summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* msgb_trim(): actually trim to an absolute length, as the comment statesHarald Welte2012-01-141-5/+3
| | | | | The previous commit introduced a new msgb_trim() but the implementation differed from the specification.
* msgb: introduce msgb_trim() and msgb_l3trim() to trim msgb'sHarald Welte2012-01-141-0/+28
|
* Merge remote-tracking branch 'origin/zecke/lapd-fixes'Holger Hans Peter Freyther2012-01-144-13/+40
|\
| * lapdm: Make sure that the msgb_l3len(msg) == length...Holger Hans Peter Freyther2012-01-123-13/+17
| | | | | | | | | | | | | | | | | | | | This code should not play with the internals of the msgb like this, this code got introduced in af48bed55607931307 and is breaking the osmo-bts usecase of forwarding an RSL message. Add a test case that fails without the new code. I would prefer if we could get rid of the manipulating the msgb like this, it is prone to errors like this one.
| * lapd: Warn if someone attempts to send an empty messageHolger Hans Peter Freyther2012-01-122-0/+22
| | | | | | | | | | | | DATA REQ with a msgb_l3len(msg) == 0 message does not make any sense, log an error and return immediately before attempting to send an empty I frame in lapd_send_i.
| * lapd: Mention the L3 size of the payload being sentHolger Hans Peter Freyther2012-01-121-1/+2
|/
* libosmocore/lapd: Fixed handling of sequence errors at lapd_core.cAndreas.Eversberg2012-01-101-3/+21
| | | | | | | | | | | | | | | If a sequence error is received, the N(R) variable must still be used to acknowledge previously transmitted frames. If there are two subsequent sequence errors received, ignore it. (Ignore every second subsequent error.) This happens if our reply with the REJ is too slow, so the remote gets a T200 timeout and sends another frame with a sequence error. Test showed that replying with two subsequent REJ messages could the remote L2 process to abort. Replying too slow shouldn't happen, but may happen over serial link between BB and LAPD. Written-by: Andreas.Eversberg <jolly@eversberg.eu> Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* Fix generation of GPRS Routing Area in case of 3-digit MNCGus Bourg2011-12-141-1/+1
| | | | | This fixes a simple typo that causes erroneous GPRS Routing Area encoding if you use 3-digit MNCs.
* lapd: Disable one part of the test as it fails on FreeBSDHolger Hans Peter Freyther2011-12-121-1/+2
| | | | | I thought this number would be wrong in the same way but it depends on OS/ARCH. Disable this part of the test for now.
* lapd_test: Look at the content of LAPD messages and compare themHolger Hans Peter Freyther2011-12-122-16/+61
| | | | Verify the output, accept two known errors in the generation of it.
* lapdm: Create a simple test case to transmit data between both endsHolger Hans Peter Freyther2011-12-117-1/+294
| | | | | | | This is testing the establishment of two ends (BTS, MS) one is in the polling mode as used by osmo-bts. Transfer data between the two ends. The sent data is not verified though. This should be followed up soon.
* auth_core: add missing string.h include to reduce warningsHarald Welte2011-12-071-0/+1
|
* auth_core: don't use anonymous unions to make certain gcc versions happyHarald Welte2011-12-075-20/+20
|
* milenage_test: make sure test output is what .ok file expectsHarald Welte2011-12-071-1/+1
|
* auth_milenage: we now copy RAND in auth_core, no need to do it twiceHarald Welte2011-12-071-1/+0
|
* milenage_test: make sure test result matches expectationsHarald Welte2011-12-071-1/+1
|
* add 'osmo-auc-gen' tool that can be used to generate auth vectorsHarald Welte2011-12-072-1/+187
| | | | | | Thsi tool usses libosmocore in order to generate 2G and 3G authentication vectors from private key material, exactly like an AuC would produce them in a real network.
* auth_core: Make sure we always save the RAND in the vectorHarald Welte2011-12-071-1/+8
|
* auth_core: add functions for parsing algorithm namesHarald Welte2011-12-072-0/+23
|
* whitespace fixes in milenage_test.okHarald Welte2011-12-071-7/+7
|
* add milenage internal headers to ensure 'make dist' worksHarald Welte2011-12-071-0/+5
|
* add autotest script for milenage/auth testingHarald Welte2011-12-076-1/+102
|
* Auth: Import milenage implementation from hostap (Jouni Malinen)Harald Welte2011-12-0614-3/+1659
| | | | ... and add integration into the osmo_auth core.
