AgeCommit message (Collapse)AuthorFilesLines
2018-02-28pcu_sock: LOG + drop DATA.req from PCU for non-PDCH timeslotHarald Welte1-0/+6
Change-Id: I3ae496eca96bc0823dfeca16ce36c200ce000895
2018-02-28pcu_sock: Log an error message and discard PCU primitives for BTS != 0Harald Welte1-0/+4
In OsmoBTS, we (so far?) only have a single BTS inside each process, let's make sure we log an error message if the PCU should ever want to transmit primitives to a non-zero BTS number. Change-Id: I158f935fed12941737c806c0677a8192ea3418a0
2018-02-28pcu_sock: Don't overflow the timeslot arrayHarald Welte1-0/+6
Don't blindly trust that the ts_nr received on the PCU socket will be small enough to not overflow our timeslot array! Change-Id: Ie9964c8dc0ca7b049da7dfec0ac0a0d3f1aedd45
2018-02-28pcu_sock: Discard messages that are too shortHarald Welte1-0/+6
The downstream code of pcu_sock.c doesn't contain any length checks, so let's discard any messages that are shorter than the primitive length. Change-Id: I35ac84d4db6d21ea61afbd1864c810bbf601d69b
2018-02-28pcuif_proto: correct indention of gsm_pcu_if_dataAlexander Couzens1-5/+5
Change-Id: I39f3bc1f0a1e238f8f00cb00e2d1e5193f118c16
2018-02-27Add high-accuracy ToA value to Uplink Measurement ReportsHarald Welte3-0/+41
Normal Abis RSL MEasurement Results contain only the "MS Timing Offset IE" in units of full symbols. In some use cases it is important to have higher-accuracy timing information exposed to the BSC. We do this by adding the average timing offset value during the last measurement interval in 1/256th symbol accuracy to the "Supplementary MEasuremen Information" part of the TS 48.058 9.3.25 Uplink Measurements IE. In order to avoid any compatibility issues, this feature is only enabled if the new vty config command "supp-meas-info toa256" at the bts node is enabled. Change-Id: Ie85e53b47d4041cc4e6d7b78406ae8b79b2d9397
2018-02-27measurement: Keep average of high-accurate ToA value in lchanHarald Welte2-0/+2
At the end of a measurement processing window, we currently compute the ToA / timing offset at 1/256th symbol accuracy, but we only print it to the log. Let's store the value in the lchan to make it usable by other code in follow-up patches. Change-Id: I5f00a16ac966b627d9452a98b8fa70984bed684a
2018-02-27L1SAP: Increase resolution of reported burst timingHarald Welte8-19/+19
Before this patch we had: * osmo-bts-trx internally using 1/256th bit/symbol period * osmo-bts-sysmo internally using 1/4 bit/smbol period * PCU interface using 1/4 * L1SAP interface using 1/4 * measurement processing code on top of L1SAP using 1/256 So for sysmo/lc15/octphy we are not loosing resolution, but for osmo-bts-trx we're arbitrarily reducing the resolution via L1SAP only then to compute with higher resolution again. Let's change L1SAP to use 1/256 bits and hence not loose any resolution. This requires a corresponding change in libosmocore for l1sap.h, which is found in Change-Id Ibb58113c2819fe2d6d23ecbcfb8b3fce4055025d Change-Id: If9b0f617845ba6c4aa47969f521734388197c9a7
2018-02-27measurement.c: higher-precision TA/TOA mathHarald Welte3-8/+8
Change-Id: I0dc8e78545465dfc5c93691a49b86b6b8b56b432
2018-02-27trx/scheduler: Use integer math for TOA (Timing of Arrival)Harald Welte10-71/+74
There's no need to express TOA as a float: * We receive it as signed 16bit integer in units 1/256 symbol periods * We pass it to L1SAP as signed integer in 1/4 symbol periods So turn it into an int16_t with 1/256 symbol period accuracy throughout the code to avoid both float arithmetic as well as loosing any precision. Change-Id: Idce4178e0b1f7e940ebc22b3e2f340fcd544d4ec
2018-02-27RACH decoding: Use BER threshold for RACH ghost detectionHarald Welte4-1/+24
