New page on Puppy and the OLPC project
Thanks for the pointer Sit Heel Speak.
I wish things where so simple...
You see besides the lack or proper BIOS in OLPC-XO hardware, they also have components that require specific kernel patches to allow a standard linux to run.
A properly crafted olpc.fth file (if the above is not good enough) can solve the BIOS problem but the kernel is a different story. Even the SugarLabs Fedora11 and Fedora12-based Sugar builds (the standard XO-1 UI) that use the official Fedora kernel, can not run properly on the XO. Patched are pushed *slowly* upstream but I do not know if they ever going to make it all into the official kernel branch.
So if puppy is to ever run on the XO-1 hardware (a perfect fit I might say) has to start with a proper kernel and then see what else is needed.
I do not know if puppy kernel will ever have OLPC patches. It is more likely for OLPC kernel to include puppy required modules. So what I'm hopping for is to get something running on the XO, so a pupplet can be generated, and debugged and then see if the required changes can be incorporated in the "official" OLPC kernel.
There are more than a million XO-1s out there and although most of those are "State owned" and run, they are given to school kids to have and keep with then all the time... Despite its ingenuity the Fedora/python-based Sugar UI (or any standard Linux distro) is really *heavy* on the XO hardware. Puppy linux can give a new level of experience to its young users. Ideally the Sugar UI (that now can be packaged for many distros) could run as an option on the top of puppy and get rejuvenated (speed-wise)
But I'm blubbering... Back to the kernel... Any takers...? pointers?....
I wish things where so simple...
You see besides the lack or proper BIOS in OLPC-XO hardware, they also have components that require specific kernel patches to allow a standard linux to run.
A properly crafted olpc.fth file (if the above is not good enough) can solve the BIOS problem but the kernel is a different story. Even the SugarLabs Fedora11 and Fedora12-based Sugar builds (the standard XO-1 UI) that use the official Fedora kernel, can not run properly on the XO. Patched are pushed *slowly* upstream but I do not know if they ever going to make it all into the official kernel branch.
So if puppy is to ever run on the XO-1 hardware (a perfect fit I might say) has to start with a proper kernel and then see what else is needed.
I do not know if puppy kernel will ever have OLPC patches. It is more likely for OLPC kernel to include puppy required modules. So what I'm hopping for is to get something running on the XO, so a pupplet can be generated, and debugged and then see if the required changes can be incorporated in the "official" OLPC kernel.
There are more than a million XO-1s out there and although most of those are "State owned" and run, they are given to school kids to have and keep with then all the time... Despite its ingenuity the Fedora/python-based Sugar UI (or any standard Linux distro) is really *heavy* on the XO hardware. Puppy linux can give a new level of experience to its young users. Ideally the Sugar UI (that now can be packaged for many distros) could run as an option on the top of puppy and get rejuvenated (speed-wise)
But I'm blubbering... Back to the kernel... Any takers...? pointers?....
- Sit Heel Speak
- Posts: 2595
- Joined: Fri 31 Mar 2006, 03:22
- Location: downwind
I understand. I'm developing a program to provide visual comparison across dotconfig's of several distro's, specifically Fedora 11, Ubuntu 9.04, and several Pup(pie)s. Thus I hope to learn what I need to add to the Puppy 2.6.29.6 Aug 22 kernel, to run all the gadgets my client wants to, on the Mini-9.mavrothal wrote:...components that require specific kernel patches...So if puppy is to ever run on the XO-1 hardware...has to start with a proper kernel...
I could do a similar analysis for the XO-1, if I had the dotconfigs for the kernel(s) which have been patched to run most sweetly (or at least, least awfully) on it. This is the first I've heard of Fedora 12 (didn't know it had advanced beyond 11; 12 isn't supposed to come out until November 3rd, says Wikipedia) --are you referring to a Fedora 11 on which the kernel has been upgraded through version stepping and / or driver patches? If so, I would like a dotconfig for it, if you would be so kind as to attach it. And pointers to the patches. And then perhaps I can better grasp the problem.
Also any list you can point me to, of which CONFIG_ switch in menuconfig goes with which patch, would be helpful.
Is there a particular reason why a 2.6.30 kernel would be superior to a 2.6.29 kernel, on the XO-1?
