Porteus ARM from ARMedSlack for The Future!

New features which should be implemented in Porteus; suggestions are welcome. All questions or problems with testing releases (alpha, beta, or rc) should go in their relevant thread here, rather than the Bug Reports section.
User avatar
WildeGeist
White ninja
White ninja
Posts: 11
Joined: 05 Mar 2012, 23:42
Distribution: V1.2
Location: Austin Texas
Contact:

Porteus ARM from ARMedSlack for The Future!

Post#1 by WildeGeist » 15 Mar 2012, 06:45

I have been reading up on where computers are headed, and I think ARM is an important niche to stay on top of. I am not sure what all is entailed, but I do have a love for Porteus, and would like to be with Porteus as She grows.

Reading some papers written some time ago, from prophets and sages in the computing world back then, they said basically what I personally have been desiring as I move from the dreaded end of OS X to the F/OSS world of wonder.

The Anti-Mac papers http://www.advogato.org/article/130.html and http://www.useit.com/papers/anti-mac.html basically sum up a lot of truths.

ARM computers are going to be more wanting from the poor souls who went on the iPad and Android kick. When they come back to realizing real computers have more power, they will be desiring the low power consumption and portability capabilities of ARM, but want POWER, which lies in the Terminal. And so far for most of us trying to learn the command line, its not an easy transition, unless born and raised into it like so many currently are today; and which these Anti-Mac papers spoke of back in the day.

Slackware is about the only original OS that still accommodates all levels of Operators. Arch, FreeBSD, Crux, they still remain hard to learn relatively. Though Arch and FreeBSD are active in ARM, they remain what they are, not accommodating to the run of the mill computer user. ARMedSlack on the other hand, being of Slackware, is Slackware. And the Porteus I believe should have an ARM Porteus, since its such a groovy handy tool that could perform its wonders in the ARM world which is coming.

I beg of the Porteus gang to take a peek at ARMedSlack http://www.armedslack.org/ and see what all is needed to get an ARM project rolling for Porteus. I emailed the Trim-Slice people showing them Porteus, and kind of pulled at them to consider funding for Porteus. They have not replied yet. I have seen some work Arch ARM is doing on the Trim-Slice. The Trim-Slice is not exactly "new", but it is a forerunner for what is to come.

I really believe in Porteus, and personally would like it to be my OS of choice whether on a Think Pad or an ARM machine when you all get that partition manager going. As is it is still a wonderful tool. Saved my butt when I encountered a huge disaster when attempting to be rid of OS X on this MacBook Pro once and for all, it was horrible. But then again, not so bad, because Porteus was the only means I had to get a System going so I could get online and research how to fix my broken machine. I got suckered into trying a pre-release of the PC-BSD 9.1 . And the ZFS scheme along with GPT/MBR made it very hard to boot anything in my arsenal, especially the OS X. Porteus rescued me.

Computers in the near future are not going to be KDE centric, but more into the Terminal. I am currently investing my energy into getting a personal vision going of an interactive Terminal/Editor to handle people like me who are not born into the Terminal. Like a massive talking and listening Emacs. I think it a travesty for F/OSS to expend so much effort to build massive Desktop Environments full of GUI just like OS X and Windows, like Ubumtu for instance. As people flock again to F/OSS, they want the Power, and that is in the command line. Making that more accessible is my aim. I am more an Open Box kind of guy wishing I knew the Terminal better. So I want to build a massive talking farting singing Emacs. I know, frightening, but for us new to it, its needed.

So anyway, back to ARM. Its so cheap now, just imagine in a couple years, even people in the most poor places will be able to pick up ARM machines for peanuts. And if Porteus were there, everybody could be up and running in no time with a System not lacking a thing, but giving them a tool to do anything.

Anything I can do to help, I am all ears. As I said, I am kinda naive as to what all is entailed. But I know the miracle you all have pulled off with Porteus 1.2rc, and its truly amazing. I doubt it is too far off for you to take off from ARMedSlack and Arch ARM, and run with it making a Porteus ARM version. As I learn things and can feel safe in applying them, I hope to be of help here ARM or not. Outside of building my own Slackware, Arch and FreeBSD Think Pad when I get back to Hawaii, I will always try to keep up with Porteus. The others are for learning purposes for me mainly. As is, all my wishes of expertise in building a System, are already here in Porteus.

Please give it some serious consideration, and I will keep plugging along on my end to see where I can fit in. Any suggestions or criticism is welcome. I like being where the winners are. Porteus is all that! I would like to see it dominate the ARM Race too!