When decoding RACH bursts, we should use a BER threshold in order to help distinguish 'ghost' RACH bursts from real RACH bursts. The theoretical ideal threshold according to some papers is 7 out of 41 bits qhich aquals to Eb/N0 of 0 dB = 0.1707 (17.07%) We add a new 'ber10k' parameter to the RACH indication l1sap primitive (needs separate change for libosmocore), and then fill this value from osmo-bts-{sysmo,lc15,trx,octphy}. The common part above L1SAP then applies the threshold, which can be changed from vty using the "max-ber10k-rach <0-10000>" command available at the BTS node. The unit is BER in 1/10000, i.e. a value of 100 equals 1% bit error rate. Change-Id: Ic41c11f6312a36baa2738547e8dcec80829457f8
2018-02-27Move rach_busy counting above L1SAPHarald Welte5-48/+44
In the past, rach_busy counting was performed below L1SAP, while reporting was handled above. This lead to subtle differences between the BTS models, such as osmo-bts-trx missing to increment rach_busy. Let's move the rach_busy counting above L1SAP to share more code. This means we need libosmocore Change-Id I9439810c3a3ad89ea0302753617b850749af887c for the additional required parameters in ph_rach_ind_param, as well as libosmocore Change-id I2b1926a37bde860dcfeb0d613eb55a71271928c5 for osmo-bts-trx to determine the RACH bit error rate. Change-Id: I3b989580cb38082e3fd8fc50a11fedda13991092 Closes: OS#3003
2018-02-26scheduler: Add missing \n at end of LOG statementHarald Welte1-1/+1
Change-Id: I55419dfa884b4170dfed696a7e1334940a46ba82
2018-02-26measurement.c: Don't silently copy "FULL" measurements to "SUB"Harald Welte1-2/+5
The existing code contained an ugly hack that if we didn't have any "SUB" measurements we would simply use the "FULL" values. That's wrong as TS 45.008 contains quite detailed rules on how the "SUB" values are to be computed. In some cases, they are identical to "FULL", but in most they are not. Let's remove the hack and replace it with an ERROR message, as clearly something is wrong if we ever encounter a measurement period end in which no single "SUB" measurement was received. The only situation in which this can occur is if the related uplink burst/block was missing, so let's set BER to 100% and level to lowest possible. Change-Id: I358f7b97fd4ea19264a77eff7abef13da7d5fbcd
2018-02-26measurement: Compute RX{LEV,QUAL}-SUB for SDCCH and non-AMR TCHHarald Welte1-3/+99
The rules on how to compute RX{LEV,QUAL}-SUB are rather convoluted, and depend on the detailed channel type and mode. For SDCCH and TCH/H in signalling mode, it's easy: No DTX is allowed, and all measurements are used in SUB. For non-AMR (TCH/F and TCH/H in non-signalling mode), we need to count the SACCH block measurements, as well as any SID/SID_UPDATE/L3_FILL/DUMMY blocks received in the blocks of table 8.3 of TS 45.008. Only AMR (TCH/AFS + TCH/AHS) are more difficult, as there are no fixed blocks/bursts/frames that always contain uplink messages, but the L1 will have to determine if a valid SID_UPDATE was received or not. This patch implements the above rules (except AMR related) in the common part of OsmoBTS. The AMR specific bits will have to follow as a later patch, likely in a BTS specific way, i.e. separate changes to sysmo/lc15/octphy/trx code. Related: OS#2978 Change-Id: I16eb3747a1c23df935a4c50dafe46abce512a474
2018-02-26split scheduler_mframe.c from scheduler.cHarald Welte6-818/+857
There are use cases for the multiframe scheduler tables outside the context of the entire scheduler. Let's prepare for that. Related: OS#2978 Change-Id: I6a501e66c47809ae3cdc55bef2cb6390ee0096b1
2018-02-26{sysmo,lc15}: Fix RACH reporting in combined CBCH caseHarald Welte2-2/+4
Combined CCCH with CBCH is a separate PCHAN type and hence we must accept it in the list of RACH-carrying pchan types in order to correctly report the RACH chan_nr when handing RACH requests up to L1SAP. The bug this fixes likely might have rendered cells with combined CBCH impossible to use. Change-Id: I9537463e5eedd2b8b30f298e0d3b308367c5b1fb
2018-02-26{sysmo,lc15}: Correctly report BER to L1SAP in INFO_MEAS_INDHarald Welte2-2/+2
