[Qemu-devel] The QEMU Accelerator Module

[Qemu-devel] The QEMU Accelerator Module

[Fabrice Bellard]

Hi,

I just commited the first alpha release of the QEMU Accelerator Module 
(aka KQEMU) in the CVS. It gives better performance for the "x86 on x86" 
case by running most of the application code as is. It works only for a 
Linux x86 host running a 2.4 or 2.6 kernel. Linux 2.4 and 2.6 kernels 
and Windows 2000 have been runnning as guest OSes, but other OSes may 
work as well. As with every alpha kernel driver testing, it is better to 
backup your data before trying it.

KQEMU is _not_ open source as the rest of QEMU. It is a proprietary 
kernel module (read the LICENSE file) and will stay so until a gentle 
company decides to subsidy the QEMU project.

KQEMU usage is optional: you can disable it at compilation or run time, 
so no one is forced to use it.

Fabrice.

Re: [Qemu-devel] The QEMU Accelerator Module

[Grzegorz Kulewski]

Hi,

On Fri, 11 Feb 2005, Fabrice Bellard wrote:
> I just commited the first alpha release of the QEMU Accelerator Module (aka 
> KQEMU) in the CVS. It gives better performance for the "x86 on x86" case by 
> running most of the application code as is. It works only for a Linux x86 
> host running a 2.4 or 2.6 kernel. Linux 2.4 and 2.6 kernels and Windows 2000 
> have been runnning as guest OSes, but other OSes may work as well. As with 
> every alpha kernel driver testing, it is better to backup your data before 
> trying it.
>
> KQEMU is _not_ open source as the rest of QEMU. It is a proprietary kernel 
> module (read the LICENSE file) and will stay so until a gentle company 
> decides to subsidy the QEMU project.

1. How much? Can the money be collected by comunity (like blender)?
2. What if you will quit making qemu?
3. Are you planning to make and support it completly alone?
4. [Technical question.] Does savannah allow for not open source content 
in their CVS?


Thanks for your work,

Grzegorz Kulewski

Re: [Qemu-devel] The QEMU Accelerator Module

[Lennert Buytenhek]

On Fri, Feb 11, 2005 at 12:57:21AM +0100, Grzegorz Kulewski wrote:

> >I just commited the first alpha release of the QEMU Accelerator Module 
> >(aka KQEMU) in the CVS. It gives better performance for the "x86 on x86" 
> >case by running most of the application code as is. It works only for a 
> >Linux x86 host running a 2.4 or 2.6 kernel. Linux 2.4 and 2.6 kernels and 
> >Windows 2000 have been runnning as guest OSes, but other OSes may work as 
> >well. As with every alpha kernel driver testing, it is better to backup 
> >your data before trying it.
> >
> >KQEMU is _not_ open source as the rest of QEMU. It is a proprietary kernel 
> >module (read the LICENSE file) and will stay so until a gentle company 
> >decides to subsidy the QEMU project.
> 
> 1. How much? Can the money be collected by comunity (like blender)?

I'm interested in this as well.  How much would you need?


> 4. [Technical question.] Does savannah allow for not open source content 
> in their CVS?

I don't think so.

	This web site (called Savannah) is a central point for
	development, distribution and maintenance of Free Software
	that runs on free operating systems.


> Thanks for your work,

Same for me here.


--L

Re: [Qemu-devel] The QEMU Accelerator Module

[Tim]

> 1. How much? Can the money be collected by comunity (like blender)?

I would also contribute, if this were an option to keep the source
completely free (as in freedom).

tim

Re: [Qemu-devel] The QEMU Accelerator Module

[Fabrice Bellard]

Grzegorz Kulewski wrote:

> 4. [Technical question.] Does savannah allow for not open source content 
> in their CVS?

Sorry I made a mistake here. I am removing the module from the CVS and I 
am putting in the download area of the QEMU web site.

Fabrice.

Re: [Qemu-devel] The QEMU Accelerator Module

[Fabrice Bellard]

Hi,

I just removed the QEMU Accelerator Module (aka KQEMU) from the QEMU CVS 
repository. You can now download it from the QEMU web site 
(http://bellard.org/qemu). QEMU and KQEMU are two different projects 
with different licenses - I hope that it is clearer now.

Fabrice.

Re: [Qemu-devel] The QEMU Accelerator Module

[Darryl Dixon]

[-- Attachment #1: Type: text/plain, Size: 1700 bytes --]

Hi Fabrice!

    Congratulations on the release of this very important code, it has
been much anticipated!  :)  A quick question: if the module does not
allow for redistribution, how does this affect the nightly source
tarballs that I am used to grabbing from dad-answers - will these not be
able to contain this module?  CVS access is very difficult for some (me
in particular :) because of fascist firewalls :(

    Also, can I add my $0.02c and say that I too would be happy to
donate a la blender to seeing this module put under the GPL.

Many regards, and thank you for your fantastic work
Darryl Dixon

On Fri, 2005-02-11 at 00:31 +0100, Fabrice Bellard wrote:

> Hi,
> 
> I just commited the first alpha release of the QEMU Accelerator Module 
> (aka KQEMU) in the CVS. It gives better performance for the "x86 on x86" 
> case by running most of the application code as is. It works only for a 
> Linux x86 host running a 2.4 or 2.6 kernel. Linux 2.4 and 2.6 kernels 
> and Windows 2000 have been runnning as guest OSes, but other OSes may 
> work as well. As with every alpha kernel driver testing, it is better to 
> backup your data before trying it.
> 
> KQEMU is _not_ open source as the rest of QEMU. It is a proprietary 
> kernel module (read the LICENSE file) and will stay so until a gentle 
> company decides to subsidy the QEMU project.
> 
> KQEMU usage is optional: you can disable it at compilation or run time, 
> so no one is forced to use it.
> 
> Fabrice.
> 
> 
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel

-- 
Darryl Dixon <esrever_otua@pythonhacker.is-a-geek.net>

[-- Attachment #2: Type: text/html, Size: 2805 bytes --]

Re: [Qemu-devel] The QEMU Accelerator Module

[Hetz Ben Hamo]

Hi,

The snapshots in my web site (dad-answers.com) are grabbed nightly
from the main CVS server at savannah.

In case the code will be needed to remove due to some conflictions in
licensing with the Savannah team, then I'll make sure that the kernel
code will be available at my web site as well.

Thanks,
Hetz


On Fri, 11 Feb 2005 13:38:40 +1300, Darryl Dixon
<esrever_otua@pythonhacker.is-a-geek.net> wrote:
>  Hi Fabrice!
>  
>      Congratulations on the release of this very important code, it has been
> much anticipated!  :)  A quick question: if the module does not allow for
> redistribution, how does this affect the nightly source tarballs that I am
> used to grabbing from dad-answers - will these not be able to contain this
> module?  CVS access is very difficult for some (me in particular :) because
> of fascist firewalls :(
>  
>      Also, can I add my $0.02c and say that I too would be happy to donate a
> la blender to seeing this module put under the GPL.
>  
>  Many regards, and thank you for your fantastic work
>  Darryl Dixon
> 
>  
>  On Fri, 2005-02-11 at 00:31 +0100, Fabrice Bellard wrote: 
>  Hi, I just commited the first alpha release of the QEMU Accelerator Module
> (aka KQEMU) in the CVS. It gives better performance for the "x86 on x86"
> case by running most of the application code as is. It works only for a
> Linux x86 host running a 2.4 or 2.6 kernel. Linux 2.4 and 2.6 kernels and
> Windows 2000 have been runnning as guest OSes, but other OSes may work as
> well. As with every alpha kernel driver testing, it is better to backup your
> data before trying it. KQEMU is _not_ open source as the rest of QEMU. It is
> a proprietary kernel module (read the LICENSE file) and will stay so until a
> gentle company decides to subsidy the QEMU project. KQEMU usage is optional:
> you can disable it at compilation or run time, so no one is forced to use
> it. Fabrice. _______________________________________________ Qemu-devel
> mailing list Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel 
>  -- 
>  Darryl Dixon <esrever_otua@pythonhacker.is-a-geek.net> 
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel
> 
> 
>

Re: [Qemu-devel] The QEMU Accelerator Module

[Leigh Dyer]

Fabrice Bellard wrote:
> Hi,
> 
> KQEMU is _not_ open source as the rest of QEMU. It is a proprietary 
> kernel module (read the LICENSE file) and will stay so until a gentle 
> company decides to subsidy the QEMU project.