As I said kernel hacking is far from my abilities but I think that this mail http://lists.laptop.org/pipermail/devel ... 24939.html has the info you want. You can also look here for kernel rpms http://dev.laptop.org/~dilinger/olpc-2.6.30-xo1/.Sit Heel Speak wrote:I could do a similar analysis for the XO-1, if I had the dotconfigs for the kernel(s) which have been patched to run most sweetly (or at least, least awfully) on it.
You are right. The F12 are development builds.Sit Heel Speak wrote:This is the first I've heard of Fedora 12 (didn't know it had advanced beyond 11; 12 isn't supposed to come out until November 3rd, says Wikipedia)
I'm not sure I understand what exactly you are asking for. Sorry. If you can tell me where I should be looking for these (eg in my XO?) I could try to find them... And I certainly have no idea if and why 2.6.30 may be superior to 2.6.29. All I know is that is used successfully on OLPC hardware.Sit Heel Speak wrote:If so, I would like a dotconfig for it, if you would be so kind as to attach it. And pointers to the patches. And then perhaps I can better grasp the problem.
Also any list you can point me to, of which CONFIG_ switch in menuconfig goes with which patch, would be helpful.
Is there a particular reason why a 2.6.30 kernel would be superior to a 2.6.29 kernel, on the XO-1?
Thanks for your help.
- Sit Heel Speak
- Posts: 2595
- Joined: Fri 31 Mar 2006, 03:22
- Location: downwind
Ah so. Well, if Mr. Saxena is the lead developer, and he says "all development for both XO-1 and XO-1.5 will be done on the olpc-2.6.30 branch of the olpc-2.6 repository" then that means 2.6.30 will be from here on out superior!mavrothal wrote:You can also look here for kernel rpms http://dev.laptop.org/~dilinger/olpc-2.6.30-xo1/...no idea if and why 2.6.30 may be superior to 2.6.29. All I know is that is used successfully on OLPC hardware.
Back when I have something practical to contribute.
Any progress on that? I'm trying to use olpc-kernel source with squashfs support to rebuild the puppy 4.2.1 kernel and also meet olpc-xo1 requirements/patches (they are both 2.6.25) but the menuconfig is a nightmare...Sit Heel Speak wrote:I understand. I'm developing a program to provide visual comparison across dotconfig's of several distro's, specifically Fedora 11, Ubuntu 9.04, and several Pup(pie)s. Thus I hope to learn what I need to add to the Puppy 2.6.29.6 Aug 22 kernel, to run all the gadgets my client wants to, on the Mini-9.mavrothal wrote:...components that require specific kernel patches...So if puppy is to ever run on the XO-1 hardware...has to start with a proper kernel...
Alternatively, can anyone point me to a post that describes the minimal puppy kernel configuration, eg without the components that assure compatibility with a wide range of hardware/peripherals/network/video/modems etc. The olpc_config should take care of the hardware compatibility part but what else (other than squashfs) is _mandatory_ for the full puppy experience?
I did look around the forum and BK's pages/blog bun no luck...
Thx
xopuplet
As mentioned, I'm trying to get puppy 4.2.1 on the OLPC XO-1 computer. Given the many hardware requirements of the XO I decided to go with the 2.6.25-derived OLPC kernel patched with the official Fedora-9 squashfs-3.3 patch. I used a full install of puppy4.2.1 with devx_421.sfs, kernel-src-2.6.25.16-patched-puppy4.1.sfs as well as puppy-unleashed-core-421 all in full install in the HD.
Comparing the Puppy DOTconfig and OLPC_defconfig I made a kernel that will keep OLPC requirements and will incorporate many puppy settings (I left out the firmware, PCI etc things that are useless for the specific hardware). The vmlinuz will boot the XO fine but will not load puppy 421. I would guess because the initrd.gz is not from the same kernel. And here is the problem.
The OLPC kernel will not boot puppy on the development machine. I'm using puppy on Parallels VM on a Mac (so I can go back after the inevitable mess-ups) and the OLPC kernel will not boot in a VM (Parallels, VMWare or Qemu) nor in a modern Intel CPU! Of course puppy's mkinitrd script will make an init from the running kernel, so I'm stuck.