L1SAP uses 'ber10k' values, i.e. BER in 1/10000 units, where 10000 is all errors are bit-errors (= 100%). The PHY on osmo-bts-sysmo and osmo-bts-lc15 is reporting a float fBer value scaled to 1, i.e. 1.00 = 100% and hence must be 10000 as ber10k. Before this patch, BER values reported on those BTS models were too low by a factor of 100, resulting in way too optimistic RxQual values reported to the BSC. Closes: OS#3005 Change-Id: I17e2f8fe8055f613da1e554cd36ed13289f56fb3
2018-02-26octphy: Remove code duplication for BER / RSSI conversionHarald Welte1-18/+22
Let's introduce some functions to hide the details of BER and RSSI conversion from OCTPHY representation to L1SAP representation. Change-Id: I517669c87a97b2ba164a2812811c8802fe0b92e8
2018-02-26counters: split rach:sent into rach:drop, rach:ho, rach:cs and rach:psHarald Welte3-4/+14
Change-Id: I51e9938df0e05b8fdb12686b9a9bb6994546deed
2018-02-26measurement.c: Fix sdcch4_meas_rep_fn102 / sdcch8_meas_rep_fn102Harald Welte1-15/+14
Since Change-Id: I23fba50f48415314da40cf5bf86fce2ed3e66af6 we were not reporting measurements for SDCCH channel types due to the wrong encoding of the sdcch{4,8}_meas_rep_fn102 table. Let's fix the table by encoding the needed information: "What is the modulo-102 remainder of the first burst of the last block before fn%102 reaches 37?" (SDCCH/4) "What is the modulo-102 remainder of the first burst of the last block before fn%102 reaches 12?" (SDCCH/8) The TS 45.002 Clause 7 tables have to be consulted carefully to determine this information. Change-Id: Icf02354872670126ab3297b787b216981ca6c309 Related: OS#2965
2018-02-25RSL IPA DLCX: Avoid another null-pointer dereferenceHarald Welte1-6/+5
In case a DLCX _with conn-id_ is issued without any CRCX before, we ran into a NULL pointer dereference in adding the connection statistics. Let's handle this gracefully and simply return empty statistics. Change-Id: If8b71266c847b90cdc51695b9f47b527c51bd70c Closes: OS#2996
2018-02-25RSL: Fix encoding of ConnectionID in IPA_DLCX_ACKHarald Welte1-1/+1
Change-Id: Ifdea747e78f191902870af53692f7a70fa7ece93 Closes: OS#2999
2018-02-25RSL IPA DLCX: Avoid null-pointer dereferenceHarald Welte1-5/+7
In case a DLCX is issued without any CRCX before, let's handle this gracefully and simply ack the DLCX anyway. Change-Id: I7c5bedccfc5a7cf552a9ce3a2dc712081c7ce177 Closes: OS#2996
2018-02-25paging: Fix encoding of PAGING TYPE 3 Rest OctetsHarald Welte1-8/+12
The RR PAGING TYPE 3 Rest Octets IE contains (among other things) the channel type needed for Mobile Identity 3 + 4 in the paging message. We did not only "forget" to encode those channel type needed field, but we have a completely wrong definition of those rest octets in libosmocore/include/gsm/protocol/gsm04_08.h "struct gsm48_paging3" Change-Id: I3a0bca6707ce95b68459c89f5b2b07f1590a1ab3 Closes: OS#2994
2018-02-24rsl: Improve ERROR REPORTingHarald Welte1-44/+81
Let's make sure all useful optional IEs of the RSL ERROR REPort aare present Change-Id: I5ecb98f8c72f472ac23c1e4e0f606b75e2cf032c
2018-02-24paging.c: Fix encoding of optional Mobile ID RR PAGING TYPE 1 / 2Harald Welte1-2/+2
It seems we have been encoding PAGING REQUEST TYPE 1 and PAGING REQUEST TYPE 2 erroneously all the time. The optional last Mobile Identity in those messages are TLV, not just LV. This is a quite serious bug in one of the most fundamental parts of the Radio Resource layer, and it has likely stayed hidden for a long time as usually in small networks there's a low paging load, reducing the amount of pressure to put multiple identities in one PAGING REQUEST message. Change-Id: Icc320ed130d0c29e9260a6a2aabe52e7346c3888 Closes: OS#2993
2018-02-24paging: Drop + Log paging requests for non-existant paging groupsHarald Welte1-0/+9
Change-Id: Id02c8fea51c1260f1baae8566c67bafc5bdb970e Related: OS#2992
2018-02-24BTS: add rate_ctr about CCCH (paging, agch, pch)Harald Welte9-0/+54
Change-Id: Id6c833746150a3c2e32b00ea6604669f16b84bc4
