Wheezy, Debian 7, aka Maemo Reloaded
 

This thread is continuum of my "Rebasing Fremantle on top of Debian 6" [http://talk.maemo.org/showthread.php?t=89416]

I have reached one milestone and now I'm looking for testers and starting discussion how to utilize all of this.

Here are packages:
http://www.cs.helsinki.fi/u/rantalai.../wheezy/repo2/
http://www.cs.helsinki.fi/u/rantalai.../wheezy/waemo/

(there are also older repository: http://www.cs.helsinki.fi/u/rantalai...5/wheezy/repo/
and here is squeeze packages: http://cc.oulu.fi/~rantalai/maemo5/squeeze/repo/ )

--------------
Motivation:
*Claim1: new version of package is better than old (todo: proof for every package)
*Compiling new code will be easier if libraries and tools are newer.
*It would be possible to install packages straight from Wheezy-armel repository.
--------------
Goal:
(Term package refers to 'binary'-package, which means installable package, not source package, which produce one or more (even tens) binary-packages).

Just installed/flashed Maemo5 has 1745 packages (http://wiki.maemo.org/User:Aapo/Waemo/DefaultPackages), 355 of them are closed (http://wiki.maemo.org/Fremantle_closed_packages) and not scope of this.

Most open packages are made by Nokia and there are no newer Debian versions (but there could be newer upstream or Meego code, but it is not scope of this.)

There are 181 Fremantle packages which are also in Wheezy ( http://wiki.maemo.org/User:Aapo/Waemo/Upgradeable ).
(There are also 26 Fremantle packages in Squeeze but not in Wheezy. These might or might not be in scope of this).


First task for every package is to get it compiled (and its build dependencies). Then installed (and its dependencies) without bricking the device. After that starting to look if it breaks some functionality or if it is causing any side effects (CPU, memory, disk space).

---------
Goal2:
Recompile ~1200 Nokia packages using Wheezy toolchain. (This comes very close to the CSSU and thumb-recompiling.)

This step might be out of this scope, because many package works without recompiling (thanks to good backward compatibility). There are at least one counter example: hildon-application-manager must be compiled against newer libapt (which is now done and it is not bricking the device).

---------
Status on 1 Nov 2013: http://wiki.maemo.org/User:Aapo/Waemo/Status

These packages can be installed without immediate brick: http://www.cs.helsinki.fi/u/rantalai...mo_install.txt
(Note: I have installed top part of the list more often than end of the list (so they are 'more tested'))

You can download script to the /home/user before flashing. After flashing rootfs you need either ssh-server or rootsh installed via Application manager and then just run the script, it will take ~30minutes to run.

Next steps (maybe in parallel):
* Testers: upgrade your device and report what breaks (for sure something breaks, prepare for reflash!)
* Maemo developers: How to use this all (or part) to improve Fremantle?
* How to upgrade hard packages? (hal, udev, gtk, gstreamer)

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Three suggestions:

Goal3: provide a sane building environment to be able for developers to build against wheezy

Goal4: (after goal3) create a maemo-reloaded autobuilder and extras - they would benefit from the newer compiler and newer libs (if statically linked) plus community components which need e.g. libc6 2.10

Goal5: (somewhere inbetween) integrate with CSSU (its qt 4.7.4 and maemo-specific stuff)

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

dammm , you're good at what you're doing

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

ahh, the Neo900 should be running on this foundation...
actually, even android devices should! lol.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

0. When is the step of building flashable image from these packages?
i. e. completely free maemo image, without proprietary nokia packages. And then we'll see how it looks and what can be done, what needs to be rewritten.

1. If I upgrade using your repository, and device get bricked I have no idea how to check what's wrong, which package causes that. If something breaks, but device boots, and it's possible to open a terminal, then yes, something could be done to investigate the problem.

2. Most important. Anyway, I think you are very close to creation of flashable image. And not only for n900, because n900 is a dying hardware, but for any modern device, which allows booting non vendor made operating system. Of course you probably have to start with n900. Ask here or in irc about how to create the image.
I believe you should follow mer strategy and separate the kernel (hardware adaptation) from the rest of the system. So that you can build all maemo packages for different architectures, and in the same time, when preparing an installable image, at this step you unite those packages with vendor specific kernel / hardware adaptation, which is usually non-free because it contains non-free software, drivers, and usually we can do nothing about it.

