aboutsummaryrefslogtreecommitdiffstats
path: root/openbsc
AgeCommit message (Collapse)AuthorFilesLines
2010-03-29[misc] Remove whitespace from the end of the line.Holger Hans Peter Freyther17-57/+57
2010-03-28remove gsm48_mi_to_string() as it is now in libosmocore 0.1.3Harald Welte3-47/+1
2010-03-28chan_alloc: Support allocating TCH/F of a dynamic TCH/F + PDCHHarald Welte2-1/+34
This code simply enables the channel allocator to understand the dynamic TCH/F / PDCH channel type as used by ip.access nanoBTS. It does not actually try to switch the dynamic mode, but instead sends signals to a (not yet present) dynamic switching algorithm.
2010-03-28RSL: keep track of ip.access dynamic TCH/PDCH activationHarald Welte3-5/+14
We use the (currently unusued) flags member of the bts_trx_ts structure to track if a dynamic TCH/PDCH is currently on PDCH mode or not.
2010-03-26debug: ensure no overlap between LOG_FILTER_ALL and FLT_IMSIHarald Welte1-1/+2
2010-03-26move log/debug codebase to libosmocoreHarald Welte14-529/+210
The logging/debugging code is generic enough to move it into libosmocore while keeping OpenBSC specific definitions in openbsc itself. This commit uses the logging support present in libosmocore-0.1.2, you will have to update your library to this version.
2010-03-25db: Fix a bug where no pending SMS were foundHolger Hans Peter Freyther2-4/+4
The "sms send pending" VTY command did not work due a mismatch of types. We are specifying a unsigned long long in the query and provided DBI with a signed integer type. The result was a failure do find any information. Change the API to operate on unsigned long long that is matching the id of the SMS and the Subscriber and the mismatch with the query string is gone and pending SMS are sent.
2010-03-25sms: Remove some more occurences of the GSM lchanHolger Hans Peter Freyther1-4/+2
2010-03-25Store the GSM BTS in the gsm subscriber connectionHolger Hans Peter Freyther3-8/+9
Remove further usage of lchan from the gsm 04.11 bits
2010-03-25bsc: Start creating 08.08 like APIHolger Hans Peter Freyther5-14/+45
The 08.08 API will interface with the internal BSC code and it is the boundary between MSC and BSC. So nothing that calls the BSC functionality should know about lchan or such.
2010-03-25sms: First run of removing lchan usage from MSC code paths.Holger Hans Peter Freyther2-12/+20
2010-03-25lchan: Change transaction to work on the GSM Subscriber ConnectionHolger Hans Peter Freyther7-63/+56
Change the MSC transaction code to operate on a GSM Subscriber Connection instead of the lchan. This will help us to separate the two commands properly.
2010-03-25lchan: Create a structure for MSC data of the lchanHolger Hans Peter Freyther13-154/+203
Prepare to split the BSC and the MSC part by putting the MSC data for a connection into a "gsm_subscriber_connection" struct and renaming the macros.
2010-03-25convert openbsc to use libosmocore-0.1.1 APIHarald Welte4-7/+8
the 0.1.1 API is cleaned up and removes all exported global static arrays (like rlm_cause_strs). There are now proper accessor functions.
2010-03-25current version needs exactly 0.1.0 version of libosmocoreHarald Welte1-1/+1
2010-03-25increase the use of 'struct value_string' in abis_nm and gsm_dataHarald Welte2-247/+183
2010-03-25debug: Make the global target loglevel overwrite the category levelHolger Hans Peter Freyther1-1/+1
Together with the previos -e Number option one can easily turn on debugging without needing to remember the category masks for a quick check on what is going on.
2010-03-25bsc_hack: Add other known options to the listHolger Hans Peter Freyther1-1/+3
2010-03-25bsc_hack: Add option to set the global log level.Holger Hans Peter Freyther1-1/+6
2010-03-24rll_establish: When on a TCH always use the SACH for SAPI!=0Holger Hans Peter Freyther1-2/+1
When we are on a TCH/H or TCH/F always use the SACH for SAPI!=0 for the establishment otherwise it will never be answered. This can be tested by starting to page with a traffic channel and then trying to submit the SMS.
2010-03-24ipa: Fix the previous patch and initialize the fds to -1Holger Hans Peter Freyther1-0/+6
The code in 354ef81d8062510765b7280023b3f4ef98d0613a checked for fd >= 0 but on startup the struct is memset to 0 so this test is true. Initialize the fds to -1 to make the code work and be able to handle all ranges of the fd.
2010-03-24Move the version/copyright string to a separate fileHolger Hans Peter Freyther3-15/+33
It didn't really belong into the bsc_init.c... now we could even easily autogenerate this file.
2010-03-24Make the compiler happy and use a putsHolger Hans Peter Freyther1-1/+1
2010-03-24Include bscconfig.h without ../ for the srcdir != build dir caseHolger Hans Peter Freyther2-2/+2
2010-03-24Move the copyright and version header to bsc_init.cHolger Hans Peter Freyther2-10/+16
The copyright/version will be used from within the VTY code which might be linked into binaries even when bsc_hack.c is not linked into it.