Thanks, Fabrice, for your amazing work with QEMU. I can certainly 
understand why you've made KQEMU closed for the moment, and I too would 
probably be willing to contribute to a bounty on opening it.

In the meantime, what are the chances of getting KQEMU running on an 
AMD64 kernel? I'm happy to run 32-bit binaries on my 64-bit system, but 
I don't relish the thought of having to use a 32-bit kernel to run 
accelerated QEMU.

Thanks
Leigh

Re: [Qemu-devel] The QEMU Accelerator Module

[James Mastros]

Fabrice Bellard wrote:
> KQEMU usage is optional: you can disable it at compilation or run time, 
> so no one is forced to use it.
Would it be feasible to isolate the code that is GPL'd vs non-GPL'd, and 
clearly mark what files are non-GPL'd, preferably by putting them in a 
different CVS module, or sepperate subdirectory of the present CVS 
module, and a sepperate tarball in the next released version?

Doing that would make life /vastly/ easier for distributions, OS 
mirrors, and other such people, as well as people, like me, who want not 
to be tainted.

I notice that there already is a kqemu directory extant -- does it 
include all the non-GPL code?

	-=- James Mastros

Re: [Qemu-devel] The QEMU Accelerator Module

[Tom Marble]

[-- Attachment #1: Type: text/plain, Size: 2069 bytes --]

Fabrice Bellard wrote:

> KQEMU is _not_ open source as the rest of QEMU. It is a proprietary
> kernel module (read the LICENSE file) and will stay so until a gentle
> company decides to subsidy the QEMU project.

I'd like to make an appeal that KQEMU be licensed under the GPL.

I'm not going to try to roll out a general FOSS position statement, but
speak from a more practical level:
- This technology is extremely important for faciliting automated
  testing of the full (open) stack (including BIOS, ACPI, storage,
  graphics, OS, swsusp).  And that testing is vital to the continuing quality
  and vibrance of the entire stack.
- non-GPL kernel modules will taint the kernel and completely
  change the trajectory of the technology (it's support, innovation,
  evolution, etc.)
- Clearly Fabrice is (currently) in the comfortable position of
  being the Benevolent Dictator ( http://en.wikipedia.org/wiki/Benevolent_dictator ).
  I'm sure that's a position that could lead to speaking engagements,
  books, consulting and other forms of monetization which
  are quite compatible with the GPL.

Is KQEMU valuable?  Absolutely!  Could it be proprietary?  Of course
(vis VMware).  In this strange calculus of open source, as counterintuitive
as it seems, I'm convinced that it can be even *more* valuable as
GPL technology (to the community) and more valuable to Fabrice
personally.  While there may be many shades of 'open' (e.g. blender)
I'm sure that that community would be much more productive
with a GPL license and consequently would provide much more
enthusiastic support of whatever Fabrice needs.

I'm sure everyone would be willing to help Googlebomb and /.
Qemu, right?  (Or better yet, help make KQEMU *even better*
through many eyes so that it can really reach a new level of success).

Sincerely,

--Tom

P.S. Some press already
  http://slashdot.org/article.pl?sid=04/11/10/1320231&tid=201&tid=190&tid=1
  http://yro.slashdot.org/article.pl?sid=04/10/21/176235&tid=93&tid=158
  http://en.wikipedia.org/wiki/Qemu
  http://freshmeat.net/projects/qemu/

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 256 bytes --]

Re: [Qemu-devel] The QEMU Accelerator Module

[Karel Gardas]

Tom,

On Thu, 10 Feb 2005, Tom Marble wrote:

> Fabrice Bellard wrote:
>
> > KQEMU is _not_ open source as the rest of QEMU. It is a proprietary
> > kernel module (read the LICENSE file) and will stay so until a gentle
> > company decides to subsidy the QEMU project.
>
> I'd like to make an appeal that KQEMU be licensed under the GPL.
>
> I'm not going to try to roll out a general FOSS position statement, but
> speak from a more practical level:
> - This technology is extremely important for faciliting automated
>   testing of the full (open) stack (including BIOS, ACPI, storage,
>   graphics, OS, swsusp).  And that testing is vital to the continuing quality
>   and vibrance of the entire stack.
> - non-GPL kernel modules will taint the kernel and completely
>   change the trajectory of the technology (it's support, innovation,
>   evolution, etc.)
> - Clearly Fabrice is (currently) in the comfortable position of
>   being the Benevolent Dictator ( http://en.wikipedia.org/wiki/Benevolent_dictator ).
>   I'm sure that's a position that could lead to speaking engagements,
>   books, consulting and other forms of monetization which
>   are quite compatible with the GPL.
>
> Is KQEMU valuable?  Absolutely!  Could it be proprietary?  Of course
> (vis VMware).  In this strange calculus of open source, as counterintuitive
> as it seems, I'm convinced that it can be even *more* valuable as
> GPL technology (to the community) and more valuable to Fabrice
> personally.  While there may be many shades of 'open' (e.g. blender)
> I'm sure that that community would be much more productive
> with a GPL license and consequently would provide much more
> enthusiastic support of whatever Fabrice needs.
>
> I'm sure everyone would be willing to help Googlebomb and /.
> Qemu, right?  (Or better yet, help make KQEMU *even better*
> through many eyes so that it can really reach a new level of success).

I think that you makes your appeal on the wrong side. Please read
Fabrice's note above. You should rather contact your Linux distribution
vendor and try to convience it to sponsor kqemu development and switch to
GPL. You can also contact some other linux friendly companies like IBM/HP
to do the same.

I'm afraid you are not right with your preassure to how much this
technology is important. No, it is not important for testing of
BIOS/ACPI/kernel etc. It's just very important for end user because of
possible (as I hope) performance benefit.

Please consider how Fabrice is really nice to Qemu community. He makes the
core of kqemu free (as a beer) and just limits its _redistribution_. I can
imagine, he also could make it completely proprietary and force you to
spend $$$ for its download. So if I understand this message right, it is
just to Linux commercial distributors: `Would you like to distribute
_fast_ Qemu? OK, so please pay for its development'. Please consider that
the same these vendors do for many other applications, even core like
kernel, gcc, binutils, wine/crossover (which is even wholy proprietary)
and such. Linux vendors seems to pay for every important technology which
makes them more competitive in linux distributions market, so why also
don't pay for Qemu a bit?

Once again I would like to thank Fabrice for making kqemu enough free that
Qemu user community can benefit from it and provide some bugreports, which
hopefully will be also benefit to kqemu itself. I hope that members of
Qemu community will understand this Fabrice's step and will support him in
finding appropriate company willing to pay for (K)Qemu's development.

Cheers,
Karel
--
Karel Gardas                  kgardas@objectsecurity.com
ObjectSecurity Ltd.           http://www.objectsecurity.com

Re: [Qemu-devel] The QEMU Accelerator Module

[Lionel Ulmer]

On Fri, Feb 11, 2005 at 10:23:59AM +0100, Karel Gardas wrote:
> (...) Please consider that
> the same these vendors do for many other applications, even core like
> kernel, gcc, binutils, wine/crossover (which is even wholy proprietary)
> and such.