I tried to make a new initrd just by manually issuing the relevant script commands but ended up with an initrd that gave me a kernel panic when I booted the XO.
So the short question is: is it possible to to make a pupplet with a rebuild kernel when this kernel is not running on the development machine?
I do not care if the development machine will be bricked at the end of the process since I can always go back to a previous snapshot. As long as I can get vmlinuz, initrd.gz and "xopup".sfs in a stick, I'm fine.
Oh... did I mentioned that I'm new to Linux development and this is my first attempt in puppy development?...
Comparing the Puppy DOTconfig and OLPC_defconfig I made a kernel that will keep OLPC requirements and will incorporate many puppy settings (I left out the firmware, PCI etc things that are useless for the specific hardware). The vmlinuz will boot the XO fine but will not load puppy 421. I would guess because the initrd.gz is not from the same kernel. And here is the problem.
The OLPC kernel will not boot puppy on the development machine. I'm using puppy on Parallels VM on a Mac (so I can go back after the inevitable mess-ups) and the OLPC kernel will not boot in a VM (Parallels, VMWare or Qemu) nor in a modern Intel CPU! Of course puppy's mkinitrd script will make an init from the running kernel, so I'm stuck.
I tried to make a new initrd just by manually issuing the relevant script commands but ended up with an initrd that gave me a kernel panic when I booted the XO.
So the short question is: is it possible to to make a pupplet with a rebuild kernel when this kernel is not running on the development machine?
I do not care if the development machine will be bricked at the end of the process since I can always go back to a previous snapshot. As long as I can get vmlinuz, initrd.gz and "xopup".sfs in a stick, I'm fine.
Oh... did I mentioned that I'm new to Linux development and this is my first attempt in puppy development?...
new kernel
Puppy 4.3beta3 has new kernel. You may want to try this.
http://puppylinux.com/blog/?viewDetailed=01072
http://puppylinux.com/blog/?viewDetailed=01072
Puppy user since Oct 2004. Want FreeOffice? [url=http://puppylinux.info/topic/freeoffice-2012-sfs]Get the sfs (English only)[/url].
Re: new kernel
Well, the pup4.3b1 gave me some hope on this and got me started (see previous page). However even the olpc-2.6.30 kernel is not working properly on the olpc-xo yet, so I opted for the "tested" 2.6.25. Also the patched 2.6.30.5 kernel source is not available yet.raffy wrote:Puppy 4.3beta3 has new kernel. You may want to try this.
http://puppylinux.com/blog/?viewDetailed=01072
However 2.6."whatever" kernel will not solve the problem of remastering from a non-booted kernel. Any thoughts on that?
Please Barry - Work your magic!
Any luck getting Puppy to work on the OLPC?
All my rudimentary attempts have failed, and I have yet to read a successful post anywhere on the vast Internet.
Thanks!
All my rudimentary attempts have failed, and I have yet to read a successful post anywhere on the vast Internet.
Thanks!
Message to BarryK
Any chance of you editing this thread's first page link to http://www.puppyos.com/olpc/ ....
a now defunct 'parked' webpage/IP address
and title....'new' ....page?
Any news since
http://www.olpcnews.com/forum/index.php ... 384.0;wap2
http://www.olpcnews.com/forum/index.php ... 356.0;wap2
Would it not be possible to 'make overtures' for renewing funding/partnership with the OLPC team as you, ttuuxxx, and SHS to name a few, are already 'involved in this development', as I understand it....?
IMHO, this would be a huge kudos for Puppy/woof if it could be pulled off to get the XO-1 to boot/run Puppy
....and a huge boost to the apparently dying, but million+ component XO-1 project
thanks, Barry, ....& good luck, team!
Aitch
Any chance of you editing this thread's first page link to http://www.puppyos.com/olpc/ ....
a now defunct 'parked' webpage/IP address
and title....'new' ....page?
Any news since
also, have you seen mavrothal's pages hereBarryK wrote:I am now experimenting with the 2.6.27.x kernel, which is "OLPC aware".
http://www.olpcnews.com/forum/index.php ... 384.0;wap2
http://www.olpcnews.com/forum/index.php ... 356.0;wap2
Would it not be possible to 'make overtures' for renewing funding/partnership with the OLPC team as you, ttuuxxx, and SHS to name a few, are already 'involved in this development', as I understand it....?
