March 2023 - gerrit-log - lists.osmocom.org

[S] Change in osmocom-bb[master]: layer23: modem: Forward GMMRR primitives between GMM and RLCMAC layers
by pespin 28 Mar '23

28 Mar '23

Attention is currently required from: pespin. Hello Jenkins Builder, fixeria, I'd like you to reexamine a change. Please visit https://gerrit.osmocom.org/c/osmocom-bb/+/32080 to look at the new patch set (#2). Change subject: layer23: modem: Forward GMMRR primitives between GMM and RLCMAC layers ...................................................................... layer23: modem: Forward GMMRR primitives between GMM and RLCMAC layers Related: OS#5501 Change-Id: I7603d8bbda0cdd437eb7557267efb7bbc3552565 --- M src/host/layer23/src/modem/gmm.c M src/host/layer23/src/modem/rlcmac.c 2 files changed, 40 insertions(+), 4 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmocom-bb refs/changes/80/32080/2 -- To view, visit https://gerrit.osmocom.org/c/osmocom-bb/+/32080 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmocom-bb Gerrit-Branch: master Gerrit-Change-Id: I7603d8bbda0cdd437eb7557267efb7bbc3552565 Gerrit-Change-Number: 32080 Gerrit-PatchSet: 2 Gerrit-Owner: pespin <pespin(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Attention: pespin <pespin(a)sysmocom.de> Gerrit-MessageType: newpatchset

1 0

[S] Change in osmocom-bb[master]: layer23: rlcmac: Support GMM detaching from VTY
by pespin 28 Mar '23

28 Mar '23

pespin has submitted this change. ( https://gerrit.osmocom.org/c/osmocom-bb/+/32051 ) Change subject: layer23: rlcmac: Support GMM detaching from VTY ...................................................................... layer23: rlcmac: Support GMM detaching from VTY Related: OS#5501 Depends: libosmo-gprs.git Change-Id If6cbb1d425b3a9f713348f1dea4747e2b6be0a44 Change-Id: Idca0bff6fc3e6a16b5e8379eae3315e3af012d78 --- M src/host/layer23/src/modem/vty.c 1 file changed, 39 insertions(+), 1 deletion(-) Approvals: fixeria: Looks good to me, approved Jenkins Builder: Verified diff --git a/src/host/layer23/src/modem/vty.c b/src/host/layer23/src/modem/vty.c index 0d13c93..69848df 100644 --- a/src/host/layer23/src/modem/vty.c +++ b/src/host/layer23/src/modem/vty.c @@ -172,7 +172,33 @@ gmm_prim->gmmreg.attach_req.attach_type = OSMO_GPRS_GMM_ATTACH_TYPE_GPRS; if (osmo_gprs_gmm_prim_upper_down(gmm_prim) != 0) { - vty_out(vty, "Failed to enqueue an GMM PDU%s", VTY_NEWLINE); + vty_out(vty, "Failed to enqueue a GMM PDU%s", VTY_NEWLINE); + return CMD_WARNING; + } + + return CMD_SUCCESS; +} + +DEFUN_HIDDEN(test_gmm_reg_detach, + test_gmm_reg_detach_cmd, + "test MS_NAME gmm detach", + TEST_CMD_DESC MS_NAME_DESC GMM_CMDG_DESC + "Enqueue a GMM GMMREG-DETACH.req for transmission\n") +{ + struct osmo_gprs_gmm_prim *gmm_prim; + const uint32_t tlli = 0xe1c5d364; + struct osmocom_ms *ms; + + if ((ms = l23_vty_get_ms(argv[0], vty)) == NULL) + return CMD_WARNING; + + gmm_prim = osmo_gprs_gmm_prim_alloc_gmmreg_detach_req(); + gmm_prim->gmmreg.detach_req.ptmsi = tlli; + gmm_prim->gmmreg.detach_req.detach_type = OSMO_GPRS_GMM_DETACH_MS_TYPE_GPRS; + gmm_prim->gmmreg.detach_req.poweroff_type = OSMO_GPRS_GMM_DETACH_POWEROFF_TYPE_NORMAL; + + if (osmo_gprs_gmm_prim_upper_down(gmm_prim) != 0) { + vty_out(vty, "Failed to enqueue a GMM PDU%s", VTY_NEWLINE); return CMD_WARNING; } @@ -363,6 +389,7 @@ install_element_ve(&test_llc_unitdata_req_hexpdu_cmd); install_element_ve(&test_llc_unitdata_req_gmm_attch_cmd); install_element_ve(&test_gmm_reg_attach_cmd); + install_element_ve(&test_gmm_reg_detach_cmd); install_element(CONFIG_NODE, &l23_cfg_ms_cmd); install_element(MS_NODE, &cfg_ms_apn_cmd); -- To view, visit https://gerrit.osmocom.org/c/osmocom-bb/+/32051 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmocom-bb Gerrit-Branch: master Gerrit-Change-Id: Idca0bff6fc3e6a16b5e8379eae3315e3af012d78 Gerrit-Change-Number: 32051 Gerrit-PatchSet: 2 Gerrit-Owner: pespin <pespin(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: pespin <pespin(a)sysmocom.de> Gerrit-MessageType: merged