3. Even active n810 users will benefit from that. For instance, I use n810 every day, and nokia did not provide fremantle image for n810, I believe mostly for marketing purposes. Openmoko community will benefit from that - they could run fremantle like operating system on their devices. Nexus users could benefit, they could run canonical GNU/Linux system instead of their inferior, sorry, androids.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Aapo: did you count the sdk/dev stuff in?

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

How hard would it be to install this system on an μSD card so as not to have to reflash? :-)

I already have Arch Linux ARM there as a backup system.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Look here and here :)
But don't follow it... (no crazy copy&paste please! :D)

PS.: Going to spend my today on that! :)

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Hi,
as far as I understood you start off with PR1.3.1 and then add Debian packages to the mix. Why do you do it that way round? In my understanding "Rebasing Fremantle on top of Debian 6" would mean that you start with Debian 6 and then put Fremantle parts on top of it. That way you start with something that is fully Debian (apart from the kernel and some modules) and diverge from it with fremantle step by step. Your method seems to be prone to never have a "real" Debian system installed because of all the leftover cruft from PR1.3.1 with which you started.

Back in the days I booted Debian on my N900 using instructions from here: http://elektranox.org/n900/index.html

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Aapo: are there any status updates on this? Development halted? Simply too few testers too continue? Or maybe no time / focusing on bringing home the bacon / whatever? Or simply is it as silent as sound wave with amplitude approaching 0 as 1/t for reasonably big t? :)

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Good question. Status is very same than on my last post. After that I haven't compiled anything more for Maemo Reloaded".

Lately I have had less time for Maemo and I haven't used that time for Maemo Reloaded.

I think there must be some bigger plan until I will continue with this.
Maemo5 is outdated, which means there are plenty of new and interesting floss-code we can't utilize. But Maemo5 is not pathologically outdated, there are lots of floss-code (both new and old) which we can use!
(How else you define 'outdated'? I define uptodate="How easily you can compile any random floss-package for it")
-> Issue1: Compiling package for Maemo is harder

There are too few Maemo developers to make miracles. But there are still developers working on Maemo!
-> Issue2: No new packages for Maemo.

I thought that 'merging' Maemo back to the Debian those issues would be solved. Another solution is toss whole Maemo and use pure Debian (People have done much for this, but I think it is not ready). And yet another way is toss Maemo, and use pure Arc (or what not distro).

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

I'll agree with issue 1, for me uptodate="most FOSS apps from an LTS distro will build without complicated tweaks." Using workarounds for a newer debhelper - I can cope with it, but if debhelper is too old to build an app - it's not up to date. If I need to port n libs, but they'll allow themselves to be ported - ok. If the lib is a core lib and can't be upgraded - not up to date.

But I don't think rebasing Maemo will fix the issue 2. Most devs need only Qt4/Qt5/Gtk plus some core libraries. The problems is that too few devs want to develop for Maemo. A major blocker could be Qt5 unavailable in extras, because of lack of extras-thumb

But let's stop for a moment... Do we need some new apps for Maemo? I guess we could find some, but these are not myriads. Mostly these are lacks of particular features, for example thread view in Modest, offline nav in modRana, full-featured voice nav, so big apps. Or a decent browser upgrade. Having the system upgraded would possibly improve their performance, but if no one wants to add threads in modest now, he won't do it when he has libc6 2.15 and glib 2.35. It could ease up porting software - if you can use SparQl for tracker, or the Harmattan signon libraries, it'll be easier for you to port it and there's greater chance you'll proceed to do it.

The problem is how to motivate people to develop the needed software. Unfortunately there's no Maemo Coding Competition. Maybe we should make a list of what we need, make a community fund to pay the dev a couple of beers (after they complete a task! - might be with other's help), in other words - make developing for Maemo attractive. FOSS is great, but you have to have money be able to develop it.

From another point of view, I see no use in throwing Maemo away - who's gonna redo the UI from scratch?