I do not see what Wine (fully LGPL'ed and without any - that I know of -
Linux distribution sponsors) and CrossOver (which is not wholly proprietary
as 95 % of the code is pure LGPL'ed Wine code and that, except for some
packagings deals, is only paid for by CodeWeavers itself) does in this list
:-)

    Lionel (who can't wait to go back home to test QEMU / KQEMU :-) )

-- 
		 Lionel Ulmer - http://www.bbrox.org/

Re: [Qemu-devel] The QEMU Accelerator Module

[Karel Gardas]

On Fri, 11 Feb 2005, Lionel Ulmer wrote:

> On Fri, Feb 11, 2005 at 10:23:59AM +0100, Karel Gardas wrote:
> > (...) Please consider that
> > the same these vendors do for many other applications, even core like
> > kernel, gcc, binutils, wine/crossover (which is even wholy proprietary)
> > and such.
>
> I do not see what Wine (fully LGPL'ed and without any - that I know of -
> Linux distribution sponsors) and CrossOver (which is not wholly proprietary
> as 95 % of the code is pure LGPL'ed Wine code and that, except for some
> packagings deals, is only paid for by CodeWeavers itself) does in this list
> :-)

It was my impression that if you like to have SuSE distro with CrossOver,
you have to buy some kind of SuSE product/cd/dvd instead of just do plain
ftp install. From this I've wrongly assumed that CrossOver is proprietary,
sorry for that mistake and thanks for the clarification.

Anyway, if it is (L)GPL, then even better, since then you are in the same
situation like with kernel/gcc/binutils which are also sponsored by
someone/someparty and which approach probably Fabrice would like to apply
to Qemu too. :-)

Thanks,
Karel
--
Karel Gardas                  kgardas@objectsecurity.com
ObjectSecurity Ltd.           http://www.objectsecurity.com

Re: [Qemu-devel] The QEMU Accelerator Module

[Lionel Ulmer]

On Fri, Feb 11, 2005 at 01:05:12PM +0100, Karel Gardas wrote:
> It was my impression that if you like to have SuSE distro with CrossOver,
> you have to buy some kind of SuSE product/cd/dvd instead of just do plain
> ftp install. From this I've wrongly assumed that CrossOver is proprietary,
> sorry for that mistake and thanks for the clarification.

Well, this is the 'packaging' deal I was talking about (some SuSE
distributions came with CrossOver installed natively).

As for the proprietariness (?) of CrossOver, the 'core' of it (i.e. Wine) is
a modified version of Wine (so you can get the source for it as it is
LGPL'ed) but all the 'glue' around it (configuration file creation, pretty
UI to install application, desktop integration, ...) *is* closed source.

I was just reacting to your 'wholly' in your sentence as it's only 'partly'
closed source (a bit like QMEU actually :-) ).

And as they actually sell CrossOver (plus support / porting contracts) I did
not see this as 'sponsoring' (although for a Wine user / developper like me,
it looks like it as a LOT of patches to the Wine tree come from their
employees).

     Lionel

-- 
		 Lionel Ulmer - http://www.bbrox.org/

Re: [Qemu-devel] The QEMU Accelerator Module

[Grzegorz Kulewski]

Hi all,

On Fri, 11 Feb 2005, Karel Gardas wrote:
> On Thu, 10 Feb 2005, Tom Marble wrote:
>> Fabrice Bellard wrote:
>>> KQEMU is _not_ open source as the rest of QEMU. It is a proprietary
>>> kernel module (read the LICENSE file) and will stay so until a gentle
>>> company decides to subsidy the QEMU project.
>>
>> I'd like to make an appeal that KQEMU be licensed under the GPL.

I hope it will happen soon.


>> I'm not going to try to roll out a general FOSS position statement, but
>> speak from a more practical level:
>> - This technology is extremely important for faciliting automated
>>   testing of the full (open) stack (including BIOS, ACPI, storage,
>>   graphics, OS, swsusp).  And that testing is vital to the continuing quality
>>   and vibrance of the entire stack.

Ok, but it does not stop you from testing. Look at the binary-only part 
size and its interface in .c and .h files. If I understand it correclty 
Fabrice (as usual) done things in extremely right way. The proprietary 
module is _only_ the accelerator. All emulation is done in userspace. Both 
with and without kernel module you get the same qemu (modulo 
userspace-only CPU implementation bugs on unpriviledged instructions if 
there are any left). Only some virtualization is done in kernel. So every 
IO device is exactly the same in both versions of qemu.


>> - non-GPL kernel modules will taint the kernel and completely
>>   change the trajectory of the technology (it's support, innovation,
>>   evolution, etc.)

I think that you are not 100% right. I think that adding non-GPL module to 
the kernel is the (nearly only) problem of this aproach because you will 
loss support from LKML and Linux Community. You will need to reproduce all 
problems on vanilla kernel. (It is not clear if even with GPL licence 
kernel developers will want to support kqemu enabled kernels but the 
module is so small and probably so unintrusive that maybe in could be 
merged into -mm very fast or at least be considered as not-the-cause of 
your issues.)

But saying that it will "change the trajectory of the technology" is I 
think a little too much. Ok, you will not be able to hack on the module 
itself but (if I understand it correctly) Fabrice did everything to place 
as little as possible amount of code in it. After several trivial bugfixes 
of its child age it will probably be left untouched. So you can change 
nearly everything in qemu if you like. It is not 100% freedom but say 
99.999%... :-)

Besides, maybe I am not right, but I do not think that more than 1 or 2 
developers of qemu will want to hack on its kernel module. I do not even 
think it will be good. It should remain as small and simple as possible.


>> Is KQEMU valuable?  Absolutely!  Could it be proprietary?  Of course
>> (vis VMware).

There is _big_ difference. In VMware you can change nothing. Not even add 
new device.


>> In this strange calculus of open source, as counterintuitive
>> as it seems, I'm convinced that it can be even *more* valuable as
>> GPL technology (to the community) and more valuable to Fabrice
>> personally.  While there may be many shades of 'open' (e.g. blender)
>> I'm sure that that community would be much more productive
>> with a GPL license and consequently would provide much more
>> enthusiastic support of whatever Fabrice needs.

But I do not see any milionaire on this list? I am wrong?


>> I'm sure everyone would be willing to help Googlebomb and /.
>> Qemu, right?  (Or better yet, help make KQEMU *even better*
>> through many eyes so that it can really reach a new level of success).
>
> I think that you makes your appeal on the wrong side. Please read
> Fabrice's note above. You should rather contact your Linux distribution
> vendor and try to convience it to sponsor kqemu development and switch to
> GPL. You can also contact some other linux friendly companies like IBM/HP
> to do the same.

Exactly. Question if Fabrice will want gigant to support qemu or rather 
several (possibly) smaller companies? (IIRC author of plex86 was fired 
from Mandrake because of some reduction so maybe several companies would 
be better?)


> I'm afraid you are not right with your preassure to how much this
> technology is important. No, it is not important for testing of
> BIOS/ACPI/kernel etc. It's just very important for end user because of
> possible (as I hope) performance benefit.

And BIOS and ACPI and other IO and even priviledged instructions are done 
in userspace in open source code (or I am completly wrong...)

And if you are really briliant programmer you can write that 36k of code 
yourself. The only thing that stops me from (trying) doing so is that 
it would be not fair to Fabrice. Maybe I am too self-confident but it 
really should not be that hard to do if you know kernel programming a 
little (and assembler of course) or have somebody to anwser your 
questions. This is maybe good exercise to everybody: write your own and 
sent it to Fabrice... ;-)  Of course any form of reverse engineering is 
considered cheating and does not produce the main good: learning how to 
make exciting things in kernelspace.