1 0

[S] Change in osmo-bts[master]: trx: detect UL SID in EFR just like in FR
by falconia 28 Mar '23

28 Mar '23

falconia has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-bts/+/32095 ) Change subject: trx: detect UL SID in EFR just like in FR ...................................................................... trx: detect UL SID in EFR just like in FR The TCH/F Rx code in osmo-bts-trx uses osmo_fr_check_sid() to detect when the MS sends SID, and passes the flag to lchan_set_marker(). However, equivalent logic was missing for EFR, as until recently there was no EFR SID check function in libosmocodec. Now that we have osmo_efr_check_sid(), use it. Change-Id: Ib043e00dbf92145c2a6c32f6365517244472a922 --- M src/osmo-bts-trx/sched_lchan_tchf.c 1 file changed, 17 insertions(+), 0 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-bts refs/changes/95/32095/1 diff --git a/src/osmo-bts-trx/sched_lchan_tchf.c b/src/osmo-bts-trx/sched_lchan_tchf.c index eff0f63..256125b 100644 --- a/src/osmo-bts-trx/sched_lchan_tchf.c +++ b/src/osmo-bts-trx/sched_lchan_tchf.c @@ -150,6 +150,8 @@ break; case GSM48_CMODE_SPEECH_EFR: /* EFR */ rc = gsm0503_tch_fr_decode(tch_data, *bursts_p, 1, 1, &n_errors, &n_bits_total); + if (rc == GSM_EFR_BYTES) /* only for valid *speech* frames */ + lchan_set_marker(osmo_efr_check_sid(tch_data, rc), lchan); /* DTXu */ break; case GSM48_CMODE_SPEECH_AMR: /* AMR */ /* the first FN 0,8,17 defines that CMI is included in frame, -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/32095 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ib043e00dbf92145c2a6c32f6365517244472a922 Gerrit-Change-Number: 32095 Gerrit-PatchSet: 1 Gerrit-Owner: falconia <falcon(a)freecalypso.org> Gerrit-MessageType: newchange

1 0

[S] Change in osmo-bts[master]: sysmo: fix handling of SID in EFR
by falconia 28 Mar '23

28 Mar '23

falconia has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-bts/+/32096 ) Change subject: sysmo: fix handling of SID in EFR ...................................................................... sysmo: fix handling of SID in EFR Handling of SID in EFR mode was broken in osmo-bts-sysmo. l1_to_rtppayload_efr(), the function for UL Rx, was using completely bogus logic (passing bits in ETSI TS 101 318 EFR format to an AMR decoding function), whereas l1if_tch_encode(), the function for DL Tx, had missing SID logic for EFR while supporting SID detection for all other codecs. The fix is to use the new osmo_efr_check_sid() function in libosmocodec. Change-Id: Ia56c1bb7432968685110456961d24a907b0a201f --- M src/osmo-bts-sysmo/tch.c 1 file changed, 23 insertions(+), 7 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-bts refs/changes/96/32096/1 diff --git a/src/osmo-bts-sysmo/tch.c b/src/osmo-bts-sysmo/tch.c index a390c8c..8c9557c 100644 --- a/src/osmo-bts-sysmo/tch.c +++ b/src/osmo-bts-sysmo/tch.c @@ -131,12 +131,8 @@ cur[0] |= 0xC0; #endif /* USE_L1_RTP_MODE */ - enum osmo_amr_type ft; - enum osmo_amr_quality bfi; - uint8_t cmr; - int8_t sti, cmi; - osmo_amr_rtp_dec(l1_payload, payload_len, &cmr, &cmi, &ft, &bfi, &sti); - lchan_set_marker(ft == AMR_GSM_EFR_SID, lchan); + + lchan_set_marker(osmo_efr_check_sid(l1_payload, payload_len), lchan); return msg; } @@ -403,7 +399,10 @@ *payload_type = GsmL1_TchPlType_Efr; rc = rtppayload_to_l1_efr(l1_payload, rtp_pl, rtp_pl_len); - /* FIXME: detect and save EFR SID */ + if (rc && lchan->ts->trx->bts->dtxd) + is_sid = osmo_efr_check_sid(rtp_pl, rtp_pl_len); + if (is_sid) + dtx_cache_payload(lchan, rtp_pl, rtp_pl_len, fn, -1); break; #endif case GSM48_CMODE_SPEECH_AMR: -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/32096 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ia56c1bb7432968685110456961d24a907b0a201f Gerrit-Change-Number: 32096 Gerrit-PatchSet: 1 Gerrit-Owner: falconia <falcon(a)freecalypso.org> Gerrit-MessageType: newchange

1 0

[M] Change in osmo-bts[master]: common: implement rtp always-output mode
by falconia 28 Mar '23