2018-02-24RSL: Send ERROR REPORT on too short/truncated messages + wrong discriminatorHarald Welte1-0/+6
Change-Id: I6a0e63999f39592474064e2f05df450aec8e37fe
2018-02-24RSL: Implement DELETE INDICATION on AGCH overflowHarald Welte1-0/+17
This patch adds generation of a DELETE INDICATION when the BTS AGCH queue overflows due to too many IMMEDIATE ASSIGN CMDs, as required by the specs. The AGCH queue length in OsmoBTS so far is at 1000 entries, which I consider way too high. But that is for another patch. Change-Id: Ied3306e85cbdc6f3476b10dc4bb0463cd728b274 Related: OS#2990
2018-02-24load_indication: Fix start of load indication timerHarald Welte2-6/+9
Starting the timer in bts_init() may result in it expiring already before the load indication period is set via OML. Let's make load_timer_start() safe to call several times and call it from OML code. Change-Id: I295d91413542014aa2507d5f09e01243fc3916fa Fixes: OS#2991
2018-02-23scheduler.c: Print message when burst substitution happensHarald Welte1-1/+4
Whenever we receive discontinuous frame numbers from the TRX socket, osmo-bts-trx is substituting zero-filled bursts for those frame numbers which we missed. Don't just do this silently, but actually log about it, as it is an error. Note: This [currently] happens when using a virtual air interface with trxcon as opposed to a real SDR receiver with osmo-trx. Change-Id: If79eab37c80647c9ab64f399fa4676d97af3e9ad
2018-02-23osmo-bts-trx: Add missing frame number to l1if_process_meas_res()Harald Welte1-1/+1
Change-Id: I7f1322b025ce733f78ae2b958c7a2bb38b6aa6f5
2018-02-23l1sap: Pass is_sub from L1 primitive into the Uplink MeasurementHarald Welte1-0/+1
info_meas_ind on the L1SAP always allowed the lower layers to pass in whether a given measurement is part of the "SUB", or not. However, the existing l1sap code before this patch simply drops this information, despite the measurement.c code also having "is_sub" state. Let's make sure this state is passed from L1SAP into measurement processing as intended. Fact is, none of our current lower-layers actually set this is_sub flag for their primitives passed up in L1SAP, but at least now *if* they would set that flag, the measurement code would process it as intended. Related: OS#2978 Change-Id: Ibed2e8d7563b471c6b5dd2214ac4765caf31ed2a
2018-02-23measurement.c: Hand Frame Number into measurement computationHarald Welte3-6/+6
This is currently only used for logging, but will be needed for proper RX{LEV,QUAL}-SUB reporting in upcoming patches. Related: OS#2978 Change-Id: I07fd06e8a379cab7c0c2eb111c3f5600037d3c9e
2018-02-23Revert "measurement: fix measurement computation"Harald Welte2-49/+1
This reverts commit d5fdcfe6d95f52fb76c4f4201969347a062fc9fd, which introduces a new function lchan_meas_update_ordered_TA whose functionality I still haven't yet managed to fully understand. It appears to be adjusting the requested timing advance (lchan->rqd_ta) but outside osmo-bts-trx/loops.c code. This is odd, as rqd_ta is a state variable of that loops.c code. So for one part, it is a failure of encapsulation. The TA loop code should be self-contained, particularly as it is only used for omso-bts-trx, and not for the other BTS models. The new lchan_meas_update_ordered_TA() function is used from common code, applicable to all BTS models. The resulting interaction between loops.c code and this new (now reverted) function cause the TA value to only ever grow, despite the MS never moving at all. Change-Id: I5a5adac6f18f94a5b51758a5ace8ef6ddfd23e80 Related: OS#2989
2018-02-23scheduler: add trx_sched_is_sacch_fn() functionHarald Welte2-0/+43
For proper measurement processing of RX{LEV,QUAL}-SUB, we will need this information. Related: OS#2978 Change-Id: I768fde62452a74dce471ebf946e56eb1e4de1abc
2018-02-23scheduler.c: Factor out find_sched_mframe_idx() functionHarald Welte1-15/+22