Rock on!

Georg
Porteus, Slackware, FreeBSD on Habey MITX-6500 Development (ARM)

User avatar
fanthom
Moderator Team
Moderator Team
Posts: 5666
Joined: 28 Dec 2010, 02:42
Distribution: Porteus Kiosk
Location: Poland
Contact:

Re: Porteus ARM from ARMedSlack for The Future!

Post#2 by fanthom » 16 Mar 2012, 08:04

@WildeGeist

from my point of view this is not going to happen soon. main reason: i dont own any ARM based PC so wont be able to work on it myself.
i'm ready to share my knowledge with someone who wants to make ARM port of Porteus.
Please add [Solved] to your thread title if the solution was found.

User avatar
brokenman
Site Admin
Site Admin
Posts: 6105
Joined: 27 Dec 2010, 03:50
Distribution: Porteus v4 all desktops
Location: Brazil

Re: Porteus ARM from ARMedSlack for The Future!

Post#3 by brokenman » 22 Mar 2012, 01:57

Computers in the near future are not going to be KDE centric, but more into the Terminal. I am currently investing my energy into getting a personal vision going of an interactive Terminal/Editor to handle people like me who are not born into the Terminal.
Personally i don't think things will regress back to terminal operation. I started my computer life in a terminal on dual color screens. Things progressed and trends show a clear migration to GUI as being favored. The geeks will always prefer direct terminal access and getting there hands dirty, but progress demands ease of use and intuitive interface development ... both of which the terminal is not ... and DE's are.

People wanting to play with ARM should remember that virtual machines can emulate this quite well ... but virtual won't cut it when it comes to real life development of a releasable system.

I would love to jump into ARM as i believe it IS where things are headed ... but the learning curve is steep, and the clock stops for no man.
How do i become super user?
Wear your underpants on the outside and put on a cape.

User avatar
Ahau
King of Docs
King of Docs
Posts: 1331
Joined: 28 Dec 2010, 15:18
Distribution: LXDE & Xfce 32/64-bit
Location: USA

Re: Porteus ARM from ARMedSlack for The Future!

Post#4 by Ahau » 23 Mar 2012, 19:41

Exciting news for me: I've just ordered an android tablet with an arm processor (nvidia tegra 2 SoC, like almost all the other last-year model android tablets). I doubt I'll be able to create a functional porteus rom (can we please refer to this as the Arm port Porteus arm?), as XFCE, documentation and support for the standard edition needs to take precedence and my free time is limited, but I will likely take a crack at something along these lines.

I should spell out that, based on my research thus far, creating a unified ARM distro appears to be a nearly impossible task, as each kernel needs to be tailored to the hardware on the board. So, I might be able to build an image to boot porteus on my specific tablet, but anyone wishing to run the userland I produce would have to either own the same tablet or compile their own kernel and install their own firmware, then add the userland on top of it. That's not a very user-friendly approach.
Please take a look at our online documentation, here. Suggestions are welcome!

quotaholic
Black ninja
Black ninja
Posts: 75
Joined: 15 May 2011, 16:20
Location: denver
Contact:

Re: Porteus ARM from ARMedSlack for The Future!

Post#5 by quotaholic » 27 Jun 2013, 01:23

I may have some arm kernel development work coming in soon. One project for vision system application. Saw a slackbuild for opencv out there somewhere. The other is more industrial in nature. I have a high desire of interest in learning how to package what is needed to make porteus work in ARM. Now that I have gotten past making working kernels for Porteus this seems like a good next step.

I have not looked at the ARMedSlack project nor what chips it supports but I think I have a bagelbone black coming. I'll have to look in to device overlay trees and see if they can be incorporated in to the build. Most developers in ARM are using DOT's frequently and changing them out for others. I'll have more info in a week but I also have interest in this.

quotaholic

User avatar
brokenman
Site Admin
Site Admin
Posts: 6105
Joined: 27 Dec 2010, 03:50
Distribution: Porteus v4 all desktops
Location: Brazil

Re: Porteus ARM from ARMedSlack for The Future!

Post#6 by brokenman » 27 Jun 2013, 01:39

I must be getting old. Most of what you just said in your second paragraph made absolutely no sense to me. i thought Bilbo Bagelbones was a hobbit! You should check out Ahau's work on ARM. He has done the ground work for Porteus already.
How do i become super user?
Wear your underpants on the outside and put on a cape.

quotaholic
Black ninja
Black ninja
Posts: 75
Joined: 15 May 2011, 16:20
Location: denver
Contact:

Re: Porteus ARM from ARMedSlack for The Future!

