Holger Hans Peter Freyther
c2d3e45571
sysmobts: The array size for the clocksources has increased to 10.
2012-07-20 15:30:06 +02:00
Holger Hans Peter Freyther
27baa4c3de
sysmobts-calib: Add support for pre-production revb hardware
...
The board version wasn't exposed in the revb DSP interface.
2012-07-20 15:30:06 +02:00
Holger Hans Peter Freyther
b3eb6da2db
misc: Quote the warning to avoid additional warning
2012-07-20 15:30:06 +02:00
Andreas Eversberg
d40d4d6071
Allow L1 forward proxy to provide all 4 queues to seperate applications
...
Different applications can now connect to L1 forward proxy or access DSP
directly, if they use different message queues.
2012-07-19 20:33:37 +02:00
Andreas Eversberg
08fce19cfc
Allocate correct message size for L1 primitives
...
This is required for using firmware v2.4
2012-07-19 20:29:56 +02:00
Andreas Eversberg
0390d54ade
logging: Fixed order of logging categories in enum list
...
The enum list must have the same order as the logging description
structure. Otherwiese libosmocore will crash when writing loglevels at
VTY.
2012-07-16 18:50:55 +02:00
Andreas Eversberg
0c470759da
PCU: Add verion number of PCU interface to PCU INFO IND message
...
The client (PCU) can check if it is compiled with a different version.
2012-07-16 18:50:26 +02:00
Andreas Eversberg
5a53eff4cb
sysmobts L1: fix memory leaks for GPRS
...
we have to hand off the PH-RTS.ind to the PCU interface _before_
we allocate a response msgb/primitive.
2012-07-16 18:48:55 +02:00
Andreas Eversberg
990d1da8a4
PCU interface: fix memory leaks in error paths
2012-07-16 18:48:37 +02:00
Holger Hans Peter Freyther
65d4d5108a
calib: Create a new header file and move it.
2012-07-12 09:08:13 +02:00
Holger Hans Peter Freyther
0cfefa0e12
calib: Add code to change the BSIC/TSC before following the BCCH.
2012-07-12 09:08:13 +02:00
Holger Hans Peter Freyther
4253150bab
calib: Use base 16 encoding for the dsp trace flags
2012-07-11 23:14:35 +02:00
Harald Welte
38420fb951
add new sysmobst-mgr daemon
...
This daemon is taking care of counting the number of hours in operation
and to watch the system temperature as determined by internal
temperature sensors.
Later, it will export an external interface for firmware reload, as well
as a way to raise OML ALARMs in case of temperature issues or other
problems.
2012-07-11 01:32:42 +02:00
Harald Welte
3696c6946d
OML: add missing ntohs() for UL/DL_TBF_EXT
...
Thanks to Andreas for spotting this.
2012-07-11 01:26:32 +02:00
Harald Welte
438a28714d
l1_if: skip processing of measurement results on PDTCH
...
In case of PDTCH, the PCU has to process measurements, not the BTS.
2012-07-09 15:51:42 +02:00
Harald Welte
c1368d4ebe
PCU: remove german warnings from the code
2012-07-08 23:53:32 +02:00
Andreas Eversberg
744f745d7a
PCU: Add PCU socket interface to BTS.
...
A special command line option "-P" is used to enable socket interface
and signal available GPRS MO object to BSC.
2012-07-08 20:50:02 +02:00
Andreas Eversberg
8169b0bd85
Add BTS to list at the beginning of bts_init()
...
During init process, signals might be sent. PCU receives these signals and
requires that BTS instance is already in the list.
2012-07-08 20:20:51 +02:00
Andreas Eversberg
07b37853a4
PCU: Add PCU socket interface prototype header file
2012-07-08 19:59:41 +02:00
Andreas Eversberg
bf2a18e623
debug: Add new debugging class for PCU interface (DPCU)
2012-07-08 19:41:41 +02:00
Andreas Eversberg
66f1fe15e9
signal: Add signals for setting/change of GPRS MO attributes
2012-07-08 19:38:39 +02:00
Andreas Eversberg
07891a0908
paging: Alow to store CCCH messages in paging records
...
This is required for PCU to send IMMEDIATE ASSIGNMENT messages on PCH.
A message in a paging record is sent only once.
2012-07-08 18:55:45 +02:00
Andreas Eversberg
343cae60b6
lchan: Activate PTCCH/PRACH/PDTCH/PACCH when activating PDCH
2012-07-08 18:48:58 +02:00
Andreas Eversberg
ea15101896
Fixed check for RACH (random access) delay
2012-07-08 18:44:40 +02:00
Andreas Eversberg
b57e17394b
Fixes for handling of GPRS NSE/NSVC/CELL MO
2012-07-08 18:03:04 +02:00
Holger Hans Peter Freyther
b19592f713
paging: Update the unit test that would have caused the previous
...
The unit test created and used the paging request in the same
second and was passing because of that. Add a second test with
a delay to force now to not be equal to the expiration time.
2012-07-05 23:31:18 +02:00
Holger Hans Peter Freyther
cb7697074e
paging: Expire paging requests after the expiration time
...
The paging needs to expire when the expiration time is smaller
than the current time.