2010-03-24ipa: Handle corrupt incoming messages without crashingHolger Hans Peter Freyther1-6/+14
Attempt to read the three byte IPA header. If we read 0 then the socket can be closed otherwise we need to read three bytes, if we don't we do have a failure. After having parsed the header we need to evaluate the length, if the length is bigger than would fit into our buffer we will ignore that and print an error. This is fixing a crash when the BTS is crashing...
2010-03-24ipa: Reduce the throttling of the IPA msgesHolger Hans Peter Freyther1-1/+1
This code used to be a sleep, it was changed to be a timer by Andreas but this timer does not seem to have any use. When doing the sw load this timer is increasing the upload time dramatically, reduce it to make it work faster.
2010-03-24ipa: Make it possible to specify -w and -f in any orderHolger Hans Peter Freyther1-3/+9
Currently one has to put -w in front of -f to really write the firmware file to disk. Change the config handling to first take all parameters and then execute on them. This means -f can now be used with any other parameter.
2010-03-24ipa: Use two defines instead of magic numbersHolger Hans Peter Freyther1-8/+11
We spearate OML/RSL by putting a magic number into the priv_nr. Use a define instead of ordinary integers.
2010-03-24ipaccess: Handle the case of replacing RSL connectionsHolger Hans Peter Freyther1-0/+7
Fix a infinite loop when establishing a new RSL connection and the BSC is identifying itself with a unit id of an already established RSL connection. The infinite loops happens because we are corrupting the the linuxlist inside the bsc_fd when registering the bfd twice. Due the lack of proper authentication favor the new RSL connection as the real one and close the previous one.
2010-03-23depend on libosmocore 0.1.0Harald Welte1-1/+1
this creates a hard dependency to a certain base version of the library
2010-03-23automatically include program version and print it from vty and --versionHarald Welte8-19/+221
2010-03-22gprs: remove (the already unused) si13_templateHarald Welte1-16/+0
system_information.c and rest_octets.c don't need a template...
2010-03-22GPRS: make NSEI configurable from VTYHarald Welte3-0/+25
2010-03-22GPRS: don't allow VTY configuration of GPRS parameters if GPRS inactiveHarald Welte1-0/+30
2010-03-22GPRS: remove outdated FIXME about patching CGIHarald Welte1-1/+0
2010-03-22GPRS: set SI3 GPRS indicator dependent on bts->gprs.enabledHarald Welte1-1/+3
2010-03-22GPRS: remove "#define GPRS" compile-time defineHarald Welte1-16/+14
It is now possible to enable/disable GPRS support purely based on the vty / config file.
2010-03-22GPRS: make NSVC parameters VTY-configurableHarald Welte3-10/+88
2010-03-22Create new gprs-conf branch with the non-SGSN part of the gprs branchHarald Welte8-22/+270
This new gprs-conf branch is intended to contain everything needed to configure GPRS in the nanoBTS, but without implementing the SGSN/GGSN functionality. The SGSN/GGSN development will happen in a branch based on this branch called "gprs-sgsn"
2010-03-22ipa-firmware: Fix the write out of the firmware partsHolger Hans Peter Freyther2-2/+2
Use the start address inside the header entry, the start is relative to the surrounding SDP record which is located in our base offset, when writing it out also ignore four bytes of something (crc?).
2010-03-22ipa-firmware: Write the firmware parts to a fileHolger Hans Peter Freyther3-5/+61
The first non working version to dump the content of the sdp to the terminal..
2010-03-22ipa-firmware: Add the entry first and then seekHolger Hans Peter Freyther1-3/+4
The sub entry might not be a SDP file so it is best to add the read entry, add it to the list and then attempt to recursively parse this one.
2010-03-22ipa-firmware: Lift the requirement of having a table_offset of zeroHolger Hans Peter Freyther1-8/+4
Make the code work/skip the other table and find the entries at the given offset. This appears to work.
2010-03-22ipa-firmware: Remove the part_length from the struct sdp_firmwareHolger Hans Peter Freyther2-7/+21
* The length of the table is not at a fixed position. We will need to read the offset, seek there, read the data, convert it to the host endianes. * Prepare the code to work with offsets of 0...
2010-03-22ipa-firmware: Look at the table offsetHolger Hans Peter Freyther1-0/+7
* The two dummy bytes appear to be an offset, check that they are zero and refuse to work when they are not.
2010-03-22ipa-firmware: Warn/Exit when the required ids can not be found.Holger Hans Peter Freyther1-5/+13
2010-03-22ipa-firmware: Remove dummy check...Holger Hans Peter Freyther1-3/+0
We have derefenced that way before and would have crashed already.
2010-03-22ipa-firmware: The more_more_magic is to indicate the payloadHolger Hans Peter Freyther1-4/+0
* Ignore the value of the field there is no sematnic change to the content of the file..
2010-03-22debug: Use the right enum value for the ref counting group.Holger Hans Peter Freyther1-1/+1