> Once again I would like to thank Fabrice for making kqemu enough free that
> Qemu user community can benefit from it and provide some bugreports, which
> hopefully will be also benefit to kqemu itself. I hope that members of
> Qemu community will understand this Fabrice's step and will support him in
> finding appropriate company willing to pay for (K)Qemu's development.

Me too. :-)


Thanks Fabrice.

Grzegorz Kulewski

[Qemu-devel] KQemu logic and marketing goals

[Jean-Michel POURE]

Le Vendredi 11 Février 2005 06:15, Tom Marble a écrit :
> I'd like to make an appeal that KQEMU be licensed under the GPL.

Dear Tom,

I agree with you. As far as I am concerned, I thought Qemu could be used to 
ease the migration of MS Windows and other OSes to GNU/Linux.

I don't understand the logic behind KQemu.
Fabrice, can you tell us more about your marketing projects ? 

Kind regards,
Jean-Michel Pouré

Re: [Qemu-devel] KQemu logic and marketing goals

[Lionel Ulmer]

On Fri, Feb 11, 2005 at 10:54:31AM +0100, Jean-Michel POURE wrote:
> I agree with you. As far as I am concerned, I thought Qemu could be used to 
> ease the migration of MS Windows and other OSes to GNU/Linux.

And what would the fact of the 'closeness' of the KQEMU module hurt this
goal ?

> I don't understand the logic behind KQemu.
> Fabrice, can you tell us more about your marketing projects ?

Well, maybe the simple logic of being able to pay one's bills with the work
one produces ?

And you will always have the fully LGPL'ed / BSD licensed 'normal' QEMU to
use if, for political / practical /technical reasons, you do not wish to
taint your kernel.

           Lionel

-- 
		 Lionel Ulmer - http://www.bbrox.org/

Re: [Qemu-devel] KQemu logic and marketing goals

[Julian Seward]

On Friday 11 February 2005 09:54, Jean-Michel POURE wrote:

> I don't understand the logic behind KQemu.
> Fabrice, can you tell us more about your marketing projects ?

I think it's clear -- qemu is a lot of work and after a
while, doing a day job and running an open-source project on 
the side is just impossible, if you want to have any kind of
a life too.  I suspect Fabrice would like to be able to work
on Qemu full time, and so it has to generate money somehow.
This seems eminently reasonable to me.

A possible alternative is to convince some open-source company
to hire Fabrice and have him work full-time on qemu.

Just my E 0.02,

J

Re: [Qemu-devel] KQemu logic and marketing goals

[Lennert Buytenhek]

On Fri, Feb 11, 2005 at 12:27:14PM +0000, Julian Seward wrote:

> > I don't understand the logic behind KQemu.
> > Fabrice, can you tell us more about your marketing projects ?
> 
> I think it's clear -- qemu is a lot of work and after a
> while, doing a day job and running an open-source project on 
> the side is just impossible, if you want to have any kind of
> a life too.  I suspect Fabrice would like to be able to work
> on Qemu full time, and so it has to generate money somehow.
> This seems eminently reasonable to me.
> 
> A possible alternative is to convince some open-source company
> to hire Fabrice and have him work full-time on qemu.

Or convince some university that his work is academically interesting
and make them hire him as a PhD student or something.  I'm sure one
could write an interesting dissertation on dynamic translation.  (The
pay will naturally not be all that great though.)


--L

Re: [Qemu-devel] KQemu logic and marketing goals

[Johannes Schindelin]

Hi,

On Fri, 11 Feb 2005, Lennert Buytenhek wrote:

> Or convince some university that his work is academically interesting
> and make them hire him as a PhD student or something.

Believe me, QEmu is worth much more than a PhD salary.

Ciao,
Dscho

Re: [Qemu-devel] KQemu logic and marketing goals

[Lennert Buytenhek]

On Fri, Feb 11, 2005 at 02:17:04PM +0100, Johannes Schindelin wrote:

> > Or convince some university that his work is academically interesting
> > and make them hire him as a PhD student or something.
> 
> Believe me, QEmu is worth much more than a PhD salary.

Sure, I'll agree with that, but it all depends on what your goals in
life are.

I walked away from a well-paid, high-responsibility position myself
recently, and will likely go back to the academic world, just for the
reason that I wasn't happy living the corporate lifestyle.  A number
of my friends did the exact same thing, now earn one fourth of what
they used to, but generally feel much happier.


--L

Re: [Qemu-devel] KQemu logic and marketing goals

[Jean-Michel POURE]

Le Vendredi 11 Février 2005 13:27, Julian Seward a écrit :
>  I suspect Fabrice would like to be able to work
> on Qemu full time, and so it has to generate money somehow.
> This seems eminently reasonable to me.

I thought that Fabrice was doing reseach for the French Government, but I may 
be wrong. Maybe Fabrice can explain his marketing project a little more.

Kind regards,
Jean-Michel

Re: [Qemu-devel] The QEMU Accelerator Module

[Johannes Schindelin]

Hi,

On Thu, 10 Feb 2005, Tom Marble wrote:

> I'd like to make an appeal that KQEMU be licensed under the GPL.

Me, too. But not without compensation for the wonderful work of Fabrice!
It is just plain wrong to ask anybody to work for free. I personally know
programmers who get paid a lot of money, but their work is just a POC.

Fabrice, to the contrary, came up with a very usable *and* elegant
program. Everybody who benefitted from his work should be content with
what license he chose IMHO (the "Put up or shut up" principle).

Having said that, I think that it might be a good idea if the readers of
this list decide on a target company, and hit them with polite requests.

Target companies I suggest:

	- IBM (they should be very much interested in using QEmu on their
		new Cell processors; imagine that: a processor which is
		10x faster than Pentium natively, with a great emulator
		like QEmu => still faster than Pentium!)
	- Novell (just because they do great work, and are great sponsors)
	- RedHat (just because they are the best known Linux company)

Of course, sponsorship always has its downsides, too, but I think that
Fabrice has well earned some big bucks for this great project!

Just them noo 2 Eurocents of yours truly,
Dscho

Re: [Qemu-devel] The QEMU Accelerator Module

[Jens Arm]

> Target companies I suggest:
> 
> 	- IBM (they should be very much interested in using QEmu on their
> 		new Cell processors; imagine that: a processor which is
> 		10x faster than Pentium natively, with a great emulator
> 		like QEmu => still faster than Pentium!)
> 	- Novell (just because they do great work, and are great sponsors)
> 	- RedHat (just because they are the best known Linux company)

or MDK (Mandrake) they sponsored that bochs get under GPL :)

Jens

Re: [Qemu-devel] The QEMU Accelerator Module

[Derek Fawcus]

Well,
  rather than whinging because Fabrice has not chosen to distribute his
work under your preferred licence.  People could simply reimplement it.

Mind - this involves effort on their behalf,  and some thinking.  So it's
certainly easier to moan.  However I suggest that such moaning is simply
a waste of time and effort.

Now I don't know how Fabrice has done the kqemu module,  but the obvious
approach that springs to mind is simply moving the qemu-fast processing
into the kernel with checks for the address boundary.  So if I was to
attempt to reimplement it,  my starting point would be to approach it
in that fashion.

Namely placing a version of cpu_exec() and/or main_loop() into the kernel
together with the use of the USE_CODE_COPY facility and some bounds checks
such that if the machine being emulated attempted to have accessable memory
above 0xc0000000 it would fall back to the user-space SOFT_MMU emulation.
One could then manipulate the process space such that while the kernel
module was running user space code,  it's process address space (< 0xc0000000)
reflected the emulated machine space.

However,  I've got other things to do,  so the above is not a priority,
and I'm quite happy to use Fabrice's module.  Mind - I need to update
to a more recent version of the code,  my current work is in a version
dating back to ~ September / October last year.