Let's split the look-up of the multiframe scheduler from the asignment to a given l1ts in trx_sched_set_pchan. Related: OS#2978 Change-Id: I79548b25aae647ce993a9d83c771d22b08cb1c74
2018-02-23osmo-bts-virtual: Make sure PRIM_INFO_MEAS have non-zero frame numberHarald Welte1-2/+3
Measurement reports fed into L1SAP so far had their frame number always set to zero, resulting in higher-layer common code above L1SAP to never detect the end of the measurement period, which in turn caused no RSL MEAS REP to be sent. Related: OS#2978 Change-Id: I67837d19515ea335614928570c12dd5027104c6b
2018-02-22osmo-bts-virtual: Make use of LOGL1S() macro for contextHarald Welte1-67/+40
In Change-Id I5703b46c8a59fe00a3cdc063bcf72872980ec5e5 we introduced LOGL1S and starte to use in in common/scheduler.c as well as osmo-bts-trx but somehow we didn't introduce it in osmo-bts-virtual at the time. Let's catch up. Change-Id: I0b5fd3b7982b9119becda844531108f64c68d19f
2018-02-22Introduce + use LOG/DEBUGP with frame number prefixing/printingHarald Welte9-125/+118
Let's make sure whenever we do have a frame number, we print it as context in the related log line Change-Id: I751d5ddb3322fce489bc241459738cbcc55c890b
2018-02-22osmo-bts-virtual: Generate PRIM_INFO_MEAS (with bogus values)Harald Welte1-2/+5
The problem is that measurement processing above L1SAP requires/expects those PRIM_INFO_MEAS indications from the bts specific parts. Otherwise it will never generate even uplink-only measurement reports to the BSC, which is a violation of Abis protocol specs. Change-Id: I48f73293cb4f0ab4c657dfd00e7ddd032a3c030f
2018-02-22osmo-bts-virtual: Shut down gracefully on socket creation failureHarald Welte2-2/+7
Don't simply crash if creation of the multicast socket fails Change-Id: Ie87b6684b3aa7f21742e4cf21533e980485c1230
2018-02-22bts: use feature list instead of speech codec tablePhilipp Maier8-80/+54
osmo-bts has a table of pchan/channel mode combinations for every bts. This table models the codec capabilitys of the BTS hardware. However, having the speech codec apabilities modeled inside the BTS feature list would be much more comfortable and since the feature list is communicated back to the BSC we would get the codec capabilities inside the BSC domain as well. - remove the pchan/channel mode tables - set speech codec variants for each BTS type - fix bts_supports_cm so that it queries the feature list Change-Id: I977dc729ba856631245aedf76afd48eac92166f7
2018-02-22vty: display bts features in vty command show btsPhilipp Maier1-0/+20
The VTY command show bts does not display the bts specific features yet. - Also display the feature list in snow-bts Change-Id: I509f2a7bbfa96c70bdfea4ff2488ee371e914620
2018-02-22octphy: ensure all BTS models set featuresPhilipp Maier5-0/+28
Most of the BTS models do not or do register not all of thier features to the the feature list. - Update/extend the feature lists for all BTS-Models Change-Id: I26765a64153368016921c2ac115b1c4aec9bc5e4
2018-02-22gsm_data_shared: extend bts feature list with speech codecsPhilipp Maier2-1/+12
The feature list does not cover any speech codec related information. - Add speech codec related items to feature list. Change-Id: If6d50c6f4e2348b23f31c3415b0f5577a3f5be50
2018-02-21RSL: Ensure we don't accept DCHAN messages for CCHANHarald Welte1-0/+22
If the Channel Number IE points to a common channel, we cannot accept such messages in code paths that only process dedicated channels, such as RLL/DCHAN/IPA. Related: OS#2972, OS#2971 Change-Id: I43a78bec63aeb36dd67043d237b27fe880209349
2018-02-21RSL: Properly reject RSL CHAN_NR IE for incompatible PCHANHarald Welte1-34/+78
If we receive a message for a dedicated channel, whose channel number structure doesn't match with the physical channel (timeslot) type, we must properly reject this. For RSL CHAN ACT it means sending a NACK, and for all other cases it means sending an RSL ERROR REPORT. Related: OS#2972, OS#2971 Change-Id: Iebd2571726d1284a7431b3f9b23ad3185e832ed1