Puppy linux - ARM

[qkall]

when i first heard of this raspberry pi project i immediately thought of puppy linux. its faster than any nix distro i've tried and very resource light with the minimum required specs are well below the specs of the raspberry pi. the problem is puppy currently isn't arm compatible... but there are puppy devs very interested in developing for the raspi as it falls perfectly with puppy development. the huge stumbling block they see, however, is the boot loader. puppy boots differently they typical nix os's.

so i was wondering if any of the devs could help the puppy community out by helping them get puppy to boot and/or some board donations to Barry Kauler (lead puppy dev).

i know nothing of anything in developing .. .i'm just the messenger.

two threads about raspi
http://www.murga-linux.com/pup.....369#572369
http://www.murga-linux.com/pup.....hp?t=70475

[Svartalf]

Heh... Seems I need to get over to their forums... I'm in the process of working out what needs to be done to get a mutant Puppy going.

[emercer]

I believe a mutant puppy (or maybe a raspberry puppy or some puppy pi?) is a great addition to the list of available distro choices for the RasPi.

[Svartalf]

Quote from emercer on October 11, 2011, 23:59
I believe a mutant puppy (or maybe a raspberry puppy or some puppy pi?) is a great addition to the list of available distro choices for the RasPi.

It's why I'm working at trying to make it happen. Raspberry Puppy sounds nice...Puppy Pi?? Ewww.

I should be able to get back to more than the initial hacking at it sometime this weekend. I accomplished most of my puttering around to get OE back running and to spin an initial cut of the discussed Raspberry Linux I'd said I'd help make happen back in another discussion thread. I've got a console userland that I suspect will work out of box with one of the kernel images from the bootable SD's out there. The result built cleanly- which is an ARMv6 tuned build. I'll be tinkering with trying to bring it all up via QEMU here shortly.

I'll try, initially, at least, to use Bones to do the deed (It should be easy enough, just change the scripting for the X86 builds to do the pull to see what will pull across, what won't- and what I'll need to do to get it to build the missing binaries...) If it seems...problematic...at this time, I'll do a non-Bones built mutant that uses the packaging list and OpenEmbedded to jam the thing out.

[Jongoleur]

Puppy Crumble?

(Its a Pi, but different... )

[Svartalf]

Quote from Jongoleur on October 13, 2011, 21:12
Puppy Crumble?

(Its a Pi, but different... )



Still sounds like something...unpleasant...done to the poor Puppy...

[Jongoleur]

I know, but have you ever seen a Stargazy pie?
http://londonfood.typepad.com/.....c00015.jpg

Normal service will shortly be resumed.......

[obarthelemy]

OSes are fine and all, and I've played with, and liked, Puppy Linux.
In the end though, I'd rather have one single Linux distro for the Pi, with a lot of optimizations, documentation, packages, and support, than a gaggle of so-so stuff.
As far as I know, there will be one lead distro (Debian ?) and a couple of secondary official ones (Arch and Fedora ?). That seems like plenty to me. Especially on something as constrained as the Pi, which won't leave much room for differentiation anyway.
What's needed the most is ready-to-run, polished packages for software development / earning, fun stuff (emulators, HTPC, hardware hacking...) and semi-useful stuff (office, server, interwebz...). These need to be distro-specific, so the fewer distros, the better.
Let's not fragment the Pi ^^
PS: This is my personal opinion, not the foundation's.

[kme]

Let's not fragment the Pi ^^
Well raspi fans are likes Jews - whereever two meet there'll be three opinions...
I agree that it's important to have a conservative standard distro like Debian (I'm a big Debian fan myself) for raspi, but I also think you can make a much more suited distro directly for the raspi. Debian doesn't use busybox or ulibc and it does use X. Is X really needed for raspi's intended purpose? Wouldn't a framebuffer do - DirectFB? I think there is a huge potential in cutting fat off a standard distro. And yes, it will impose other limitations. That's life.

If you can spend 2½ minute, watch this and wonder:

(note it is an ARM variant). Unfortunately they don't state the size of the result, but I'm willing to bet a pint it's one digit MBs. Oh and you want a text mode editor? How about e3? 13 kB (yes kilobytes, no libs) on x86 and impersonate WordStar (TurboPascal), pico, vi, emacs on most trivial keyboard commands. Obviously not a complete emulation of each editor, but enough for simple editing tasks.

And so you can go on. It makes perfect sense to make a Raspi Linux, but it is a huge task.

EDIT: I wasn't my intension to get the YouTube window - I just pasted an URL. Hope it's OK with you.

[lobster]

We now have sufficient developer interest to announce the Puppy ARM or PARM project which will be developed for the Raspberry Pi
http://puppylinux.org/wikka/PARM

What does Puppy bring to the table?
Wide experience of optimising low resources
Specialised programs - for example PuppyPhone
Rapid development cycle - small focus team - often one or two developers
Fun

[Svartalf]

Wicked... Guess I'm part of the PARM project, then...

[scologic]

sorry folkes i just coulnt resist....

PUPPY POWER!

Fab to hear the Puppy distro is being developed... personally i thought someone would have already been looking at this as it does ply on the vortex x86 machines.

[CharlieDelta]

Hi Lobster

You need to PUSH the Puppy connection. I have seen Puppy evolve over several years and you guys certainly know what you are doing (well at least in x86). Yes you get a lot into a small space and run out of RAM (plug the run out of RAM speed/MIPS) and you have the accumulated EXPERIENCE behind you.

OK you can argue the pros and cons of each different flavour of Linux until the cows come home but that is not the point here. It is knowledge, the "How to" and the understanding that is important. We do not want to reinvent the wheel and with Puppy there have been many "wheels". The objectives of Puppy are the same as those required of the Linux to run on Raspberry Pi. I am sure that there has already been some very valuable work done on different Linux distros and so there will of course be some resistance to adopting Puppy but for my money Puppy is is the best way forward.

BTW I have no connection with Puppy I am just a hardware engineer (spaceship and windmill power systems).

Oooooh and a name for Puppy on Raspberry Pi - how about some derivation of pied Puppy! A good Logo would be "Spot the dog"

I agree with the moderator keep the distros to a minimum but certainly one of them should be Puppy. That way we get to build on firm foundations. With the greatest respect an OS is just an OS; it is just part of the package (just as needed as the RAM or the USB). It is what we build on top of the OS that is the important thing here.

[FredRenner]

RaspupPi?

[grumpyoldgit]

I think the point about Puppy is that it seems to have been designed to run within the confines of a low spec machine and a memory stick. I tried it out a few days ago and it zipped along on my nine year old Dell. You can try it out on your existing PC without installing or changing anything on the hard disk. The backing up process to a memory stick at the end of the session was very well thought out. If you've not seen it in action it is well worth giving it a go with even the most decrepit kit.

[bredman]

Although I am a regular Debian user, I am very impressed with Puppy Linux. In fact I always use Puppy if I need a boot CD for a strange PC.

I expect Puppy to make the RPi fly! It will give the other distros a benchmark to work towards.

My suggestion for the name is PupPi.

[toxibunny]

Out of all the versions of linux I've installed/ran off CD/USB, puppy linux is the only one I've ever been impressed with. This is good news.