28 Mar '23

falconia has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-bts/+/32098 ) Change subject: common: implement rtp always-output mode ...................................................................... common: implement rtp always-output mode In some environments it is highly desirable for the RTP stream coming from each GSM call UL on a BTS to be fully continuous, without any gaps, with _some_ RTP packet emitted every 20 ms, even if there is no speech or SID frame to be sent in that frame time window. The present change adds an rtp always-output vty option which, when enabled, causes the BTS to emit RTP packets with a zero-length payload, instead of producing gaps in the RTP stream, when it has nothing else to send. Related: OS#5975 Change-Id: Ic0e2edf2ed90ba0ac6bee5e7d9629bf0255e256d --- M include/osmo-bts/bts.h M src/common/l1sap.c M src/common/vty.c 3 files changed, 57 insertions(+), 2 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-bts refs/changes/98/32098/1 diff --git a/include/osmo-bts/bts.h b/include/osmo-bts/bts.h index ed3da1d..14c78a3 100644 --- a/include/osmo-bts/bts.h +++ b/include/osmo-bts/bts.h @@ -314,6 +314,8 @@ int rtp_ip_dscp; int rtp_priority; + bool rtp_nogaps_mode; /* emit RTP stream without any gaps */ + struct { uint8_t ciphers; /* flags A5/1==0x1, A5/2==0x2, A5/3==0x4 */ } support; diff --git a/src/common/l1sap.c b/src/common/l1sap.c index 2a1196c..f8320b0 100644 --- a/src/common/l1sap.c +++ b/src/common/l1sap.c @@ -1247,6 +1247,11 @@ static bool rtppayload_is_valid(struct gsm_lchan *lchan, struct msgb *resp_msg) { + /* A zero-length payload is never valid, it is merely a BFI marker + * and a timing tick, which we are not interested in because we run + * on our own TDMA timing. */ + if (!resp_msg->len) + return false; /* Avoid sending bw-efficient AMR to lower layers, most bts models * don't support it. */ if (lchan->tch_mode == GSM48_CMODE_SPEECH_AMR && @@ -1571,6 +1576,7 @@ static int l1sap_tch_ind(struct gsm_bts_trx *trx, struct osmo_phsap_prim *l1sap, struct ph_tch_param *tch_ind) { + struct gsm_bts *bts = trx->bts; struct msgb *msg = l1sap->oph.msg; struct gsm_time g_time; struct gsm_lchan *lchan; @@ -1600,8 +1606,10 @@ /* Low level layers always call us when TCH content is expected, even if * the content is not available due to decoding issues. Content not - * available is expected as empty payload. */ - if (msg->len) { + * available is expected as empty payload. We send out an RTP packet + * if the lower layers sent us a non-empty payload OR if + * rtp always-output mode is configured. */ + if (msg->len || bts->rtp_nogaps_mode) { /* hand msg to RTP code for transmission */ if (lchan->abis_ip.osmux.use) { lchan_osmux_send_frame(lchan, msg->data, msg->len, diff --git a/src/common/vty.c b/src/common/vty.c index dd1b07f..aa49f6b 100644 --- a/src/common/vty.c +++ b/src/common/vty.c @@ -421,6 +421,8 @@ vty_out(vty, " rtp ip-dscp %i%s", bts->rtp_ip_dscp, VTY_NEWLINE); if (bts->rtp_priority != -1) vty_out(vty, " rtp socket-priority %i%s", bts->rtp_priority, VTY_NEWLINE); + if (bts->rtp_nogaps_mode) + vty_out(vty, " rtp always-output%s", VTY_NEWLINE); vty_out(vty, " paging queue-size %u%s", paging_get_queue_max(bts->paging_state), VTY_NEWLINE); vty_out(vty, " paging lifetime %u%s", paging_get_lifetime(bts->paging_state), @@ -780,6 +782,28 @@ return CMD_SUCCESS; } +DEFUN(cfg_bts_rtp_always_output, + cfg_bts_rtp_always_output_cmd, + "rtp always-output", + RTP_STR "Always emit an RTP packet every 20 ms\n") +{ + struct gsm_bts *bts = vty->index; + + bts->rtp_nogaps_mode = true; + return CMD_SUCCESS; +} + +DEFUN(cfg_bts_no_rtp_always_output, + cfg_bts_no_rtp_always_output_cmd, + "no rtp always-output", + NO_STR RTP_STR "Always emit an RTP packet every 20 ms\n") +{ + struct gsm_bts *bts = vty->index; + + bts->rtp_nogaps_mode = false; + return CMD_SUCCESS; +} + #define PAG_STR "Paging related parameters\n" DEFUN_ATTR(cfg_bts_paging_queue_size, @@ -2667,6 +2691,8 @@ install_element(BTS_NODE, &cfg_bts_rtp_port_range_cmd); install_element(BTS_NODE, &cfg_bts_rtp_ip_dscp_cmd); install_element(BTS_NODE, &cfg_bts_rtp_priority_cmd); + install_element(BTS_NODE, &cfg_bts_rtp_always_output_cmd); + install_element(BTS_NODE, &cfg_bts_no_rtp_always_output_cmd); install_element(BTS_NODE, &cfg_bts_band_cmd); install_element(BTS_NODE, &cfg_description_cmd); install_element(BTS_NODE, &cfg_no_description_cmd); -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/32098 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ic0e2edf2ed90ba0ac6bee5e7d9629bf0255e256d Gerrit-Change-Number: 32098 Gerrit-PatchSet: 1 Gerrit-Owner: falconia <falcon(a)freecalypso.org> Gerrit-MessageType: newchange