IMHO, this would be a huge kudos for Puppy/woof if it could be pulled off to get the XO-1 to boot/run Puppy
....and a huge boost to the apparently dying, but million+ component XO-1 project
thanks, Barry, ....& good luck, team!
Aitch
- ttuuxxx
- Posts: 11171
- Joined: Sat 05 May 2007, 10:00
- Location: Ontario Canada,Sydney Australia
- Contact:
When I read those links, and a few users were talking about DSL/Puppy, DSL would not have a chance, the kernel is too old, not enough developers, they use a really old version of FF, etc. It would be a lost cause. I'm still working on Xo, like a good wine, it takes timeAitch wrote:
http://www.olpcnews.com/forum/index.php ... 384.0;wap2
http://www.olpcnews.com/forum/index.php ... 356.0;wap2
thanks, Barry, ....& good luck, team!
Aitch
ttuuxxx
http://audio.online-convert.com/ <-- excellent site
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)
I agree that the sugar interface was a terrible choice for a low-powered machine. But I also don't think that a run-from-RAM OS like Puppy is right for the job either. The XO has limited memory as well as limited CPU clocking, so running from RAM is not gonna be ideal either. That doesn't mean that Puppy has nothing to offer -it does have a lot of small utilities and some concepts which would help make such hardware more usable.
I think that any efforts to create an OS for such hardware should focus on fast-boot times, but not by using up the scarce RAM resources. Instead, using readahead, pre-linking and other such techiniques would achieve the desired result.
Besides the problems of using python or java for the (sugar) inetrface, the other projects which have built systems for the XO have all started with over-bloated systems from the start. Building an OS for specific hardware has the advantage of not needing to include support for hundreds/thousands of devices which might *possibly* be needed, so it really makes no sense to be running udev/HAL or other such services which are waiting for the user to connect nearly any kind of device.
I think that any efforts to create an OS for such hardware should focus on fast-boot times, but not by using up the scarce RAM resources. Instead, using readahead, pre-linking and other such techiniques would achieve the desired result.
Besides the problems of using python or java for the (sugar) inetrface, the other projects which have built systems for the XO have all started with over-bloated systems from the start. Building an OS for specific hardware has the advantage of not needing to include support for hundreds/thousands of devices which might *possibly* be needed, so it really makes no sense to be running udev/HAL or other such services which are waiting for the user to connect nearly any kind of device.
Amigo
I'm not sure that Sugar is a "terrible choice" but I would agree that the "heavy" VM-based (python) implementation is far from desirable on a low-specs machine. Even SugarLabs realizes that other environments like Qt might be better for low-spec machines.
Regarding puppy and running from RAM in 256 machines, that's maybe the reason that "I fall in love" with it . When I first tried it in a practically unusable old P4 machine, was screaming. When I tried it on an old PentiumM laptop downgraded to 256MB Ram was blazing and since then I have tried all shorts of pupplets on VM always with 256BM and they are happy puppies as ever...
Running from RAM is even more important on SSD machines since random writes are really terrible on these systems. More than often the XO hangs for 1-3 sec doing these writes. I would actually suggest for the XOpup to do away even from swap and try compcache instead. People are reporting "success and happiness" when implemented on the XO.
Granted running OOo3, heavy python apps, flash or video editing is always going to be problem on this hardware, but these are not the target apps for the XO and even puppy I might add. There is no reason to pay a penalty for 99% or your time/users to have the chance to use something 1% of the time. And puppy's sfs-approach is very wise on that aspect.
Indeed, many kernel modules and drivers may be useless for the XO and a carefully crafted kernel might be the most important aspect of a speedy system, but that's were ttuuxxx and (hopefully) other puppy developers come in... And you can always provide additional modules separately that can be injected on a need bases in one's system. But that's for later...
However udev/HAL is something that I think should be there, there are always peripherals to be added (an external mouse is almost mandatory) and power-management without HAL is not really easy. So I wouldn't mess with it. Besides is not a big burden anyway.
I'm not sure that Sugar is a "terrible choice" but I would agree that the "heavy" VM-based (python) implementation is far from desirable on a low-specs machine. Even SugarLabs realizes that other environments like Qt might be better for low-spec machines.
