Creating GSM Users Association (GSMUA)

This is merely a historical archive of years 2008-2021, before the migration to mailman3.

A maintained and still updated list archive can be found at https://lists.osmocom.org/hyperkitty/list/OpenBSC@lists.osmocom.org/.

Mychaela Falconia mychaela.falconia at gmail.com
Sat Jun 10 02:18:37 UTC 2017


thayyil09 yil <thayyil at yandex.com> wrote:

> michela are you kidding ( dont worrry iam a kid at twentees)

Kidding about what?

> calypso bb starts from leaked Docs ( if not sorry )

I don't know what you mean by "calypso bb", but OsmocomBB started from
not only leaked docs, but also leaked *source code* for TI's original
firmware for the chipset in question.  Even though OsmocomBB people
have chosen (for license worship reasons) not to make any direct reuse
of TI's original code beyond a couple of L1 header files, they have
extensively studied those TI firmware source leaks in order to gain
the knowledge of how to operate the Calypso hardware and how to
command the Calypso's DSP ROM to perform the functions it needs to
perform for GSM functionality.  See here for more juicy details
regarding those two OsmocomBB L1 header files I just mentioned:

https://www.freecalypso.org/pipermail/community/2017-April/000361.html

> so why qcom code leak is ugly

Where did I say that it is ugly?  I have not yet had a chance to look
at that QC code you keep talking about - I only downloaded it earlier
today and don't have time currently to look at it in any depth, thus I
am not qualified to judge whether it is ugly or beautiful or somewhere
in between.

> osmocom is for open softwear  not for any specific hardware

I am not a member of Osmocom, instead I lead my own project
(FreeCalypso) in the same general space, hence the arguments regarding
what Osmocom is or is not have no direct applicability to me or to my
non-Osmocom project.

> keeping updated with mainstream hardware is the way

There is no such thing as "the way"; it may be YOUR way, but it is not
*my* way.

> its happy to know your the mother of freecalypso.
> in softwear world you maybe first mom :)

Well, some of us (me included) subscribe to the view that our poor
planet is *way* overpopulated, hence engaging in biological procreation
(imposing more living and resource-consuming humans on the poor planet)
is akin to a crime.  As a result, those of us with maternal instincts
have to find a different outlet for those instincts, such as becoming
the mother of a FLOSS project. :)

Also leaders of FLOSS projects commonly take on titles such as
principal developer, maintainer or even BDFL - but I prefer Mother. :)

> and i here for osmocom on qcom  == osmodroidbb.wordpress.com

I assume you are talking about the QC source leak described on this page:

http://syshwid.blogspot.in/2016/10/build-qualcomm-modem-msm8626.html

(yes, I can read Russian just fine, it's one of my native languages)

When you first posted about it back in April, I went to that page and
the https://drop.me/B439WM mirror it talks about was down - appeared
to have been taken down.  But when I tried it again earlier today
while in the process of composing this response to you, it worked, so
I was finally able to download this QC source you've been talking
about.

However, even if this leak is 100% real source and not a semi-src (the
Russian hacker says he was able to make a complete build from this
source, but it doesn't mean that everything is really rebuilt from
source - there could be major important pieces that are in the form of
linkable binary objects, and it would take a lot of work to analyze
the source to see which is the case), I have my doubts that this
MSM8x26 platform would make a good replacement for FreeCalypso.  I see
the following (potential) problems:

* According to the marketing info that I could easily find for this
  chip(set), it seems to be CDMA/3GPP2-oriented, which is not what I
  am interested in.  That marketing info seems to imply that the chip
  supports UMTS/WCDMA too, but says nothing about GSM support.  I
  principally refuse to use any chip, chipset or device that does not
  support GSM, i.e., I am potentially interested in having support for
  UMTS/WCDMA and maybe even LTE *in addition* to GSM/2G, but never
  instead of.  Furthermore, one of my absolute requirements is to be
  able to invert the preference order for network search, i.e., have
  the modem always preferentially choose GSM/2G networks if such
  service is available, fall back to 3G/UMTS only if no GSM is
  available, and use LTE (or more properly VoLTE only, *without* LTE
  Internet service) only as the last resort when neither 2G nor 3G is
  available.

* This MSM8x26 chip(set) has a ton of extra hardware for making those
  smartphones which I detest: 4 CPU cores, a powerful GPU and gawd
  knows what other crap.  I desire a phone that can only make plain
  old phone calls and absolutely positively nothing else, based on a
  processor that has just enough horsepower to make those plain old
  phone calls and not one iota more, thus having all that extra
  application processor power would be a problem for me.

* Where are the chip datasheets and reference board schematics and so
  forth that would be needed in order to make our own board with the
  chip(set) you are advocating for?  Being able to build my own
  hardware starting from just chips is an absolute requirement for me,
  as I am NOT willing to use an Android slab phone, even it has a
  liberated baseband - I want a phone in my own personally preferred
  form factor (a "dumbphone" *without* Android), hence I need to be
  able to build my own hw in order to package it in the FF I desire.

Overall, it is pretty clear to me that we are on different paths with
very different interests and goals.

M~



More information about the OpenBSC mailing list