baseband-devel

baseband-devel@lists.osmocom.org

2022 discussions

T191 unlock cable
by R M 24 May '11

24 May '11

Hi This is my first mail to this list. I have downloaded the code and compiled it. My development environment is Eclipse using cygwin as I am a Windows user. I have succeeded in procuring Motorola C115 phone and also T191 unlock cable but the cable has a serial port on one end and 2.5 mm audio jack at the other. My laptop doesn't have a serial port. It only has USB ports. What sort of USB to serial converter should I use? I have not enquired the cost of FDTI cables in India. But they seem to cost a fortune online. Please do suggest one that is cheap say within 200 rupees. I am also having difficulty downloading Iota and Rita data sheets as the links provided in the wiki lead to chinese sites and I don't understand chinese. Can some one who has already downloaded them please tell me the process like which buttons I need to click etc. Regards, RM

2 3

GSMTAP now creates a locally bound receive socket
by Harald Welte 24 May '11

24 May '11

Hi! libosmocore >= 0.3.1 and the current osmocom-bb code will now generate not only the sending GSMTAP UDP socket, but also a locally-bound receive socket. This avoids the manual start of "nc -l -u -p 4729 >/dev/null" or iptables rules to drop the UDP packets. The local receive socket is only created if the GSMTAP IP address is a locally configured address on any of your network interfaces. So sending it to 127.0.0.1 should work well. Don't be surprised if you happen to see GSMTAP over IPv6, I'm now using getaddrinfo() and related functions, i.e. "loopback" may now resolve to ::1 instead of 127.0.0.1 Regards, Harald -- - Harald Welte <laforge(a)gnumonks.org> http://laforge.gnumonks.org/ ============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6)