DF

Re: [Qemu-devel] The QEMU Accelerator Module

[Jim C. Brown]

On Fri, Feb 11, 2005 at 04:09:10PM +0000, Derek Fawcus wrote:
> Well,
>   rather than whinging because Fabrice has not chosen to distribute his
> work under your preferred licence.  People could simply reimplement it.

Or do as I am, and figure out how to mix plex86 with the open source part.

Or even better, dig deep in your pockets and sponser Fabrice to completely
open it up.

> 
> Mind - this involves effort on their behalf,  and some thinking.  So it's
> certainly easier to moan.  However I suggest that such moaning is simply
> a waste of time and effort.
> 
> Now I don't know how Fabrice has done the kqemu module,  but the obvious
> approach that springs to mind is simply moving the qemu-fast processing
> into the kernel with checks for the address boundary.  So if I was to
> attempt to reimplement it,  my starting point would be to approach it
> in that fashion.
> 
> Namely placing a version of cpu_exec() and/or main_loop() into the kernel
> together with the use of the USE_CODE_COPY facility and some bounds checks
> such that if the machine being emulated attempted to have accessable memory
> above 0xc0000000 it would fall back to the user-space SOFT_MMU emulation.
> One could then manipulate the process space such that while the kernel
> module was running user space code,  it's process address space (< 0xc0000000)
> reflected the emulated machine space.

That wouldn't be complete, as kqemu uses virtalization as well. Also, it seems
that a lot of the kernel MMU support is in the open source code. Only
kqemu_delete, kqemu_exec, kqemu_get_cpu_state, and kqemu_init need to be
reimplemented.

-- 
Infinite complexity begets infinite beauty.
Infinite precision begets infinite perfection.

[Qemu-devel] Re: The QEMU Accelerator Module

[Anand Kumria]

On Fri, 11 Feb 2005 00:31:03 +0100, Fabrice Bellard wrote:

> Hi,
> 
> I just commited the first alpha release of the QEMU Accelerator Module
> (aka KQEMU) in the CVS. It gives better performance for the "x86 on x86"
> case by running most of the application code as is. It works only for a
> Linux x86 host running a 2.4 or 2.6 kernel. Linux 2.4 and 2.6 kernels and
> Windows 2000 have been runnning as guest OSes, but other OSes may work as
> well. As with every alpha kernel driver testing, it is better to backup
> your data before trying it.
> 
> KQEMU is _not_ open source as the rest of QEMU. It is a proprietary kernel
> module (read the LICENSE file) and will stay so until a gentle company
> decides to subsidy the QEMU project.

Is it really fair to distribute a proprietary module from a free software
source code repository?

Anand

Re: [Qemu-devel] Re: The QEMU Accelerator Module

[Antti-Juhani Kaijanaho]

On 20050212T031653+1100, Anand Kumria wrote:
> Is it really fair to distribute a proprietary module from a free software
> source code repository?

At least it's against the Savannah rules.  They have enforced this
particular rule in the past.

-- 
Antti-Juhani Kaijanaho                  http://antti-juhani.kaijanaho.info/

		Blogi - http://kaijanaho.info/antti-juhani/blog/
                 Toys - http://www.cc.jyu.fi/yhd/toys/

Re: [Qemu-devel] Re: The QEMU Accelerator Module

[Johannes Schindelin]

Hi,

On Sat, 12 Feb 2005, Anand Kumria wrote:

> Is it really fair to distribute a proprietary module from a free software
> source code repository?

Is it really fair to expect others to work for nothing at all?

Hth,
Dscho

Re: [Qemu-devel] Re: The QEMU Accelerator Module

[Darryl Dixon]

[-- Attachment #1: Type: text/plain, Size: 1634 bytes --]

Please Johannes,

   No-one has suggested that Fabrice should work for nothing; that was
not a fair comment at all.  Many people on this list (myself included)
have in fact offered finance to see the accelerator module GPL'ed (if
that is possible).  The question Anand asked is eminently reasonable and
for perfectly practical reasons it needs to be resolved one way or the
other (e.g. if the accel module must be hosted elsewhere, would it be
best to move the whole project so everything stays in one place?).  An
equally unfair counter-question would be 'is it fair to use freely
provided hosting that cost other people real money and break their TOS?'

    I for one am happy to see the existence of this module in any form
at all, but that doesn't mean that certain practical matters can be
ignored.  In fact, the Wine project just recently grappled with the same
issue with (IIRC) Microsoft's DCOM redistributable and their sourceforge
hosting.  In the end they now host it elsewhere.  Perhaps Fabrice could
get an exception from the Savannah team if they are asked?

Many regards,
Darryl Dixon

On Sat, 2005-02-12 at 03:01 +0100, Johannes Schindelin wrote:

> Hi,
> 
> On Sat, 12 Feb 2005, Anand Kumria wrote:
> 
> > Is it really fair to distribute a proprietary module from a free software
> > source code repository?
> 
> Is it really fair to expect others to work for nothing at all?
> 
> Hth,
> Dscho
> 
> 
> 
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel

-- 
Darryl Dixon <esrever_otua@pythonhacker.is-a-geek.net>

[-- Attachment #2: Type: text/html, Size: 2554 bytes --]

Re: [Qemu-devel] Re: The QEMU Accelerator Module

[Hetz Ben Hamo]

Hmm,

I don't mind hosting the kqemu module, whether as a CVS or as
snapshots (or both) ....

Thanks,
Hetz


On Sat, 12 Feb 2005 16:23:53 +1300, Darryl Dixon
<esrever_otua@pythonhacker.is-a-geek.net> wrote:
>  Please Johannes,
>  
>     No-one has suggested that Fabrice should work for nothing; that was not
> a fair comment at all.  Many people on this list (myself included) have in
> fact offered finance to see the accelerator module GPL'ed (if that is
> possible).  The question Anand asked is eminently reasonable and for
> perfectly practical reasons it needs to be resolved one way or the other
> (e.g. if the accel module must be hosted elsewhere, would it be best to move
> the whole project so everything stays in one place?).  An equally unfair
> counter-question would be 'is it fair to use freely provided hosting that
> cost other people real money and break their TOS?'
>  
>      I for one am happy to see the existence of this module in any form at
> all, but that doesn't mean that certain practical matters can be ignored. 
> In fact, the Wine project just recently grappled with the same issue with
> (IIRC) Microsoft's DCOM redistributable and their sourceforge hosting.  In
> the end they now host it elsewhere.  Perhaps Fabrice could get an exception
> from the Savannah team if they are asked?
>  
>  Many regards,
>  Darryl Dixon
> 
>  
>  On Sat, 2005-02-12 at 03:01 +0100, Johannes Schindelin wrote: 
>  Hi, On Sat, 12 Feb 2005, Anand Kumria wrote: > Is it really fair to
> distribute a proprietary module from a free software > source code
> repository? Is it really fair to expect others to work for nothing at all?
> Hth, Dscho _______________________________________________ Qemu-devel
> mailing list Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel 
>  -- 
>  Darryl Dixon <esrever_otua@pythonhacker.is-a-geek.net> 
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel
> 
> 
>

[Qemu-devel] Re: Re: The QEMU Accelerator Module

[Anand Kumria]

On Sat, 12 Feb 2005 03:01:53 +0100, Johannes Schindelin wrote:

> Hi,
> 
> On Sat, 12 Feb 2005, Anand Kumria wrote:
> 
>> Is it really fair to distribute a proprietary module from a free
>> software source code repository?
> 
> Is it really fair to expect others to work for nothing at all?
> 

Are you actually this stupid?

Or is it that English isn't you native language and you've missed the
subtle point that qemu is breaking the rules of savannah by distributing a
binary module?

I was trying to be polite and subtle about it so that Fabrice could
consider his options and make a considered cautious move. 

Anyway, because of your response, I've complained to the savannah
administrators.

No thanks to you.