Post#7 by quotaholic » 02 Jul 2013, 16:50

When I searched ARM on Porteus for some reason this thread came up before Ahau's ARM thread.

Device overlay trees better explained:
http://elinux.org/images/4/48/Experienc ... _SOC's.pdf
http://hipstercircuits.com/category/device-tree/
http://en.wikipedia.org/wiki/Device_con ... on_overlay
http://www.crashcourse.ca/wiki/index.php/Device_trees

If i want to take advantage of a bunch of GPIO I can write a lot of code or I can use a DTS file to predefine some config.

They are great for expansion. If you want to interact with other devices they are a great jump start. If you only want do use computer for desktop purposes they would not be of much use. Didn't know about these myself until I started to look in to industrial applications and slaving I/O to vision system events. Possibly one could use this to identify a hobbit at the door....

Based on subject popularity if Porteus was to incorporate such tools and do it "Porteus Style" I think that user base would increase quite a bit.

Still awaiting hardware.... Sigh.

User avatar
francois
Contributor
Contributor
Posts: 6434
Joined: 28 Dec 2010, 14:25
Distribution: xfce plank porteus nemesis
Location: Le printemps, le printemps, le printemps... ... l'hiver s'essoufle.

Re: Porteus ARM from ARMedSlack for The Future!

Post#8 by francois » 03 Jul 2013, 08:36

Very interesting discussion. Please keep going on. I have a Samsung Galaxy s android powered phone.
Prendre son temps, profiter de celui qui passe.

quotaholic
Black ninja
Black ninja
Posts: 75
Joined: 15 May 2011, 16:20
Location: denver
Contact:

Re: Porteus ARM from ARMedSlack for The Future!

Post#9 by quotaholic » 03 Jul 2013, 19:14

Thank you for your interest.

DTS or device tree files for existing devices are going to be located in a subfolder of the Linux kernel.

https://archive.fosdem.org/2013/schedul ... kernel.pdf

Porteus, being built on slackware, a very vanilla release, is also about the most modular linux available. PXE, boot cheats, etc. It makes an ideal candidate for ARM.

The problem with ARM is that the number of devices releasing to market is only going to get larger and larger. If its not the ARM board or SOC itself then its going to be the accessory devices that need support. Applications may be more widespread than one thinks. Sure there is the growing community of Arduino hackers but I am looking more towards "larger" applications. I have already heard of industrial "mission critical" applications where people are taking arm stick computers and making their own networking protocol stack and using the two in conjunction to make a truly secured network reassembling packets one hop before destination. Possible Byzanthium application? Industrial manufacturers are sending in specifications to Industry specific custom manufacturers asking for ARM cpu based equipment. I will not mention the number of orders that will be filled once this spec is met. All I will say is that the largest names in the electronics community are behind this spec and initiative.

Ahau or anyone else hacking away on ARM will tell you that the broad range of devices is going to make it very very hard to keep one tree of code for. If not impossible. I cant help but think, and please prove me wrong, that these DTS files are going to be invaluable to developers as well as anyone playing with Linux on ARM for extending existing functionality to new devices and or peripherals.

I spent 5 combined years working for Mikron Automation:

http://www.mikron.com/en/mikron-automation/

and Clear Automtion:

http://www.clearautomation.com/pages/welcome.asp

My inquiry about the possible support of DTS files within Porteus is not unqualified. Its too early to tell yet however in an ideal world I may be asking for help on a paid contract in coming weeks / months. Industrial applications are likely not going to stick with legacy or low performance hardware. Thus I need to know if I can plan on an easy way to change out such files on the fly (this would be needed if the machine in question ended a manufacturing run on one product and had to be retooled for another. Or simply in R&D bench testing. Other options include using openembedded which is slow to dev on and dont even get me started on bitbake support.

Sadly I am between jobs (industrial automation really is a volatile market) and my donated BeagleBone Black may not come through after all. I start a webhosting support job on Monday so eventually I will be working on one. Until then I will continue to research.

I was hoping that DTS files could be made in to a module within Porteus. I have to guess that Fanthom has some neat scripts that can carve out things like kernel-source and the likes. Possibly it wouldn't be hard to do the same on an ARM specific kernel and branch out the DTS folder or insertion point as a linkable point that a module could be aimed at???
Over the last year, ARM has gone from a constant
headache every merge window to an outstanding
citizen in the Linux community
Linus Torvalds, August 2012

Post Reply