Regarding puppy and running from RAM in 256 machines, that's maybe the reason that "I fall in love" with it . When I first tried it in a practically unusable old P4 machine, was screaming. When I tried it on an old PentiumM laptop downgraded to 256MB Ram was blazing and since then I have tried all shorts of pupplets on VM always with 256BM and they are happy puppies as ever...
Running from RAM is even more important on SSD machines since random writes are really terrible on these systems. More than often the XO hangs for 1-3 sec doing these writes. I would actually suggest for the XOpup to do away even from swap and try compcache instead. People are reporting "success and happiness" when implemented on the XO.
Granted running OOo3, heavy python apps, flash or video editing is always going to be problem on this hardware, but these are not the target apps for the XO and even puppy I might add. There is no reason to pay a penalty for 99% or your time/users to have the chance to use something 1% of the time. And puppy's sfs-approach is very wise on that aspect.
Indeed, many kernel modules and drivers may be useless for the XO and a carefully crafted kernel might be the most important aspect of a speedy system, but that's were ttuuxxx and (hopefully) other puppy developers come in... And you can always provide additional modules separately that can be injected on a need bases in one's system. But that's for later...
However udev/HAL is something that I think should be there, there are always peripherals to be added (an external mouse is almost mandatory) and power-management without HAL is not really easy. So I wouldn't mess with it. Besides is not a big burden anyway.
My name is Alejandro Lavarello and I live in Uruguay, the first contry to give XO for all scholar childrens. My daughter Emilia has a XO, and I have used it. I have some knowledge of working with computers, since 1992 I have used Z80-clones, PDPs, VAXes, XTs and now I have a Pentium 4, running Puppy 4.3.1 and XP.
My point is: Sugar is a very bad interface. Copy files from Journal is horrible. When the child grow, he needs a REAL file management application, and a standard GUI.
I hope that Puppy can be ported to XO. Here in Uruguay, the LATU centralized all about OLPC and adds blocking features. They do not allow to install another operating system in the XO. It is a dictatorial behavior, but we live in democracy.
It will be great if Puppy can be burned in a SD card and booted from it.
My point is: Sugar is a very bad interface. Copy files from Journal is horrible. When the child grow, he needs a REAL file management application, and a standard GUI.
I hope that Puppy can be ported to XO. Here in Uruguay, the LATU centralized all about OLPC and adds blocking features. They do not allow to install another operating system in the XO. It is a dictatorial behavior, but we live in democracy.
It will be great if Puppy can be burned in a SD card and booted from it.
- ttuuxxx
- Posts: 11171
- Joined: Sat 05 May 2007, 10:00
- Location: Ontario Canada,Sydney Australia
- Contact:
It will Alejandro, I'm still working on it, just getting the base solid before compiling the kernel, It won't be quick, but It will happen.alejol wrote:My name is Alejandro Lavarello and I live in Uruguay, the first contry to give XO for all scholar childrens. My daughter Emilia has a XO, and I have used it. I have some knowledge of working with computers, since 1992 I have used Z80-clones, PDPs, VAXes, XTs and now I have a Pentium 4, running Puppy 4.3.1 and XP.
My point is: Sugar is a very bad interface. Copy files from Journal is horrible. When the child grow, he needs a REAL file management application, and a standard GUI.
I hope that Puppy can be ported to XO. Here in Uruguay, the LATU centralized all about OLPC and adds blocking features. They do not allow to install another operating system in the XO. It is a dictatorial behavior, but we live in democracy.
It will be great if Puppy can be burned in a SD card and booted from it.
ttuuxxx
http://audio.online-convert.com/ <-- excellent site
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)
alejol
Thank you for letting us know of this problem with LATU
I do not think the EU will be very happy with this 'control' of a democratic freedom by LATU as they are funding the region quite heavily [18 million euros] through LATU management
see http://latu21.latu.org.uy/en/index.php? ... Itemid=260
Also, I do not think this is what was intended by OLPC, so they may not be happy, either....?
Perhaps googlebot/our posts/these pages will help to make them aware of this hindrance to your children's OLPC project, and the future for your countryfolk and bring pressure for change?