Summing up, rebasing the whole system, in other words upgrading it, won't be a panacea for all problems, but might give a little boost for what's under the hood.

Feel free to add your thoughts, and/or move it into the Brainstorm section.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

only nokia devs can work for maemo? or any developer can?

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

stupid idea - why not make facebook account as maemo or anything and just go on additing developers around the world. i know somehow it will workout. it worth a try.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

seanmcken: Feel free to do it. Anyone can develop for Maemo. And feel free to learn Qt and contribute yourself. (Or you can go for Gtk/Hildon/Cairo if you prefer them)

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Aapo: fmg seems to be working on kernel 3.x booting on Fremantle. Maybe he's the one you could talk to about some bigger plan about Maemo Reloaded.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Yeah... But is anyone also allowed to contribute? Or only those who fit some random criteria? (I seem to remember that having to be available on IRC 24/7 was mentioned at least once. That pretty much rules out about 90% of the already limited set of the potential candidates. Those who work and/or have families, for example.)

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

You mean developing for CSSU? But anyone can try developing an app. If it's a big app - a couple of tmo-ers can cooperate - just like me and Android_808 did with the microb replacement. I've never heard that someone needs to be 24/7 on IRC. About contributing to Maemo Reloaded - you'd have to ask Aapo.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Yes, CSSU. That IRC thing is what I gathered from some heated exchange between two TMOers who may not have been the best of friends (no names :)) so it may have been sexed up a bit but I remember my reaction to that was, "even if only 10% of that is true, there is no way I could possibly offer anywhere near that level of commitment, especially since there is also some learning curve to climb, so I won't even try."

But sorry, you are right, this is OT. I apologize, I misread post #13. As far as this thread's topic is concerned, I'm afraid it is "slightly" over my head, otherwise I would have tried contacting Aapo already.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Well, it's clear that CSSU is for what it is, just (considered) stable updates without (most of the time) disruption of the original Maemo experience. And the way people work on it is up to the developers involved (mostly).

But that doesn't mean that a different (or not?) group of guys can provide alternate upgrades (such as this Debian based packages) on top of that.

If some guys can agreed to set up a new repository (out of Hildon Foundation's servers) with all this stuff so users can just apt-get upgrade to it that would be a great step forward.

Or a repository that actually depends on CSS Thumb for applications that are built with newer toolchain. There's no need to set up an autobuilder, that requires a lot of work and maintenance, just a repo to store all those packages that are randomly thrown in the forums.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

I think a Thumb repo would be great to have on maemo.org... currently Thumb compiled apps are all over the place. Autobuilder would be ideal of course, so that users have at least some assurance that the binary they are installing came from the source code that was uploaded. Failing that, just a simple repo with ability to upload via maemo.org/garage/package interface would be good.

Maybe a bit harder to get setup, but surely not impossible...

If thumb is the "future" for n900 software (and I think it probably is), then this needs to be done at some point.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

I agree with everything you've said - with a small addition, thumb applications are not "future" for N900 software, they're absolutely the present. Every self-respecting developer offers thumb version where applicable, and most of even barely knowledgeable devs and users are utilizing thumb on their devices.

Some people took personal effort to make their programs easily available thumb-compiled - for example, gdizzz host personal repo with thumb versions of all programs he is maintaining, so no more download-link hunting... In this case.

Sadly, those thing works only if individuals do it on their own - none of the (supposedly?) governing bodies took steps (actively seeking volunteers, etc) to update repository system. They're busy "elsewhere" <put a link to any random mail from Maemo's mailing list, where council and board is throwing apesh|t at each other> ...

/Estel

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Er, does that mean that I should be creating thumb versions of my various apps? I had kinda assumed that cpu-bound apps would benefit the most from thumb; but my apps are generally benign. (Even Orecchiette is mostly just a wrapper around calls to GStreamer, so it really depends on that library being recompiled for thumb.)