Anand

Re: [Qemu-devel] Re: Re: The QEMU Accelerator Module

[Natalia Portillo]

Hi,

I think that this whole discussion has gone to a completely stupid 
fashion.

Fabrice is a great programmer and QEMU is a VERY great project.

KQEMU is a great idea that speed up things a lot.

Just think if KQEMU is released in BSD or LPGL license like rest of 
qemu...

They are right now a lot of "super emulators" being sold (like 
iEmulator) that use QEMU (and like the named one, without crediting 
it), and nor Fabrice nor other QEMU collaborators see money from that, 
like the WinZip program that is nothing but a commercial GUI for the 
Info-ZIP compression library.

Just imagine what will happen if someone takes KQEMU makes a dirty GUI 
and sells it. Just like the VMWare cost... Image how much could someone 
get from KQEMU just investing 5 minutes in a RAD enviroment, selling it 
and leaving Fabrice with NOTHING.

I think that the current Fabrice's movement is the correct one and I 
hope that a company will sponsor not only the KQEMU but the whole QEMU 
project as is the BEST and MOST COMPLETE non-old computer emulator, and 
also, its fast, and it's worth it.

There are a lot of companies interested in open source projects, like 
Novell, IBM, Apple, did anyone tried to contact them?

Regards,
Natalia Portillo

P.S.: Anand I think that your action of complaining Savannah could 
"hurt" (sorry I don't know the exact word) Fabrice and is based on 
another person's words, and is more a child's action than one from an 
adult. You should just have let Fabrice to see that he was breaking 
Savannah's rules.
P.S.2: Fabrice if you need a CVS server I offer my modest server.

These are my 2 eurocents ;)

El 12/02/2005, a las 6:14, Anand Kumria escribió:

> On Sat, 12 Feb 2005 03:01:53 +0100, Johannes Schindelin wrote:
>
>> Hi,
>>
>> On Sat, 12 Feb 2005, Anand Kumria wrote:
>>
>>> Is it really fair to distribute a proprietary module from a free
>>> software source code repository?
>>
>> Is it really fair to expect others to work for nothing at all?
>>
>
> Are you actually this stupid?
>
> Or is it that English isn't you native language and you've missed the
> subtle point that qemu is breaking the rules of savannah by 
> distributing a
> binary module?
>
> I was trying to be polite and subtle about it so that Fabrice could
> consider his options and make a considered cautious move.
>
> Anyway, because of your response, I've complained to the savannah
> administrators.
>
> No thanks to you.
>
> Anand
>
>
>
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>

Re: [Qemu-devel] Re: Re: The QEMU Accelerator Module

[Jean-Christian de Rivaz]

Natalia Portillo a écrit :

> Just imagine what will happen if someone takes KQEMU makes a dirty GUI 
> and sells it. Just like the VMWare cost... Image how much could someone 
> get from KQEMU just investing 5 minutes in a RAD enviroment, selling it 
> and leaving Fabrice with NOTHING.

VMWare GUI is not juste a dirty one, and will take far far more that 5 
minutes to be programmed. Now, kqemu or not, qemu will greatly benefit 
from a good GUI. So just do it...

-- 
Jean-Christian de Rivaz

Re: [Qemu-devel] Re: Re: The QEMU Accelerator Module

[René Korthaus]

Am 12.02.2005 um 17:19 schrieb Natalia Portillo:

> Hi,
>
> I think that this whole discussion has gone to a completely stupid 
> fashion.
>
> Fabrice is a great programmer and QEMU is a VERY great project.
>
> KQEMU is a great idea that speed up things a lot.
>
> Just think if KQEMU is released in BSD or LPGL license like rest of 
> qemu...
>
> They are right now a lot of "super emulators" being sold (like 
> iEmulator) that use QEMU (and like the named one, without crediting 
> it)

This comparison lacks a bit of knowledge about the license that QemuX 
is published with. See http://creativecommons.org/licenses/GPL/2.0/.

>  and nor Fabrice nor other QEMU collaborators see money from that, 
> like the WinZip program that is nothing but a commercial GUI for the 
> Info-ZIP compression library.

I don't see any money from that, too. Of course if if ever get 
donations I will pass something to Fabrice, but my software is far away 
from being worth a donation ;-). Fabrice could at least stop this 
iEmulator thing by telling them to follow GPL/LGPL and publishing 
sources for their qemu version.

>
> Just imagine what will happen if someone takes KQEMU makes a dirty GUI 
> and sells it. Just like the VMWare cost... Image how much could 
> someone get from KQEMU just investing 5 minutes in a RAD enviroment, 
> selling it and leaving Fabrice with NOTHING.

I must agree to Jean-Christian on that point, a GUI is not programmed 
in 5 minutes and all that qemu can earn for a GUI is publicity.

>
> I think that the current Fabrice's movement is the correct one and I 
> hope that a company will sponsor not only the KQEMU but the whole QEMU 
> project as is the BEST and MOST COMPLETE non-old computer emulator, 
> and also, its fast, and it's worth it.
>
> There are a lot of companies interested in open source projects, like 
> Novell, IBM, Apple, did anyone tried to contact them?
>
> Regards,
> Natalia Portillo
>
> P.S.: Anand I think that your action of complaining Savannah could 
> "hurt" (sorry I don't know the exact word) Fabrice and is based on 
> another person's words, and is more a child's action than one from an 
> adult. You should just have let Fabrice to see that he was breaking 
> Savannah's rules.
> P.S.2: Fabrice if you need a CVS server I offer my modest server.
>
> These are my 2 eurocents ;)
>
> El 12/02/2005, a las 6:14, Anand Kumria escribió:
>
>> On Sat, 12 Feb 2005 03:01:53 +0100, Johannes Schindelin wrote:
>>
>>> Hi,
>>>
>>> On Sat, 12 Feb 2005, Anand Kumria wrote:
>>>
>>>> Is it really fair to distribute a proprietary module from a free
>>>> software source code repository?
>>>
>>> Is it really fair to expect others to work for nothing at all?
>>>
>>
>> Are you actually this stupid?
>>
>> Or is it that English isn't you native language and you've missed the
>> subtle point that qemu is breaking the rules of savannah by 
>> distributing a
>> binary module?
>>
>> I was trying to be polite and subtle about it so that Fabrice could
>> consider his options and make a considered cautious move.
>>
>> Anyway, because of your response, I've complained to the savannah
>> administrators.
>>
>> No thanks to you.
>>
>> Anand
>>
>>
>>
>> _______________________________________________
>> Qemu-devel mailing list
>> Qemu-devel@nongnu.org
>> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>>
>
>
>
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>

Re: [Qemu-devel] Re: Re: The QEMU Accelerator Module

[Natalia Portillo]

Hi

You misunderstood me!

> This comparison lacks a bit of knowledge about the license that QemuX 
> is published with. See http://creativecommons.org/licenses/GPL/2.0/.
Are you selling QemuX?
If yes, is it being sold as the "VirtualPC killer that runs under G5"?
If i don't remember bad, answers for both is not, so I'm not refering 
to QemuX, but to iEmulator and similars that ARE A QUICK AND DIRTY GUI 
for QEMU and are being sold without even showing the QEMU's copyright.

I've never said that VMWare is a GUI, I refered that in 5 minutes you 
could (if KQEMU wasn't in propietary license) write a quick and dirty 
GUI for KQEMU and sell it for the practically the same $$$ as VMWare 
and people will buy it, and what will you do?

>>  and nor Fabrice nor other QEMU collaborators see money from that, 
>> like the WinZip program that is nothing but a commercial GUI for the 
>> Info-ZIP compression library.
>
> I don't see any money from that, too. Of course if if ever get 
> donations I will pass something to Fabrice, but my software is far 
> away from being worth a donation ;-). Fabrice could at least stop this 
> iEmulator thing by telling them to follow GPL/LGPL and publishing 
> sources for their qemu version.
I think that their QEMU version is simply 0.6.0 untouched, I must check 
that more in deep.
And what if they don't want to follow GPL/LGPL?
Who will pay the lawyers?