Good Luck
Forward Puppy! - thanks ttuuxxx
Aitch
Thank you for letting us know of this problem with LATU
I do not think the EU will be very happy with this 'control' of a democratic freedom by LATU as they are funding the region quite heavily [18 million euros] through LATU management
see http://latu21.latu.org.uy/en/index.php? ... Itemid=260
Also, I do not think this is what was intended by OLPC, so they may not be happy, either....?
Perhaps googlebot/our posts/these pages will help to make them aware of this hindrance to your children's OLPC project, and the future for your countryfolk and bring pressure for change?
Good Luck
Forward Puppy! - thanks ttuuxxx
Aitch
- ttuuxxx
- Posts: 11171
- Joined: Sat 05 May 2007, 10:00
- Location: Ontario Canada,Sydney Australia
- Contact:
Your right Aitch, I'm not for any sort of censorship. That will drive me even more to succeed in this project.
ttuuxxx
ttuuxxx
http://audio.online-convert.com/ <-- excellent site
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)
I hope you can port Puppy to OLPC XO
Hi folks. Sorry by my insistence, but I suffer the horrible Sugar interface.
When I tried to save the Journal's files from my daughter XO, we end with
400MB of obscure files. And I make this using command line, because in Journal you can not save multiple entries at once.
The childrens here in Uruguay are accostumed to delete all files in Journal and to start again, because when Journal is almost full, XO do not work.
It is equal to erase all the notebook. All of this work and his favourite songs, etc. will be erased regularly.
This is a shame. And a child of 10 or more years need a real GUI similar to real computers in the real world.
I hope that Puppy can work in the OLPC.
If I can help, please let me know at alejandro.lavarello (at) gmail . com
When I tried to save the Journal's files from my daughter XO, we end with
400MB of obscure files. And I make this using command line, because in Journal you can not save multiple entries at once.
The childrens here in Uruguay are accostumed to delete all files in Journal and to start again, because when Journal is almost full, XO do not work.
It is equal to erase all the notebook. All of this work and his favourite songs, etc. will be erased regularly.
This is a shame. And a child of 10 or more years need a real GUI similar to real computers in the real world.
I hope that Puppy can work in the OLPC.
If I can help, please let me know at alejandro.lavarello (at) gmail . com
LATU blocks another operating system in XO
Yesterday, I have asked to official LATU's technical support about a second operating system in uruguayan's XO. The answer is:
"
Hello Alejandro,
The XO delivered in primary schools because of its security system
not allow the installation of another operating system.
Any question you have not feel free to communicate.
Greetings,
Natalia González
"
I think that is a sort of dictatorial behavior.
The response, in Spanish, is:
---------------------------------------------
de Apoyo Técnico <sitio.tecnico@plan.ceibal.edu.uy>
para Alejandro Lavarello <alejandro.lavarello@gmail.com>
fecha 23 de noviembre de 2009 08:18
asunto Re: Ceibal: Otro SO en las XO
Hola Alejandro,
Las XO entregadas en primaria debido a su sistema de seguridad
no permiten la instalación de otro sistema operativo.
Cualquier consulta que tengas no dudes en volver a comunicarte.
Saludos,
Natalia González
Apoyo Técnico - Plan Ceibal
Avda. Italia 6201 C.P.: 11500
Montevideo - Uruguay
-----------------------------------------------
"
Hello Alejandro,
The XO delivered in primary schools because of its security system
not allow the installation of another operating system.
Any question you have not feel free to communicate.
Greetings,
Natalia González
"
I think that is a sort of dictatorial behavior.
The response, in Spanish, is:
---------------------------------------------
de Apoyo Técnico <sitio.tecnico@plan.ceibal.edu.uy>
para Alejandro Lavarello <alejandro.lavarello@gmail.com>
fecha 23 de noviembre de 2009 08:18
asunto Re: Ceibal: Otro SO en las XO
Hola Alejandro,
Las XO entregadas en primaria debido a su sistema de seguridad
no permiten la instalación de otro sistema operativo.
Cualquier consulta que tengas no dudes en volver a comunicarte.
Saludos,
Natalia González
Apoyo Técnico - Plan Ceibal
Avda. Italia 6201 C.P.: 11500
Montevideo - Uruguay
-----------------------------------------------