wrote:

> qemu is getting cpu time, but not calling into kvm.
>
> I'll try to reproduce this locally, right now I'm completely bewildered 
> by this.

FYI, if I listen to the radio in windows media player, I don't see the
effect.  KVM will take 18% or so of the CPU even when not on a visible
workspace.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Thu Apr 12 2007, David Abrahams <dave-UB3wUj7V41K5azolltMz9laTQe2KTcn/-AT-public.gmane.org> wrote:

> on Thu Apr 12 2007, Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w-AT-public.gmane.org> wrote:
>
>> qemu is getting cpu time, but not calling into kvm.
>>
>> I'll try to reproduce this locally, right now I'm completely bewildered 
>> by this.
>
> FYI, if I listen to the radio in windows media player, I don't see the
> effect.  KVM will take 18% or so of the CPU even when not on a visible
> workspace.

And since ubuntu's update to kvm 1:16-1ubuntu2, I can't see the effect
at all, no matter what I do.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Mon Apr 16 2007, David Abrahams <dave-UB3wUj7V41K5azolltMz9laTQe2KTcn/-AT-public.gmane.org> wrote:

> on Thu Apr 12 2007, David Abrahams <dave-UB3wUj7V41K5azolltMz9laTQe2KTcn/-AT-public.gmane.org> wrote:
>
>> on Thu Apr 12 2007, Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w-AT-public.gmane.org> wrote:
>>
>>> qemu is getting cpu time, but not calling into kvm.
>>>
>>> I'll try to reproduce this locally, right now I'm completely bewildered 
>>> by this.
>>
>> FYI, if I listen to the radio in windows media player, I don't see the
>> effect.  KVM will take 18% or so of the CPU even when not on a visible
>> workspace.
>
> And since ubuntu's update to kvm 1:16-1ubuntu2, I can't see the effect
> at all, no matter what I do.

I take it back. If I launch kvm and switch to another desktop
immediately, I can come back a seemingly arbitrary amount of time
later and it will not have finished booting.  It also never seems to
force the CPU frequency to 100%.  AFAICT I am seeing exactly what I
was seeing before.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

Re: kvm somehow exempt from frequency scaling?

[Avi Kivity]

David Abrahams wrote:
>>>> I'll try to reproduce this locally, right now I'm completely bewildered 
>>>> by this.
>>>>         
>>> FYI, if I listen to the radio in windows media player, I don't see the
>>> effect.  KVM will take 18% or so of the CPU even when not on a visible
>>> workspace.
>>>       
>> And since ubuntu's update to kvm 1:16-1ubuntu2, I can't see the effect
>> at all, no matter what I do.
>>     
>
> I take it back. If I launch kvm and switch to another desktop
> immediately, I can come back a seemingly arbitrary amount of time
> later and it will not have finished booting.  It also never seems to
> force the CPU frequency to 100%.  AFAICT I am seeing exactly what I
> was seeing before.
>   

Argh.

Can you try the following experiments (independently):

- run kvm with the '-no-rtc' option

- disable compiz (or beryl).  I think the menu item is called 'Desktop
Effects'.

-- 
Do not meddle in the internals of kernels, for they are subtle and quick to panic.


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Tue Apr 24 2007, Avi Kivity <avi-AT-qumranet.com> wrote:

> Can you try the following experiments (independently):
>
> - run kvm with the '-no-rtc' option

Makes no noticeable difference that I can see.

> - disable compiz (or beryl).  I think the menu item is called 'Desktop
> Effects'.

It wasn't enabled.  I did install it a while back, but it's been
disabled ever since I discovered that fglrx doesn't support AIGLX and
is mostly crippled by enabling Xgl.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

Re: kvm somehow exempt from frequency scaling?

[Avi Kivity]

David Abrahams wrote:
> on Tue Apr 24 2007, Avi Kivity <avi-AT-qumranet.com> wrote:
>
>   
>> Can you try the following experiments (independently):
>>
>> - run kvm with the '-no-rtc' option
>>     
>
> Makes no noticeable difference that I can see.
>
>   
>> - disable compiz (or beryl).  I think the menu item is called 'Desktop
>> Effects'.
>>     
>
> It wasn't enabled.  I did install it a while back, but it's been
> disabled ever since I discovered that fglrx doesn't support AIGLX and
> is mostly crippled by enabling Xgl.
>
>   

I'm stumped.  It's probably some wierd X thing, but I can't imagine what.