2012-07-05 23:15:02 +02:00
Harald Welte
71b216d995
l1_if: don't enable any GSMTAP by default
...
the user has to explicitly enable it in the VTY
2012-07-05 15:31:06 +02:00
Harald Welte
d53ae2d0f1
sysmobts_vty: Fix setting GSMTAP sapi, as well as save/restore in cfg
2012-07-05 15:31:05 +02:00
Harald Welte
5f8a3149fe
sysmobts: avoid sending duplicate RSL CHAN ACT ACK
...
This is just an intermediary hack, until we get proper lchan manager
threads...
2012-07-05 15:31:05 +02:00
Holger Hans Peter Freyther
eda6c26360
calib: Add a mode to follow the BCCH of a given cell.
2012-07-05 00:17:02 +02:00
Harald Welte
6b561bb7ba
Add 12.21 handling for GPRS NSE/NSVC/CELL MO
...
We now bring the GPRS related MO up in DEPENDENCY state and parse
the various NS, BSSGP and RLC parameters as set by the BSC via 12.21/OML.
2012-06-28 08:59:48 +02:00
Harald Welte
fa8014f181
make sure we don't send CCCH LOAD IND before we have an Abis link
2012-06-21 16:46:05 +02:00
Harald Welte
61fb64d252
rsl: use correct headroom size for load indications
2012-06-18 22:17:28 +08:00
Harald Welte
54b8af0f64
use default value of 63 for maximum timing advance
...
As the careful commitlog reader Andreas points out: When the BSC does
not sent NM_ATT_MAX_TA, then it would be zero instead of the specified
default value of 63.
2012-06-15 23:24:55 +08:00
Holger Hans Peter Freyther
9fdefc6ffe
respawn: The BTS should not be nice, make sure the BTS is the most favorable
2012-06-15 14:53:15 +02:00
Harald Welte
13e92be8bf
Implement NM_ATT_MAX_TA in sysmobts backend
2012-06-15 14:54:33 +08:00
Harald Welte
e01a47aad4
Update README and reflect that we now have CCCH LOAD IND
2012-06-15 11:17:15 +08:00
Harald Welte
babbbbf6ee
CCCH LOAD IND: Avoid divide-by-zero
...
The total count of RACH or PCH slots should never be zero, as they
constantly increment. However, just as a safeguard, we introduce
an explicit handign to avoid divide-by-zero situations
2012-06-15 11:15:11 +08:00
Harald Welte
821bf067e4
RSL: Add CCCH LOAD INDICATION for RACH
...
We now count the total number of RACH slots, the number with rx level
above the busy threshold, and the number of valid access bursts.
This data is used to generate RSL CCCH LOAD INDICATION for the RACH.
2012-06-15 11:07:03 +08:00
Harald Welte
c882b85d8c
system information: avoid modulo 0 / SIGFPE
...
As Holger pointed out, it may well be the case that there are no system
information messages to be sent at TC=4, and we should avoid a modulo by
0. I'm simply sending SI2 instead now, as it isn't forbidden to send it
more often than the minimum at TC=2...
2012-06-14 11:51:16 +08:00
Harald Welte
565cf0d8ab
attempt to make CCCH Load Indications for PCH work
2012-06-14 11:48:24 +08:00
Holger Hans Peter Freyther
a540332df3
sysmobts-calib: Add a utility to calibrate the sysmobts v2 hardware
...
It has been tested with the OCXO and the network listen mode of the
firmware. For other sources we are not required to synchronize to
the network and the tool needs to be adjusted.
2012-06-12 18:17:11 +02:00
Holger Hans Peter Freyther
ad3e31dc4b
sysmobts: The meaning of the clock value changed from v1 to v2
...
In v2 the calibration value is the clock error in ppb that needs
to be compensated. Create a V2 specific implementation. Write the
clock value unconditionally as it is initialized to 0 by default
and not 0xffff.
2012-06-05 09:34:33 +02:00
Holger Hans Peter Freyther
1c069cd0a0
sysmobts-v1: Fix compilation by using the right define
2012-06-05 09:17:26 +02:00
Harald Welte
0455e51cd5
Use git-generated PACKAGE_VERSION in IPA IDTAG_SWVERSION
...
We previously used to send the bogus string "0815" which was a hack
from early development time, but is obviously not a generally useful
idea.
2012-06-03 11:47:42 +02:00
Harald Welte
ad09615acb
add known limitations
2012-06-03 11:01:31 +02:00
Harald Welte
2100a2e16f
sysinfo: Make our SI scheduling more complete
...
We now implement the fairly complex rules for schedulign of
SI 2bis/2ter/2quater, 13 and 9 on TC=4 and TC=5 of the BCCH Norm.
The patch is currently untested.
2012-06-03 07:37:52 +02:00
Harald Welte
c58968be02
sysinfo: Schedule SI 2bis and 2ter
...
In case we have neighbor cells in different bands, we should send those
SI...
2012-06-02 22:20:58 +02:00
Harald Welte
799ea59c2f
sysmobts: set the RF ACTIVE LED when we bring RF up
...
Once we get RF-ACTIVATE.conf from L1, we now enable the corresponding
LED. We also switch it off on RF-DEACTIVATE.conf. We do _not_ switch
it off when osmo-bts crashes or terminates before RF-DEACTIVATE.conf.
The latter is intentional, as RF may very well still be active at that
point. The re-spawning script will re-set the DSP and therby turn off
the RF and then disable the LED.
A better solution might be to do all this in the kernel driver for the
DSP.
2012-06-01 00:06:58 +02:00