2 2

Re: Anybody using a BTS for debugging?
by Mohammad Hosein 24 May '11

24 May '11

most of the test systems are not built to be used for on-air operations . the ones that offer such features must be used in Shield rooms where no signal gets in or out . these rooms are expensive but usually tech universities have some rental hours for their shield and antenna rooms . its expensive as well , otherwise you are probably going to have some conflict with local regulatory sooner or later regards On Thu, May 19, 2011 at 9:04 PM, Gianni Tedesco <gianni(a)scaramanga.co.uk>wrote: > On Fri, 2011-05-13 at 15:00 +0430, Mohammad Hosein wrote: > > you could use most GSM test systems for such purpose . on ebay you > > might find very cheap ones > > regards > > I managed to track down a Racal 6103. It should arrive in next couple of > days. > > As a followup question then, I guess that these are intended for wired > rf connection or at least they transmit at very low power, and tell the > phone to do the same. > > However, one would feel a lot better using a channel unused by other > cells in my location. I've been using cell_log to put together a list of > active ARFCN's that my handset receives broadcasts from. Script for > parsing cell_log logs is attached. > > Am I to conclude that if cell_log hasn't dumped: > > [sysinfo] > arfcn 99 > > for example, after running for a reasonably long period of time (say > 24hrs) that operating test equipment on channel 99 should be safe? > > Gianni > > -- > > #!/usr/bin/python > > def getline(f): > while True: > l = f.readline() > if not l: > raise StopIteration > l = l.rstrip('\r\n') > if not l: > continue > yield l > def bsic(v): > (a,b) = v.split(',') > return (int(a),int(b)) > > def hexdump(v): > ret = bytearray() > for x in v.split(): > ret.append(int(x, 16)) > return ret > > class Parser: > pass > > class Power(Parser): > def line(self, s): > return > > class mcc: > tbl = {0x234: 'UK and Northern Ireland'} > def __init__(self, num): > x = ((num & 0xf00) >> 8, > (num & 0xf000) >> 12, > (num & 0xf)) > self.val = (x[0] << 8) | x[1] << 4 | x[2] > def __int__(self): > return self.val > def __str__(self): > return self.tbl.get(self.val, '%.3x'%self.val) > > class mnc: > tbl = {0x15: 'Vodafone', > 0x30: 'T-Mobile', > 0x10: 'O2', > 0x33: 'Orange', > } > def __init__(self, num): > x = ((num & 0xf0) >> 4, > (num & 0xf)) > self.val = (x[1] << 4) | x[0] > def __int__(self): > return self.val > def __str__(self): > return self.tbl.get(self.val, '%.2x'%self.val) > > class SystemInfo: > pass > > class SystemInfo3(SystemInfo): > def __init__(self, b): > if isinstance(b, str): > b = hexdump(b) > assert(b[0] > 9) # len > assert(b[1] == 0x06) # discr > assert(b[2] == 0x1b) # sysinfo3 > self.cell = (b[3] << 8) | b[4] > self.mcc = mcc((b[5] << 8) | b[6]) > self.mnc = mnc(b[7]) > self.lac = (b[8] << 8) | b[9] > def __str__(self): > return 'Cell(%s, %s)'%(self.mcc,self.mnc) > > class SystemInfo4(SystemInfo): > def __init__(self, b): > if isinstance(b, str): > b = hexdump(b) > assert(b[0] > 7) # len > assert(b[1] == 0x06) # discr > assert(b[2] == 0x1c) # sysinfo4 > self.cell = None > self.mcc = mcc((b[3] << 8) | b[4]) > self.mnc = mnc(b[5]) > self.lac = (b[6] << 8) | b[7] > def __str__(self): > return 'Cell(%s, %s)'%(self.mcc,self.mnc) > > class Sysinfo(Parser): > _keys = {'arfcn':int, > 'time':int, > 'rxlev':int, > 'bsic':bsic, > 'si1':hexdump, > 'si2':hexdump, > 'si2bis':hexdump, > 'si2ter':hexdump, > 'si3':SystemInfo3, > 'si4':SystemInfo4, > 'ta':int} > def line(self, s): > (k,v) = s.split(None, 1) > if self._keys.has_key(k): > setattr(self, k, self._keys[k](v)) > else: > print k, v > return > > class RFMap: > arfcns = {} > cells = {} > def commit(self, obj): > if isinstance(obj, Sysinfo): > self.arfcns.setdefault(obj.arfcn, {}) > self.arfcns[obj.arfcn][str(obj.si3.mnc)] = None > self.arfcns[obj.arfcn][str(obj.si4.mnc)] = None > > self.cells.setdefault(obj.si3.cell, {}) > self.cells[obj.si3.cell][str(obj.si3.mnc)] = None > elif isinstance(obj, Power): > return > assert(isinstance(obj, Parser)) > > def rf_map(f): > smap = {'[power]':Power, '[sysinfo]':Sysinfo} > obj = None > m = RFMap() > > for x in getline(f): > if x[0] == '[': > if obj is not None: > m.commit(obj) > obj = smap[x]() > continue > if obj is None: > raise ValueError, 'Bad state' > obj.line(x) > > if obj is not None: > m.commit(obj) > > print 'Active ARFCNs:' > x = m.arfcns.keys() > x.sort() > for i in x: > print ' %4d:'%i, ','.join(m.arfcns[i]) > print > > print 'Active Cells:' > x = m.cells.keys() > x.sort() > for i in x: > print ' 0x%.4x'%i, ','.join(m.cells[i]) > print > > def main(argv): > if len(argv) > 1: > infile = open(argv[1]) > else: > from sys import stdin > infile = stdin > > rf_map(infile) > return 0 > > if __name__ == '__main__': > from sys import argv > raise SystemExit, main(argv) > > >

4 10

Nuttx for Calypso
by l--putt 24 May '11

24 May '11

Hi list! As you can see from the patch, I made some progress with a port of Nuttx to the Calypso platform. The UART doesn't work yet. The few possible tests with just the backlight suggest that the IRQ is working. More work required but I'd like to avoid doing the work twice. Stefan

7 11

about voice
by shrek wu 23 May '11

23 May '11

hi all, About voice，I have one question. In the latest git code, when I use bb to call other phone, it will product a file voice.raw. How I can play the voice.raw? which kind of format? 16bit or 8 bit pcm?16khz and 8khz?LSE or MSE? In fact i used the cooledit to play it, both 16khz and 8khz, it has only noise. Other question, I want to input voice from pc that run BB, then the voice can be sent by C123. But I found the voice that be sent by C123 is C123 input voice. Can we realize the idea? Best Regards Shrek W

10 15

Strange error
by Sun Ich 23 May '11

23 May '11

Hi When I try to load any program from firmware/board/compal_e88/ to my C118 phone, I receive error below: ioctl(TIOCGSERIAL): Invalid argument I upload program successfully and I receive DOWNLOAD ACK from phone properly, but then I receive this strange ioctl error. I'm running Ubuntu 9 and I tried to load compal_dsp_dump.compalram.bin layer1.compalram.bin loader.compalram.bin Result is same for all of the above. Please advice. Thanks

