Does the current version of OsmocomBB supports
to run the layer 2 or the layer 3 on the baseband
chip, now?
I read through the arxiv of the mailing list but
have maybe overlooked this.
best,
\jo
mark.neuhaus(a)email.de wrote:
> Pleae, then tell me where to get phones with a Calypso
> baseband.
www.ebay.com
> And I'm talking about marked relevant prices.
> Don't waste my time with numbers like 5000$.
Several OsmocomBB-supported models are currently available on ebay for
dirt cheap.
> And to be relevant these phones must still be produced so
> I can get many, if I want.
How many? My company will be happy to produce 5000 Calypso-based
phones for you if you make a non-cancelable purchase order and prepay
the cost of parts and production. (5000 is the number of Calypso&co
chipsets I can buy immediately from my established supplier *without
doing any extensive searching*; one can probably get more with some
effort.)
> Beside, I think its better to work on reverse engineered
> GSM stacks like the Qualcom project as started in
It's easy to tell other people what they should work on. While you
are busy doing that, I'll just keep working on my own solution which
will actually result in a usable phone less than a year from now.
VLR,
SF
Someone (not clear who) said:
> > [...] The calypso is just to outdated to be interesting
> > for anything 'useable' beyond pure hacking-fun.
I violently disagree with that statement. I personally carry a
cellphone for one and only one purpose: so I can call my significant
other (soon to be wife) and she can cell me at any time. A handset
based on the Calypso chipset does this job wonderfully, and is IMO
the optimal tool for the job.
But the problem is that at the present time there does not exist any
cellphone at all, of any kind (dumb, smart, old, new, whatever) that
can make and receive cellular phone calls using only Free Software, as
defined by the Free Software Foundation, i.e., providing the user with
the most essential Four Freedoms:
http://www.gnu.org/philosophy/free-sw.html
Hence I am currently forced to use a cellphone that runs proprietary
firmware, such that I lack the ability to fix any of the UI design
flaws that constantly drive me nuts. This situation causes me severe
distress, hence I have committed my hobby time and cash budget to a
multi-year project to solve this pressing (for me) problem.
OsmocomBB is not a solution: it works wonders for "hacking-fun" (the
wording in the comment I'm responding to), but is utterly useless for
a practical phone which one can carry around in a pocket or purse: my
back just isn't strong enough to carry a supersized backpack containing
a full PC for running the L23 stack plus a bank of lead-acid batteries.
Hence I need a totally different Free Software GSM handset
implementation, one that actually runs on the phone itself with proper
power management exactly like the original proprietary firmware. And
because no one else is working on such a thing, I started my own, and
made quite a bit of progress:
https://bitbucket.org/falconian/freecalypso-sw
But I am using the same Calypso GSM chipset for my project as OsmocomBB
uses, simply because it is IMO the optimal tool for the job at hand:
allowing a person to communicate with his or her significant other by
way of cellular phone calls. The word "outdated" does not exist in my
vocabulary; I evaluate a tool based on how well it does the job, rather
than some arbitrary irrelevant criteria like manufacture date stamps
or whatever.
Viva la Revolucion,
SF
Hi Scott,
On 8/5/14, Scott Weisman <sweisman(a)pobox.com> wrote:
> I think this desire is recognized. I remember an initiative some years
> back, started with some enthusiasm, that never got passed the stage of
> deciding on NuttX as the OS to use, and forking it.
>
We got NuttX running on compal phones (Motorola C1xx, W220, etc),
including display support using NuttX's native graphic subsystem. The
code was submitted to NuttX mainline and now it is integrated.
Unfortunately nobody with more knowledge about GSM L1/L2 layers was
available to help in the stack porting. During the porting we got help
from Steve Markgraf.
Now the scenario is more complex, the original "team" is separated and
we don't have spare time to help.
Best Regards,
Alan
Steve Markgraf <steve(a)steve-m.de> wrote:
> I don't quite see how waving the Free Software flag and distributing
> proprietary source from the TSM30/Locosto leaks (pretty much everything
> in freecalypso-sw/gsm-fw/) go together.
That source is NOT proprietary, it is *ex*-proprietary. It *was*
proprietary, but not any more - it is now in the public domain.
Free Software is not just a "flag" to be waved, it is a practical
reality: in *practical* terms, software that is based on ex-proprietary
code that has been liberated through the exercise of Eminent Domain
and subsequently developed and maintained in the manner of a free sw
project gives its users all of the 4 freedoms defined by FSF.
However, I shall leave it here -- any further replies or comments or
questions in this thread will *not* elicit a further reply from me.
But if there is anyone else on this list who desires a usable cellphone
for talking to his or her significant other like I do, please be assured
that I have no plans of dropping the project; the work is progressing
at a steady pace as you can see from the Mercurial commit history, and
I have high hopes of some actually-usable result some time around the
end of 2014.
VLR,
SF
The kernel comes with a hypervisor (under BSD on Githu, too)
and you can run Linux in one virtual environment
(see http://l4linux.org/) and for example an GSM stack in
another virtual environment. So you can think of that kernel
as a kind of virtual machine that seperates the hardware
This way all virtual systems are prooven separated and I think
that is what the merkel-phone does.
owever NICTA is actually going much further. Thei are working
on a zero-bug (verified file system) and much more.
This kind of programming is entirely different from common
way to write a program. (YOU don't write the program, but you
write a proof in Coq or ISABEL ect and from this proof a
programm can be derived automatically)
> Gesendet: Dienstag, 05. August 2014 um 10:50 Uhr
> Von: "Sebastien Lorquet" <sebastien(a)lorquet.fr>
> An: mark.neuhaus(a)email.de
> Betreff: Re: Aw: Re: Re: Re: seL4 is open source now
>
> Hello,
>
> Reading through the se4l FAQ at http://sel4.systems/FAQ/ , I'm reading this:
>
>
> Unique about seL4 is its unprecedented degree of assurance, achieved through
> formal verification. Specifically, the ARM version of seL4 is the first (and
> still only) general-purpose OS kernel with a full code-level functional
> correctness proof, meaning a mathematical proof that the implementation (written
> in C) adheres to its specification. In short, the implementation is proved to be
> bug-free (see below). This also implies a number of other properties, such as
> freedom from buffer overflows, null pointer exceptions, use-after-free, etc.
>
>
> Okay with that. But also:
>
>
> There is a further proof that the binary code which executes on the hardware is
> a correct translation of the C code. This means that the compiler does not have
> to be trusted, and extends the functional correctness property to the binary.
>
>
> So that is very cool, even the binary result is proved, not only the source code !
>
> The only thing that bothers me is: what can I do with this cool software? It
> seems that little can be done without a set of userspace servers, so I guess we
> have to wait until some open source community provides useful software to run
> with this microkernel.
>
> Best regards,
>
> Sébastien Lorquet
>
> Le 02/08/2014 12:49, mark.neuhaus(a)email.de a écrit :
> >
> >> My idea is that at some point, the developers working at xda-developers
> >> (android phone hackers) will get enough knowledge of their baseband
> >> chips so that something can be done with osmocom. But this is only my
> >> opinion!
> >
> > Yes lets hope. The calypso is just to outdated to be interesting
> > for anything 'useable' beyond pure hacking-fun.
> >
>