1 0

[M] Change in osmo-bts[master]: TCH-IND interface: move link quality check into model code
by falconia 28 Mar '23

28 Mar '23

falconia has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-bts/+/32097 ) Change subject: TCH-IND interface: move link quality check into model code ...................................................................... TCH-IND interface: move link quality check into model code The l1sap_tch_ind() handler in the common code contains a check to see if the link quality is >= bts->min_qual_norm; if this condition is not met, RTP output is suppressed even if the BTS model sent a non-empty payload. It appears that the purpose of this check is to prevent emission of bogons during DTXu pauses with FR1 codec: the only speech frame integrity check in GSM 05.03 coding for FR1 is a CRC-3, radio noise interpreted as frame bits has 1/8 probability of passing that weak CRC, and if this link quality check is removed, the resulting behavior (at least with sysmoBTS) is that fairly frequent RTP bogons (garbage frames pretending to be valid speech) are emitted during DTXu pauses, and when the resulting stream is decoded per the specs, extremely unpleasant noises are heard. The present change does not alter behavior in any way, but affects the code structure: the same unchanged check is moved from the common l1sap_tch_ind() handler into model-specific code in each BTS model that feeds TCH-IND primitives to the former. This change allows individual BTS models more fine-grained control as to what checks are appropriate for their PHY design, and it paves the way for implementation of the OS#5975 feature of emitting empty timing tick RTP packets instead of packet gaps in the RTP stream. Related: OS#5975 Change-Id: Id95e48a578b7a8ce576862655abbe553c5b9a7bf --- M src/common/l1sap.c M src/osmo-bts-lc15/tch.c M src/osmo-bts-oc2g/tch.c M src/osmo-bts-sysmo/tch.c M src/osmo-bts-trx/sched_lchan_tchf.c M src/osmo-bts-trx/sched_lchan_tchh.c 6 files changed, 67 insertions(+), 31 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-bts refs/changes/97/32097/1 diff --git a/src/common/l1sap.c b/src/common/l1sap.c index efde406..2a1196c 100644 --- a/src/common/l1sap.c +++ b/src/common/l1sap.c @@ -1571,7 +1571,6 @@ static int l1sap_tch_ind(struct gsm_bts_trx *trx, struct osmo_phsap_prim *l1sap, struct ph_tch_param *tch_ind) { - struct gsm_bts *bts = trx->bts; struct msgb *msg = l1sap->oph.msg; struct gsm_time g_time; struct gsm_lchan *lchan; @@ -1601,9 +1600,8 @@ /* Low level layers always call us when TCH content is expected, even if * the content is not available due to decoding issues. Content not - * available is expected as empty payload. We also check if quality is - * good enough. */ - if (msg->len && tch_ind->lqual_cb >= bts->min_qual_norm) { + * available is expected as empty payload. */ + if (msg->len) { /* hand msg to RTP code for transmission */ if (lchan->abis_ip.osmux.use) { lchan_osmux_send_frame(lchan, msg->data, msg->len, diff --git a/src/osmo-bts-lc15/tch.c b/src/osmo-bts-lc15/tch.c index 526b2dd..0aaea22 100644 --- a/src/osmo-bts-lc15/tch.c +++ b/src/osmo-bts-lc15/tch.c @@ -363,17 +363,18 @@ uint8_t *payload, payload_type, payload_len, sid_first[9] = { 0 }; struct msgb *rmsg = NULL; struct gsm_lchan *lchan = &trx->ts[L1SAP_CHAN2TS(chan_nr)].lchan[l1sap_chan2ss(chan_nr)]; + uint16_t ber10k = data_ind->measParam.fBer * 10000; + int16_t lqual_cb = data_ind->measParam.fLinkQuality * 10; if (is_recv_only(lchan->abis_ip.speech_mode)) return -EAGAIN; - if (data_ind->msgUnitParam.u8Size < 1) { + if (data_ind->msgUnitParam.u8Size < 1 || lqual_cb < trx->bts->min_qual_norm) { LOGPFN(DL1P, LOGL_DEBUG, data_ind->u32Fn, "chan_nr %d Rx Payload size 0\n", chan_nr); /* Push empty payload to upper layers */ rmsg = msgb_alloc_headroom(256, 128, "L1P-to-RTP"); return add_l1sap_header(trx, rmsg, lchan, chan_nr, data_ind->u32Fn, - data_ind->measParam.fBer * 10000, - data_ind->measParam.fLinkQuality * 10, 0, 0, 0); + ber10k, lqual_cb, 0, 0, 0); } payload_type = data_ind->msgUnitParam.u8Buffer[0]; @@ -462,8 +463,7 @@ if (rmsg) return add_l1sap_header(trx, rmsg, lchan, chan_nr, data_ind->u32Fn, - data_ind->measParam.fBer * 10000, - data_ind->measParam.fLinkQuality * 10, 0, 0, 0); + ber10k, lqual_cb, 0, 0, 0); return 0; diff --git a/src/osmo-bts-oc2g/tch.c b/src/osmo-bts-oc2g/tch.c index 09ee1e7..0e504ad 100644 --- a/src/osmo-bts-oc2g/tch.c +++ b/src/osmo-bts-oc2g/tch.c @@ -363,17 +363,18 @@ uint8_t *payload, payload_type, payload_len, sid_first[9] = { 0 }; struct msgb *rmsg = NULL; struct gsm_lchan *lchan = &trx->ts[L1SAP_CHAN2TS(chan_nr)].lchan[l1sap_chan2ss(chan_nr)]; + uint16_t ber10k = data_ind->measParam.fBer * 10000; + int16_t lqual_cb = data_ind->measParam.fLinkQuality * 10; if (is_recv_only(lchan->abis_ip.speech_mode)) return -EAGAIN; - if (data_ind->msgUnitParam.u8Size < 1) { + if (data_ind->msgUnitParam.u8Size < 1 || lqual_cb < trx->bts->min_qual_norm) { LOGPFN(DL1P, LOGL_DEBUG, data_ind->u32Fn, "chan_nr %d Rx Payload size 0\n", chan_nr); /* Push empty payload to upper layers */ rmsg = msgb_alloc_headroom(256, 128, "L1P-to-RTP"); return add_l1sap_header(trx, rmsg, lchan, chan_nr, data_ind->u32Fn, - data_ind->measParam.fBer * 10000, - data_ind->measParam.fLinkQuality * 10, 0, 0, 0); + ber10k, lqual_cb, 0, 0, 0); } payload_type = data_ind->msgUnitParam.u8Buffer[0]; @@ -474,8 +475,7 @@ if (rmsg) return add_l1sap_header(trx, rmsg, lchan, chan_nr, data_ind->u32Fn, - data_ind->measParam.fBer * 10000, - data_ind->measParam.fLinkQuality * 10, 0, 0, 0); + ber10k, lqual_cb, 0, 0, 0); return 0; diff --git a/src/osmo-bts-sysmo/tch.c b/src/osmo-bts-sysmo/tch.c index 8c9557c..6a5d2c3 100644 --- a/src/osmo-bts-sysmo/tch.c +++ b/src/osmo-bts-sysmo/tch.c @@ -507,20 +507,20 @@ uint8_t *payload, payload_type, payload_len, sid_first[9] = { 0 }; struct msgb *rmsg = NULL; struct gsm_lchan *lchan = &trx->ts[L1SAP_CHAN2TS(chan_nr)].lchan[l1sap_chan2ss(chan_nr)]; + uint16_t ber10k = data_ind->measParam.fBer * 10000; + int16_t lqual_cb = data_ind->measParam.fLinkQuality * 10; + int8_t rssi = data_ind->measParam.fRssi; + int16_t ta_offs = data_ind->measParam.i16BurstTiming * 64; if (is_recv_only(lchan->abis_ip.speech_mode)) return -EAGAIN; - if (data_ind->msgUnitParam.u8Size < 1) { + if (data_ind->msgUnitParam.u8Size < 1 || lqual_cb < trx->bts->min_qual_norm) { LOGPFN(DL1P, LOGL_DEBUG, data_ind->u32Fn, "chan_nr %d Rx Payload size 0\n", chan_nr); /* Push empty payload to upper layers */ rmsg = msgb_alloc_headroom(256, 128, "L1P-to-RTP"); return add_l1sap_header(trx, rmsg, lchan, chan_nr, data_ind->u32Fn, - data_ind->measParam.fBer * 10000, - data_ind->measParam.fLinkQuality * 10, - data_ind->measParam.fRssi, - data_ind->measParam.i16BurstTiming * 64, - 0); + ber10k, lqual_cb, rssi, ta_offs, 0); } payload_type = data_ind->msgUnitParam.u8Buffer[0]; @@ -614,11 +614,7 @@ if (rmsg) return add_l1sap_header(trx, rmsg, lchan, chan_nr, data_ind->u32Fn, - data_ind->measParam.fBer * 10000, - data_ind->measParam.fLinkQuality * 10, - data_ind->measParam.fRssi, - data_ind->measParam.i16BurstTiming * 64, - 0); + ber10k, lqual_cb, rssi, ta_offs, 0); return 0; diff --git a/src/osmo-bts-trx/sched_lchan_tchf.c b/src/osmo-bts-trx/sched_lchan_tchf.c index 256125b..a7e45aa 100644 --- a/src/osmo-bts-trx/sched_lchan_tchf.c +++ b/src/osmo-bts-trx/sched_lchan_tchf.c @@ -75,6 +75,7 @@ { struct l1sched_chan_state *chan_state = &l1ts->chan_state[bi->chan]; struct gsm_lchan *lchan = chan_state->lchan; + struct gsm_bts *bts = lchan->ts->trx->bts; sbit_t *burst, **bursts_p = &chan_state->ul_bursts; uint8_t *mask = &chan_state->ul_mask; uint8_t rsl_cmode = chan_state->rsl_cmode; @@ -281,9 +282,12 @@ bfi: if (rsl_cmode == RSL_CMOD_SPD_SPEECH) { - /* indicate bad frame */ - if (lchan->tch.dtx.ul_sid) { - /* DTXu: pause in progress. Push empty payload to upper layers */ + /* Indicate bad frame. We push an empty payload to + * upper layers under two conditions: + * + * 1) If there is a DTXu pause in progress, or + * 2) If the link quality is too low. */ + if (lchan->tch.dtx.ul_sid || meas_avg.ci_cb < bts->min_qual_norm) { rc = 0; goto compose_l1sap; } diff --git a/src/osmo-bts-trx/sched_lchan_tchh.c b/src/osmo-bts-trx/sched_lchan_tchh.c index 37fd59d..2b6f4c3 100644 --- a/src/osmo-bts-trx/sched_lchan_tchh.c +++ b/src/osmo-bts-trx/sched_lchan_tchh.c @@ -97,6 +97,7 @@ { struct l1sched_chan_state *chan_state = &l1ts->chan_state[bi->chan]; struct gsm_lchan *lchan = chan_state->lchan; + struct gsm_bts *bts = lchan->ts->trx->bts; sbit_t *burst, **bursts_p = &chan_state->ul_bursts; uint8_t *mask = &chan_state->ul_mask; uint8_t rsl_cmode = chan_state->rsl_cmode; @@ -319,9 +320,12 @@ /* A FACCH/H frame replaces two speech frames, so we need to send two BFIs. * One is sent here, another will be sent two bursts later (see above). */ if (rsl_cmode == RSL_CMOD_SPD_SPEECH) { - /* indicate bad frame */ - if (lchan->tch.dtx.ul_sid) { - /* DTXu: pause in progress. Push empty payload to upper layers */ + /* Indicate bad frame. We push an empty payload to + * upper layers under two conditions: + * + * 1) If there is a DTXu pause in progress, or + * 2) If the link quality is too low. */ + if (lchan->tch.dtx.ul_sid || meas_avg.ci_cb < bts->min_qual_norm) { rc = 0; goto compose_l1sap; } -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/32097 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Id95e48a578b7a8ce576862655abbe553c5b9a7bf Gerrit-Change-Number: 32097 Gerrit-PatchSet: 1 Gerrit-Owner: falconia <falcon(a)freecalypso.org> Gerrit-MessageType: newchange