I guess I should try to put thumb onto one of my N900s too, then. :o I've never really pushed my N900s to the limit, so I've never really worried about it; I play podcasts, watch videos, and edit files in vim just fine without needing to mess with the OS. :)

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

If that's the case then I've been misunderstanding Thumb all this time. My understanding was that Thumb instructions are slightly less efficient than ARM. The real benefit of Thumb is that the compiled binaries are smaller and thus take less memory space and require less swapping. In other words, any given binary will not really benefit from being compiled as Thumb, especially if it is a CPU-intensive one. Thumb is more about being nice to others: I sacrifice a percent or two of my efficiency to make you running smoother (less swapping).

Please correct me if I got it wrong.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Sorry, I shouldn't post while sleepy, you are absolutely right about thumb. :) (Once upon a time, I had educated myself pretty thoroughly on this...)

And yeah, now that I've got an Android device, I guess can use it as my phone and set up both my N900s as test devices -- one for cssu/thumb, one for stock.

I guess it's also time to start trudging back through all that Sandbox stuff, and somehow scrounge up a machine (real or virtual) to run Debian or Ubuntu. Bleah. (Any chance I could just integrate the thumb toolchain into the Qt SDK? ;) )

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

if it weren't for cssu-t to cssu-s testing path I'd suggest merging thumb back into cssu-t as i think most are using it anyway. it is still a possibility, just means eventually thumb packages would land in stable.

if your then looking at "future" maemo, an update to linaro/thumb toolchain would be worth a look. a proper fix to the xlib-xcb issue with hw keyboard from qt5 work so we only need one copy of xlib. trying to port hildon to newer gtk or update qildon, see how much patching of gtk can be avoided along the way because of the tablet style focus of gtk these days.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

But gtk3 != gtk2
And gtk3 could be brought for N900 only if some of the basic libraries were upgraded (don't remember which exactly. glib was one of them, and it was a version distributed only as random debs)
And gtk2 update is what would fall into "Maemo Reloaded"
For the N900's future, I'd vote for tracker upgrade (if it's feasible at all) - our old tracker is sloooow.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

I have read that this is the case, thumb instructions are smaller but maybe less efficient, but in reality, all the thumb applications I've tried including ones I've compiled myself either perform the same, or slighter better AND are smaller.

The better/same performance may be the fact we're using a newer compiler though and not an aspect of thumb itself.

Anyway, the bottom line is, that I can see no reason not to have all applications thumb compiled and I've been using thumb on my device since FMG released it basically.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

That's why I've written 'where applicable". Don't worry, I see you as very self-respecting (and by others) developer :) (not that you should care about my humble opinion, but you know what my point is about, I hope).

From my experience, thumb-compiled memory savings is most important for programs that either run all the time, or have potential to run in background (media player), and most of your programs are rather of a kind that get "full user focus" when running.

OTOH...

...is true, as benefits of updated linaro/thumb compiler manifest itself in code running more efficient on CPU in most cases, too. So, summing it up, if you can, thumb versions for your programs wouldn't hurt (at worst case, it wouldn't provide any improvement, apart from slightly lower size/RAM usage).

If, by any chance, you could do like gdizzz and set up mini-repo that would host all your thumb programs, it would be heaven - this way, user can just define this custom repo in application manager and forget about really tiresome thread-download-link hunting, everytime new version of something is released.

Of course I realize that not everyone have resources (or time, which is most likely case, nowadays) for setting up personal repo, though.

/Estel

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

While it's better than having thumb apps in threads on TMO (I myself am guilty of that), we should not be striving for individual repos, but a central thumb repo, IMO.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Does this talk about Thumb really belong in the Maemo Reloaded thread?

Anyway. All makefiles auto-generated by qmake on my N900 have -mthumb in CFLAGS. In my limited understanding of these things, I interpret that as at least Qt binaries built on the N900 are built as Thumb by default. Am I right? Not sure about binaries built by scratchbox, never managed to set up one successfully.

Re: Wheezy, Debian 7, aka Maemo Reloaded
 

Absolutely true! Unfortunately, I'm pretty sure that it will happen just right after the hell freeze.

With the current state of semi-rogue system administrator, threats of wiping out infra "out of frustration", and power games between board, council etc*, I think we should be actually happy, that anything Maemo-related is working yet, at all.

*For details about all those "cheerish" things, see maemo's mailing list archive...