>>
>> Just imagine what will happen if someone takes KQEMU makes a dirty 
>> GUI and sells it. Just like the VMWare cost... Image how much could 
>> someone get from KQEMU just investing 5 minutes in a RAD enviroment, 
>> selling it and leaving Fabrice with NOTHING.
>
> I must agree to Jean-Christian on that point, a GUI is not programmed 
> in 5 minutes and all that qemu can earn for a GUI is publicity.
Do you know RealBASIC, Visual Studio .NET and other RAD (Rapid 
Application Developments)?

I wrote a GUI for ffmpeg in 5 minutes using them ;).

So definitly a GUI can be programmed in 5 minutes.

 From a GUI QEMU will earn publicity, but from a GUI that is sold like 
an emulator, as iEmulator, it wont earn anything.

>>
>> I think that the current Fabrice's movement is the correct one and I 
>> hope that a company will sponsor not only the KQEMU but the whole 
>> QEMU project as is the BEST and MOST COMPLETE non-old computer 
>> emulator, and also, its fast, and it's worth it.
>>
>> There are a lot of companies interested in open source projects, like 
>> Novell, IBM, Apple, did anyone tried to contact them?
>>
>> Regards,
>> Natalia Portillo
>>
>> P.S.: Anand I think that your action of complaining Savannah could 
>> "hurt" (sorry I don't know the exact word) Fabrice and is based on 
>> another person's words, and is more a child's action than one from an 
>> adult. You should just have let Fabrice to see that he was breaking 
>> Savannah's rules.
>> P.S.2: Fabrice if you need a CVS server I offer my modest server.
>>
>> These are my 2 eurocents ;)
>>
>> El 12/02/2005, a las 6:14, Anand Kumria escribió:
>>
>>> On Sat, 12 Feb 2005 03:01:53 +0100, Johannes Schindelin wrote:
>>>
>>>> Hi,
>>>>
>>>> On Sat, 12 Feb 2005, Anand Kumria wrote:
>>>>
>>>>> Is it really fair to distribute a proprietary module from a free
>>>>> software source code repository?
>>>>
>>>> Is it really fair to expect others to work for nothing at all?
>>>>
>>>
>>> Are you actually this stupid?
>>>
>>> Or is it that English isn't you native language and you've missed the
>>> subtle point that qemu is breaking the rules of savannah by 
>>> distributing a
>>> binary module?
>>>
>>> I was trying to be polite and subtle about it so that Fabrice could
>>> consider his options and make a considered cautious move.
>>>
>>> Anyway, because of your response, I've complained to the savannah
>>> administrators.
>>>
>>> No thanks to you.
>>>
>>> Anand
>>>
>>>
>>>
>>> _______________________________________________
>>> Qemu-devel mailing list
>>> Qemu-devel@nongnu.org
>>> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>>>
>>
>>
>>
>> _______________________________________________
>> Qemu-devel mailing list
>> Qemu-devel@nongnu.org
>> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>>
>
>
>
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>

Re: [Qemu-devel] Re: Re: The QEMU Accelerator Module

[Jean-Christian de Rivaz]

Natalia Portillo a écrit :

>> I must agree to Jean-Christian on that point, a GUI is not programmed 
>> in 5 minutes and all that qemu can earn for a GUI is publicity.
> 
> Do you know RealBASIC, Visual Studio .NET and other RAD (Rapid 
> Application Developments)?
> 
> I wrote a GUI for ffmpeg in 5 minutes using them ;).
> 
> So definitly a GUI can be programmed in 5 minutes.

As I say before: juste do it...

I can also program a very basic GUI to just start and stop the emulator 
(event that will require more that 5 minutes for sur), but this basic 
GUI will be nothing like VmWare GUI is. Understand that the price is 
also not the same. This is the first point I find wrong in your original 
say.

The second point is that the kqemu, qemu and a GUI are different things. 
You can start today a separate project to make a proprietary GUI that 
use qemu at it backend. So you can maybe get some money of that 
operation. Fact is that all contributors to qemu will never see that money.

So there is a probability that someone start a Open Source GUI that will 
trash for sur your proprietary one. This is also this risk that just 
take Fabrice with kqemu. And what? Did you know a buisness without any 
risk ?

-- 
Jean-Christian de Rivaz

Re: [Qemu-devel] Re: Re: The QEMU Accelerator Module

[René Korthaus]

Am 12.02.2005 um 20:48 schrieb Natalia Portillo:

> Hi
>
> You misunderstood me!
>
>> This comparison lacks a bit of knowledge about the license that QemuX 
>> is published with. See http://creativecommons.org/licenses/GPL/2.0/.
> Are you selling QemuX?
> If yes, is it being sold as the "VirtualPC killer that runs under G5"?
> If i don't remember bad, answers for both is not, so I'm not refering 
> to QemuX, but to iEmulator and similars that ARE A QUICK AND DIRTY GUI 
> for QEMU and are being sold without even showing the QEMU's copyright.

Okay, so I simply misunderstood... sorry for that.


>
> I've never said that VMWare is a GUI, I refered that in 5 minutes you 
> could (if KQEMU wasn't in propietary license) write a quick and dirty 
> GUI for KQEMU and sell it for the practically the same $$$ as VMWare 
> and people will buy it, and what will you do?
>
>>>  and nor Fabrice nor other QEMU collaborators see money from that, 
>>> like the WinZip program that is nothing but a commercial GUI for the 
>>> Info-ZIP compression library.
>>
>> I don't see any money from that, too. Of course if if ever get 
>> donations I will pass something to Fabrice, but my software is far 
>> away from being worth a donation ;-). Fabrice could at least stop 
>> this iEmulator thing by telling them to follow GPL/LGPL and 
>> publishing sources for their qemu version.
> I think that their QEMU version is simply 0.6.0 untouched, I must 
> check that more in deep.
> And what if they don't want to follow GPL/LGPL?
> Who will pay the lawyers?

That's a good question!

>
>>>
>>> Just imagine what will happen if someone takes KQEMU makes a dirty 
>>> GUI and sells it. Just like the VMWare cost... Image how much could 
>>> someone get from KQEMU just investing 5 minutes in a RAD enviroment, 
>>> selling it and leaving Fabrice with NOTHING.
>>
>> I must agree to Jean-Christian on that point, a GUI is not programmed 
>> in 5 minutes and all that qemu can earn for a GUI is publicity.
> Do you know RealBASIC, Visual Studio .NET and other RAD (Rapid 
> Application Developments)?
>
> I wrote a GUI for ffmpeg in 5 minutes using them ;).
>
> So definitly a GUI can be programmed in 5 minutes.
>
> From a GUI QEMU will earn publicity, but from a GUI that is sold like 
> an emulator, as iEmulator, it wont earn anything.

You're right again.