1 0

[M] Change in osmo-bts[master]: speech TCH, all models: implement option of Themyscira BFI packets
by falconia 28 Mar '23

28 Mar '23

falconia has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-bts/+/32099 ) Change subject: speech TCH, all models: implement option of Themyscira BFI packets ...................................................................... speech TCH, all models: implement option of Themyscira BFI packets Themyscira Wireless System Software is an external add-on to Osmocom CNI that provides interconnection between an Osmocom GSM network and regular national public telephone networks, developed initially for USA PSTN. Voice plane interconnection to PSTN is achieved by way of G.711 RTP streams, and these public-interfacing RTP streams are required to be fully continuous, emitting 160 PCM samples every 20 ms no matter what is happening on the GSM side. The transcoding boundary MGW component of ThemWi is responsible for comfort noise generation during GSM DTXu pauses and error concealment any time GSM UL traffic frames are lost for any other reason - but it requires a bit of assistance from the IP BTS that is the ultimate origin of the outgoing RTP stream: when FR or EFR codec is used, ThemWi transcoding MGW expects the IP BTS to emit special Themyscira-invented BFI packets any time a BFI condition occurs. These Themyscira BFI packets also carry a TAF bit exactly like the one in GSM 08.60 TRAU UL frames. The present change extends the rtp always-output vty option added just previously: the new option becomes rtp always-output [themyscira-bfi], and if the last qualifier is included, the special Themyscira extension mode is enabled. In this mode, which only affects FR and EFR codecs, RTP packets emitted when there is no speech or SID frame to be sent change from zero-length RTP payload to Themyscira BFI marker format, including TAF. Related: OS#5975 Change-Id: Ib935efb1f7c9c8d919a4a0ea2bef244a47e5fb24 --- M include/osmo-bts/bts.h M src/common/l1sap.c M src/common/vty.c M src/osmo-bts-trx/l1_if.c M src/osmo-bts-trx/sched_lchan_tchf.c 5 files changed, 95 insertions(+), 15 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-bts refs/changes/99/32099/1 diff --git a/include/osmo-bts/bts.h b/include/osmo-bts/bts.h index 14c78a3..56bf289 100644 --- a/include/osmo-bts/bts.h +++ b/include/osmo-bts/bts.h @@ -315,6 +315,7 @@ int rtp_priority; bool rtp_nogaps_mode; /* emit RTP stream without any gaps */ + bool rtp_themyscira_ext; /* Themyscira extension for FR/EFR BFI */ struct { uint8_t ciphers; /* flags A5/1==0x1, A5/2==0x2, A5/3==0x4 */ diff --git a/src/common/l1sap.c b/src/common/l1sap.c index f8320b0..3e4d2a0 100644 --- a/src/common/l1sap.c +++ b/src/common/l1sap.c @@ -1252,6 +1252,9 @@ * on our own TDMA timing. */ if (!resp_msg->len) return false; + /* Likewise weed out Themyscira BFI marker packets. */ + if (resp_msg->len == 2 && resp_msg->data[0] == 0xBF) + return false; /* Avoid sending bw-efficient AMR to lower layers, most bts models * don't support it. */ if (lchan->tch_mode == GSM48_CMODE_SPEECH_AMR && @@ -1610,13 +1613,27 @@ * if the lower layers sent us a non-empty payload OR if * rtp always-output mode is configured. */ if (msg->len || bts->rtp_nogaps_mode) { + uint8_t *rtp_pl = msg->data; + uint16_t rtp_pl_len = msg->len; + uint8_t bfi[2]; + + /* Generate and substitute Themyscira BFI packet if necessary */ + if (!msg->len && bts->rtp_themyscira_ext && + (lchan->type == GSM_LCHAN_TCH_F) && + (lchan->tch_mode == GSM48_CMODE_SPEECH_V1 || + lchan->tch_mode == GSM48_CMODE_SPEECH_EFR)) { + bfi[0] = 0xBF; + bfi[1] = (fn % 104 == 52); /* TAF */ + rtp_pl = bfi; + rtp_pl_len = 2; + } /* hand msg to RTP code for transmission */ if (lchan->abis_ip.osmux.use) { - lchan_osmux_send_frame(lchan, msg->data, msg->len, + lchan_osmux_send_frame(lchan, rtp_pl, rtp_pl_len, fn_ms_adj(fn, lchan), lchan->rtp_tx_marker); } else