-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

Re: kvm somehow exempt from frequency scaling?

[David Abrahams]

on Wed Apr 25 2007, Avi Kivity <avi-AT-qumranet.com> wrote:

> David Abrahams wrote:
>> on Tue Apr 24 2007, Avi Kivity <avi-AT-qumranet.com> wrote:
>>
>>   
>>> Can you try the following experiments (independently):
>>>
>>> - run kvm with the '-no-rtc' option
>>>     
>>
>> Makes no noticeable difference that I can see.
>>
>>   
>>> - disable compiz (or beryl).  I think the menu item is called 'Desktop
>>> Effects'.
>>>     
>>
>> It wasn't enabled.  I did install it a while back, but it's been
>> disabled ever since I discovered that fglrx doesn't support AIGLX and
>> is mostly crippled by enabling Xgl.
>>
>>   
>
> I'm stumped.  It's probably some wierd X thing, but I can't imagine what.

FWIW, a gnome-terminal window running 

  while echo 1 ; do echo foo ; done

still gets cycles when not on the visible workspace.

Do let me know if you want me to try any other experiments.

-- 
Dave Abrahams
Boost Consulting
http://www.boost-consulting.com

Don't Miss BoostCon 2007! ==> http://www.boostcon.com

-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/

Re: [kvm-devel] kvm somehow exempt from frequency scaling?

[Avi Kivity]

[Resurrecting old thread]

David Abrahams wrote:
> on Tue Apr 10 2007, Michael Riepe <michael-0QoEqw4nQxo-AT-public.gmane.org> wrote:
>
>   
>> Hi!
>>
>> Avi Kivity wrote:
>>
>>     
>>> Sounds like ondemand is broken.
>>>       
>> Maybe it only switches when the load rises in user space. With KVM, it
>> appears in system space. Got to cross-check that on my Core Duo.
>>     
>
> More data: it definitely seems to make a difference whether the VM's
> desktop is showing.  If I switch Gnome workspaces so it's not visible,
> it doesn't seem to drive the frequency up.
>
>   

This just happened to me.  I noticed that if I switch the workspace 
using alt-ctrl-left while the qemu window is active, qemu hangs.  On the 
other hand, if I switch using the mouse, or while some other window is 
active, everything is fine.  The qemu code even explains this as an sdl bug.

Jingle any bells?


-- 
error compiling committee.c: too many arguments to function

Re: [kvm-devel] kvm somehow exempt from frequency scaling?

[David Abrahams]

on Sun Jul 06 2008, Avi Kivity <avi-AT-qumranet.com> wrote:

> [Resurrecting old thread]
>
> David Abrahams wrote:
>> on Tue Apr 10 2007, Michael Riepe <michael-0QoEqw4nQxo-AT-public.gmane.org> wrote:
>>
>>   
>>> Hi!
>>>
>>> Avi Kivity wrote:
>>>
>>>     
>>>> Sounds like ondemand is broken.
>>>>       
>>> Maybe it only switches when the load rises in user space. With KVM, it
>>> appears in system space. Got to cross-check that on my Core Duo.
>>>     
>>
>> More data: it definitely seems to make a difference whether the VM's
>> desktop is showing.  If I switch Gnome workspaces so it's not visible,
>> it doesn't seem to drive the frequency up.
>>
>>   
>
> This just happened to me.  I noticed that if I switch the workspace
> using alt-ctrl-left while the qemu window is active, qemu hangs.  On
> the other hand, if I switch using the mouse, or while some other
> window is active, everything is fine.  The qemu code even explains
> this as an sdl bug.
>
> Jingle any bells?

I'm afraid not, sorry.  I stopped noticing the effect since I upgraded
to Ubuntu Hardy, although I admit I've stopped looking for it.

-- 
Dave Abrahams
BoostPro Computing
http://www.boostpro.com