2 1

[PATCH 0/6] osmocom-bb: get in sync with libosmocore namespace changes
by pablo＠gnumonks.org 22 May '11

22 May '11

From: Pablo Neira Ayuso <pablo(a)gnumonks.org> This patchset gets osmocom-bb in sync with the namespace fixes for libosmocore. You can find them in pablo/namespace branch. Please, merge it. BTW: I had to temporarily disable mtk62xx compilation since I hit a linking error. This problem occurs in the current git tree. I already contacted Wolfram to let him know. diff --git a/src/Makefile b/src/Makefile index d6f556f..6ff4dc4 100644 --- a/src/Makefile +++ b/src/Makefile @@ -17,7 +17,7 @@ OSMOCORE_CONFIGURE_ENV= LIBOSMOCORE_LIBS=$(TOPDIR)/shared/libosmocore/build-host LIBOSMOVTY_CFLAGS=-I$(TOPDIR)/shared/libosmocore/include \ LIBOSMOGSM_CFLAGS=-I$(TOPDIR)/shared/libosmocore/include -all: libosmocore-target nofirmware firmware mtk-firmware +all: libosmocore-target nofirmware firmware nofirmware: libosmocore-host layer23 osmocon gsmmap libosmocore-host: shared/libosmocore/build-host/src/.libs/libosmocore.la Pablo Neira Ayuso (6): src: use namespace prefix osmo_timer* src: use namespace prefix osmo_fd* and osmo_select* src: use namespace prefix osmo_signal* src: use namespace prefix osmo_wqueue* src: use namespace prefix osmo_* for utils src: use namespace prefix osmo_* for crc16 functions

4 10

Re: hmm, sorry, data cable question
by labad jala 22 May '11

22 May '11

Hi again techies. Thanks for your advises and schemas. I already bought what was announced and illustrated as a mot t191 cable at a french company but it turned to be a "flash cable" with possibly wrong pinout and voltage. If you look at http://www.cellcorner.com/xshp/unlock-phone-codes/motorola-t190-t191-t193-u…, it's looks exactly the same one as I have. Did someone buy this one and used it successfully? I measured 4 volts when putting positive pin of my voltmeter to rx and negative pin on ground of the 2.5mm jack.. As said before, don't want to fry my moto. No cable, no project:( Sorry for my noobie question on this high level mailing list...

1 0

http git for nuttx-bb
by Sébastien Lorquet 19 May '11

19 May '11

Hello, I installed a git mirror with http access at http://repo.or.cz/r/nuttx-bb.git Regards Sebastien

1 0

Re: RFC: RTOS for Calypso
by Gregory Nutt 19 May '11

19 May '11

Hi, List, I heard through the grape vine that you are considering using NuttX. I don't want to influence your decision, but if the decision is final, I would like to offer any support to the project that I can give. > Indeed. The TI C5471 DSP/ARM chip seems to be virtually a predecessor of the Calypso. I have quite a bit of experience working with these TI dual core chips and would be happy to help out as much as possible. With the ARM7/c5 family, I have developed OS support (uClinux) for the C5471, DSC21, DSC24, and DM270. Also the DM320 (ARM9/c5, Linux). I founded this business in Costa Rica (http://www.ridgerun.com, also once called Cadenux). But have not been involved with the more recent ARM11/Cortex A8/c6 chips. > I'm already working on the IRQ subsystem and took a look at the UART. > But there are open questions, see above... There is a porting guide here may be helpful to you: http://nuttx.sourceforge.net/NuttxPortingGuide.html. But it does a some "To be provided" sections -- particulary in the IRQ section. > I have a c5 DSP bridge driver somewhere around on my backup disks as > well. I never released it into NuttX because there was never any > need. Perhaps that could be useful now. These DSP bridge drivers present interfaces to load programs into the DSP and to support an integrated, seamless messaging system between the ARM7 and the DSP. The DSP interfaces differ from chip to chip and the drivers that I was referring were targeted for uClinux. However, if that is something you are interested in, I would be happy to do the port a driver and integrate DSP messaging with the NuttX messaging. Greg

2 1

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

baseband-devel