* Add a generic abstraction for GSM/3G authentication algorithmsHarald Welte2011-12-065-2/+229
| | | | | Indiidual algorithms can be implemented as plugins. libosmogsm itself only provides COMP128v1 via this generic interface.
* comp128: mark ki and srand as 'const' input parametersHarald Welte2011-12-062-2/+2
|
* increment LIBVERSION of libosmocore for new apiHarald Welte2011-12-011-1/+1
|
* SMS: export gsm411_bcdify() and gsm411_unbcdify() functionsHarald Welte2011-12-012-20/+26
| | | | | The same functions are e.g. required by the NITZ code in 04.08, and we don't want copy+paste.
* tests: Use the missing from the source directoryHolger Hans Peter Freyther2011-11-292-378/+2
| | | | | No need to keep a manual copy of the missing binary, use the one that will be in the top_srcdir.
* Update .gitignore with new testsSylvain Munaut2011-11-251-0/+2
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* libosmocore: Update LIBVERSION because of ABI chaangesSylvain Munaut2011-11-241-1/+1
| | | | | | The recent convolutional code changes break the ABI. Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* core/conv: Add regression testing codeSylvain Munaut2011-11-246-1/+555
| | | | | | | It tests both a given test vector encoding/decoding, and some random coding/decoding cycles. Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* core/conv: Add utility methods to know length of coded/decoded vectorsSylvain Munaut2011-11-242-0/+40
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* core/conv: Add support for other termination types (trunc & tail biting)Sylvain Munaut2011-11-242-38/+119
| | | | | | | | Note that this breaks the ABI and the low level API. But it shouldn't break the high level API, nor the conv code definitions (because fields default to 0, and for new fields '0' is the previous behavior) Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* core/conv: Really only consider error for non-zero soft valuesSylvain Munaut2011-11-241-3/+6
| | | | | | | | This should have been done with 1dd7c84733b20ba776510369e9daba1a822c5b44 but somehow was missed and only applied to the 'finish' method and not the 'scan' method. Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* core/conv: Minor documentation improvementsSylvain Munaut2011-11-242-12/+19
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* gsm/lapd_core: Make sure rc is initialize in lapd_rx_uSylvain Munaut2011-11-201-1/+1
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* core/logging: Remove const qualifier on return of logp2syslog_levelSylvain Munaut2011-11-201-1/+1
| | | | | | This generates a warning because it doesn't "mean" anything. Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* core/serial: Fix warning when on platform other than linuxSylvain Munaut2011-11-201-1/+1
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* gsm/a5: Bit faster way to compute parity ...Sylvain Munaut2011-11-201-3/+2
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* utils: Fix alias to weak alias to allow OSX compilationSylvain Munaut2011-11-191-1/+1
| | | | | | Using a weak alias works just as good AFAICT. Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* tests/a5: Add a5_test.ok as EXTRA_DISTSylvain Munaut2011-11-171-0/+1
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* tests: fix cut & paste errorSylvain Munaut2011-11-171-1/+1
| | | | | | | Damn ... I fixed it locally and didn't properly commit the fix before push. Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* gsm/a5: Add documentationSylvain Munaut2011-11-172-0/+82
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* gsm/a5: Rewrite A5/2 algo to avoid the delayed bitSylvain Munaut2011-11-171-23/+12
| | | | | | It's just clearer that way ... Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* tests/a5: Add a5 regression testingSylvain Munaut2011-11-176-1/+117
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* gsm/a5: Add const qualifier on the keySylvain Munaut2011-11-172-6/+6
| | | | Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
* tests: Fix the make distcheck of libosmocore with the testsHolger Hans Peter Freyther2011-11-1410-7/+7
|
* Merge branch 'zecke/gnu-autotest'Holger Hans Peter Freyther2011-11-1414-11/+620
|\
| * tests: Initialize log system to fix crash, update test resultHolger Hans Peter Freyther2011-11-143-2/+60
| | | | | | | | | | | | | | | | Use osmo_init_logging to initialize the log system to fix crashes when we attempt to parse broken ASN1 messages. Ignore stderr with parse errors, update the test result. make check is now passing.
| * tests: timer: set maximum wait time to obtain test resultsPablo Neira Ayuso2011-11-141-0/+19
| | | | | | | | | | | | | | | | If the timer test takes more than 2 * (number of steps + 10), we abort the test. This calculation is based on the maximum timeout randomly set (10 seconds) plus the number of steps (some existing timers may be reset in each step). We double this to have some extra grace time to finish.
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-12 18:44:49 +0000