if (lchan->abis_ip.rtp_socket) { osmo_rtp_send_frame_ext(lchan->abis_ip.rtp_socket, - msg->data, msg->len, fn_ms_adj(fn, lchan), lchan->rtp_tx_marker); + rtp_pl, rtp_pl_len, fn_ms_adj(fn, lchan), lchan->rtp_tx_marker); } /* if loopback is enabled, also queue received RTP data */ if (lchan->loopback) { diff --git a/src/common/vty.c b/src/common/vty.c index aa49f6b..a8155bf 100644 --- a/src/common/vty.c +++ b/src/common/vty.c @@ -421,8 +421,12 @@ vty_out(vty, " rtp ip-dscp %i%s", bts->rtp_ip_dscp, VTY_NEWLINE); if (bts->rtp_priority != -1) vty_out(vty, " rtp socket-priority %i%s", bts->rtp_priority, VTY_NEWLINE); - if (bts->rtp_nogaps_mode) - vty_out(vty, " rtp always-output%s", VTY_NEWLINE); + if (bts->rtp_nogaps_mode) { + vty_out(vty, " rtp always-output"); + if (bts->rtp_themyscira_ext) + vty_out(vty, " themyscira-bfi"); + vty_out(vty, "%s", VTY_NEWLINE); + } vty_out(vty, " paging queue-size %u%s", paging_get_queue_max(bts->paging_state), VTY_NEWLINE); vty_out(vty, " paging lifetime %u%s", paging_get_lifetime(bts->paging_state), @@ -784,12 +788,17 @@ DEFUN(cfg_bts_rtp_always_output, cfg_bts_rtp_always_output_cmd, - "rtp always-output", - RTP_STR "Always emit an RTP packet every 20 ms\n") + "rtp always-output [themyscira-bfi]", + RTP_STR "Always emit an RTP packet every 20 ms\n" + "Emit Themyscira BFI packets for FR and EFR\n") { struct gsm_bts *bts = vty->index; bts->rtp_nogaps_mode = true; + if (argc > 0) + bts->rtp_themyscira_ext = true; + else + bts->rtp_themyscira_ext = false; return CMD_SUCCESS; } @@ -801,6 +810,7 @@ struct gsm_bts *bts = vty->index; bts->rtp_nogaps_mode = false; + bts->rtp_themyscira_ext = false; return CMD_SUCCESS; } diff --git a/src/osmo-bts-trx/l1_if.c b/src/osmo-bts-trx/l1_if.c index 2ed1868..93161b2 100644 --- a/src/osmo-bts-trx/l1_if.c +++ b/src/osmo-bts-trx/l1_if.c @@ -439,8 +439,14 @@ break; } - /* attempt to allocate an Error Concealment Unit instance, if available */ - lchan->ecu_state = osmo_ecu_init(trx, lchan2ecu_codec(lchan)); + /* Attempt to allocate an Error Concealment Unit + * instance, if available - but not in Themyscira + * environment, where in-BTS ECU is undesirable. + */ + if (!trx->bts->rtp_themyscira_ext) + lchan->ecu_state = osmo_ecu_init(trx, lchan2ecu_codec(lchan)); + else + lchan->ecu_state = NULL; /* activate dedicated channel */ trx_sched_set_lchan(lchan, chan_nr, LID_DEDIC, true); @@ -473,7 +479,10 @@ /* ECU for possibly new codec */ if (lchan->ecu_state) osmo_ecu_destroy(lchan->ecu_state); - lchan->ecu_state = osmo_ecu_init(trx, lchan2ecu_codec(lchan)); + if (!trx->bts->rtp_themyscira_ext) + lchan->ecu_state = osmo_ecu_init(trx, lchan2ecu_codec(lchan)); + else + lchan->ecu_state = NULL; /* change mode */ trx_sched_set_mode(lchan->ts, chan_nr, lchan->rsl_cmode, lchan->tch_mode, diff --git a/src/osmo-bts-trx/sched_lchan_tchf.c b/src/osmo-bts-trx/sched_lchan_tchf.c index a7e45aa..7211496 100644 --- a/src/osmo-bts-trx/sched_lchan_tchf.c +++ b/src/osmo-bts-trx/sched_lchan_tchf.c @@ -301,14 +301,22 @@ switch (tch_mode) { case GSM48_CMODE_SPEECH_V1: /* FR */ - memset(tch_data, 0, GSM_FR_BYTES); - tch_data[0] = 0xd0; - rc = GSM_FR_BYTES; + if (bts->rtp_themyscira_ext) + rc = 0; + else { + memset(tch_data, 0, GSM_FR_BYTES); + tch_data[0] = 0xd0; + rc = GSM_FR_BYTES; + } break; case GSM48_CMODE_SPEECH_EFR: /* EFR */ - memset(tch_data, 0, GSM_EFR_BYTES); - tch_data[0] = 0xc0; - rc = GSM_EFR_BYTES; + if (bts->rtp_themyscira_ext) + rc = 0; + else { + memset(tch_data, 0, GSM_EFR_BYTES); + tch_data[0] = 0xc0; + rc = GSM_EFR_BYTES; + } break; case GSM48_CMODE_SPEECH_AMR: /* AMR */ rc = osmo_amr_rtp_enc(tch_data, -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/32099 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ib935efb1f7c9c8d919a4a0ea2bef244a47e5fb24 Gerrit-Change-Number: 32099 Gerrit-PatchSet: 1 Gerrit-Owner: falconia <falcon(a)freecalypso.org> Gerrit-MessageType: newchange