>
>>>
>>> I think that the current Fabrice's movement is the correct one and I 
>>> hope that a company will sponsor not only the KQEMU but the whole 
>>> QEMU project as is the BEST and MOST COMPLETE non-old computer 
>>> emulator, and also, its fast, and it's worth it.
>>>
>>> There are a lot of companies interested in open source projects, 
>>> like Novell, IBM, Apple, did anyone tried to contact them?
>>>
>>> Regards,
>>> Natalia Portillo
>>>
>>> P.S.: Anand I think that your action of complaining Savannah could 
>>> "hurt" (sorry I don't know the exact word) Fabrice and is based on 
>>> another person's words, and is more a child's action than one from 
>>> an adult. You should just have let Fabrice to see that he was 
>>> breaking Savannah's rules.
>>> P.S.2: Fabrice if you need a CVS server I offer my modest server.
>>>
>>> These are my 2 eurocents ;)
>>>
>>> El 12/02/2005, a las 6:14, Anand Kumria escribió:
>>>
>>>> On Sat, 12 Feb 2005 03:01:53 +0100, Johannes Schindelin wrote:
>>>>
>>>>> Hi,
>>>>>
>>>>> On Sat, 12 Feb 2005, Anand Kumria wrote:
>>>>>
>>>>>> Is it really fair to distribute a proprietary module from a free
>>>>>> software source code repository?
>>>>>
>>>>> Is it really fair to expect others to work for nothing at all?
>>>>>
>>>>
>>>> Are you actually this stupid?
>>>>
>>>> Or is it that English isn't you native language and you've missed 
>>>> the
>>>> subtle point that qemu is breaking the rules of savannah by 
>>>> distributing a
>>>> binary module?
>>>>
>>>> I was trying to be polite and subtle about it so that Fabrice could
>>>> consider his options and make a considered cautious move.
>>>>
>>>> Anyway, because of your response, I've complained to the savannah
>>>> administrators.
>>>>
>>>> No thanks to you.
>>>>
>>>> Anand
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Qemu-devel mailing list
>>>> Qemu-devel@nongnu.org
>>>> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> Qemu-devel mailing list
>>> Qemu-devel@nongnu.org
>>> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>>>
>>
>>
>>
>> _______________________________________________
>> Qemu-devel mailing list
>> Qemu-devel@nongnu.org
>> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>>
>
>
>
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel
>

[Qemu-devel] Re: The QEMU Accelerator Module

[Alexander E. Patrakov]

Fabrice Bellard wrote:

> I just commited the first alpha release of the QEMU Accelerator Module
> (aka KQEMU) in the CVS. It gives better performance for the "x86 on x86"
> case by running most of the application code as is. It works only for a
> Linux x86 host running a 2.4 or 2.6 kernel. Linux 2.4 and 2.6 kernels
> and Windows 2000 have been runnning as guest OSes, but other OSes may
> work as well. As with every alpha kernel driver testing, it is better to
> backup your data before trying it.

I think it's a good idea to test all known bugs on this new qemu. This will
probably filter out some CPU-related ones.

BTW, it's still impossible to install win2k on a 1.5G disk. I will test the
following previously-known bugs tomorrow - but feel free to race with me:

1) It's impossible to install SP4 on top of win2k SP0 installed on 2G disk
2) Impossible to install Adobe Reader 6.0 on win2k SP4 installed from
sliptreamed CD on a 2G disk
3) Firefox segfaults on win2k SP0 when entering text on www.google.com and
pressing Enter

-- 
Alexander E. Patrakov

[Qemu-devel] Re: The QEMU Accelerator Module

[Ronald]

Le Fri, 11 Feb 2005 00:31:03 +0100, Fabrice Bellard a écrit :

> Hi,
> 
> I just commited the first alpha release of the QEMU Accelerator Module
> (aka KQEMU) in the CVS. It gives better performance for the "x86 on x86"
> case by running most of the application code as is. It works only for a
> Linux x86 host running a 2.4 or 2.6 kernel. Linux 2.4 and 2.6 kernels and
> Windows 2000 have been runnning as guest OSes, but other OSes may work as
> well. As with every alpha kernel driver testing, it is better to backup
> your data before trying it.
> 
> KQEMU is _not_ open source as the rest of QEMU. It is a proprietary kernel
> module (read the LICENSE file) and will stay so until a gentle company
> decides to subsidy the QEMU project.
> 
> KQEMU usage is optional: you can disable it at compilation or run time, so
> no one is forced to use it.
> 
> Fabrice.

Strange that nobody has posted some comparisons between qemu with and
without this module, maybe it's uninteresting but I do that:

I have tried to play a big divx file with mplayer under qemu, host is
running 2.6.10, guest 2.6.9, mplayer is build without optimisations (no
mmx or sse) and launched with -ao alsa -framedrop.

Without using the module the file is far away to be played correctly: no
picture in the 3 minutes before I stop, sound is not even played correctly
: a huge number of "resetting stream".
When using the kqemu module that's a lot better, sound is decoded and
played perfectly, and the video stream is played too, just some a/v sync
problems that's why I have used -framedrop but I bet this is more a
display thing than a decoding one. I have tried some different -vo x11,dga
and sdl, the first two are ok, sdl one makes mplayer crash with a "bad
usage of CPU/FPU/RAM" message (with or without kqemu), but this may be
related to the flags¹ I have used when building qemu.

The following is the output from mplayer for the streams properties:

AVI file format detected.
VIDEO:  [DIVX]  640x368  16bpp  25.000 fps  682.8 kbps (83.4 kbyte/s)
==========================================================================
Opening audio decoder: [mp3lib] MPEG layer-2, layer-3
MP3lib: init layer2&3 finished, tables done
AUDIO: 48000 Hz, 2 ch, 16 bit (0x10), ratio: 40000->192000 (320.0 kbit)
Selected audio codec: [mp3] afm:mp3lib (mp3lib MPEG layer-2, layer-3)
==========================================================================
[...]
==========================================================================
Opening video decoder: [ffmpeg] FFmpeg's libavcodec codec family Selected
video codec: [ffodivx] vfm:ffmpeg (FFmpeg MPEG-4)
==========================================================================
Checking audio filter chain for 48000Hz/2ch/16bit -> 48000Hz/2ch/16bit...
AF_pre: af format: 2 bps, 2 ch, 48000 hz, little endian signed int AF_pre:
48000Hz 2ch Signed 16-bit (Little-Endian)


I can imagine that some more games could have a chance to run on a windows
guest now :)


¹This is what I have passed to make for this one (kqemu should be build
again after a clean distclean): 
OP_CFLAGS='-Wall -O2 -g -fno-strict-aliasing -fomit-frame-pointer
-mpreferred-stack-boundary=2 -falign-functions=0 -fno-gcse
-fno-reorder-blocks -fno-optimize-sibling-calls -march=pentium
-mcpu=athlon-xp -mmmx' CFLAGS='-Wall -O2 -g -fno-strict-aliasing
-fomit-frame-pointer -march=athlon-xp -mcpu=athlon-xp -mfpmath=sse -msse
-m3dnow' HELPER_CFLAGS='-Wall -O2 -g -fno-strict-aliasing
-fomit-frame-pointer -march=pentium -mcpu=athlon-xp -mmmx'

Re: [Qemu-devel] The QEMU Accelerator Module

[Felipe Sanchez]

On Fri, 11 Feb 2005, Fabrice Bellard wrote:

> KQEMU is _not_ open source as the rest of QEMU. It is a proprietary
> kernel module (read the LICENSE file) and will stay so until a gentle
> company decides to subsidy the QEMU project.

OK, so how much would a "gentle subsidy" be?

My company (like many others) is starting to use QEMU as a part (value add
in some cases, core in others) of some products. Although not required
too, I've personally taken interest in the QEMU project, contributed small
patches and intend to contribute a lot more once I can fully grasp QEMU's
code. That's the spirit of free software I think. We use your project and
we contribute our code and testing to it because it's in everyone's
interest to do so.

I understand Fabrice's new position on KQEMU (Specially after reading
Natalia's insightful comment). So we (my company) would really like to
help. If coding/testing is not enough that's ok. However we are not a big
operation (I'd say we are a rather small one) so we have not much buck to
spare in subsidizing a project like big fish like Novell or IBM would. So
how about setting up a fund, or a community raising effort or something so
all the so-far happy QEMU users and companies can help Fabrice continue to
develop this project?

It has been seen before. Community fund rasing efforts are often
succesful. All we need is a goal (Hence the subject of this e-mail). How
much is it? Maybe even a Slashdot front page story would be in order,
don't you think so?  ;-)


Felipe.