1 0

[M] Change in osmo-bts[master]: sysmo: emit empty RTP ticks during FACCH stealing on TCH/F
by falconia 28 Mar '23

28 Mar '23

falconia has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-bts/+/32100 ) Change subject: sysmo: emit empty RTP ticks during FACCH stealing on TCH/F ...................................................................... sysmo: emit empty RTP ticks during FACCH stealing on TCH/F When FACCH stealing occurs and sysmoBTS L1 delivers GsmL1_Sapi_FacchF instead of GsmL1_Sapi_TchF, that 20 ms unit still needs to be accounted for in the RTP timestamp cadence, and if we run with rtp always-output output, then an actual BFI packet needs to be emitted. The original code failed to do either; the present change implements correct behavior for TCH/F. The case of TCH/H is still unhandled, and it may or may not be possible to handle correctly, particularly for rtp always-output mode. The fundamental difficulty is that in order to produce truly undisrupted RTP timing, two BFI packets will need to be sent for one FACCH, spaced 20 ms apart - but if the PHY only sends one GsmL1_Sapi_FacchH message surrounded by a 40 ms time gap, then we may be out of luck. Related: OS#5974 Change-Id: I39d15faade28fb7d670493a99a0e0bdb654e2a4a --- M src/osmo-bts-sysmo/l1_if.c M src/osmo-bts-sysmo/l1_if.h M src/osmo-bts-sysmo/tch.c 3 files changed, 56 insertions(+), 0 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-bts refs/changes/00/32100/1 diff --git a/src/osmo-bts-sysmo/l1_if.c b/src/osmo-bts-sysmo/l1_if.c index 646cf01..06347cf 100644 --- a/src/osmo-bts-sysmo/l1_if.c +++ b/src/osmo-bts-sysmo/l1_if.c @@ -980,6 +980,12 @@ return rc; } + /* If we got FACCH, the RTP clock needs to account for it, + * and if we have rtp always-output enabled, an actual BFI packet + * will be emitted. */ + if (data_ind->sapi == GsmL1_Sapi_FacchF) + l1if_tch_rx_facch(trx, chan_nr, l1p_msg); + /* fill L1SAP header */ sap_msg = l1sap_msgb_alloc(data_ind->msgUnitParam.u8Size); l1sap = msgb_l1sap_prim(sap_msg); diff --git a/src/osmo-bts-sysmo/l1_if.h b/src/osmo-bts-sysmo/l1_if.h index 8691eef..5b2da04 100644 --- a/src/osmo-bts-sysmo/l1_if.h +++ b/src/osmo-bts-sysmo/l1_if.h @@ -129,6 +129,8 @@ const uint8_t *rtp_pl, unsigned int rtp_pl_len, uint32_t fn, bool use_cache, bool marker); int l1if_tch_rx(struct gsm_bts_trx *trx, uint8_t chan_nr, struct msgb *l1p_msg); +int l1if_tch_rx_facch(struct gsm_bts_trx *trx, uint8_t chan_nr, + struct msgb *l1p_msg); int l1if_tch_fill(struct gsm_lchan *lchan, uint8_t *l1_buffer); struct msgb *gen_empty_tch_msg(struct gsm_lchan *lchan, uint32_t fn); diff --git a/src/osmo-bts-sysmo/tch.c b/src/osmo-bts-sysmo/tch.c index 6a5d2c3..cecb4dd 100644 --- a/src/osmo-bts-sysmo/tch.c +++ b/src/osmo-bts-sysmo/tch.c @@ -625,6 +625,29 @@ return -EINVAL; } +/*! \brief provide an RTP empty payload "tick" to upper layers upon FACCH */ +int l1if_tch_rx_facch(struct gsm_bts_trx *trx, uint8_t chan_nr, + struct msgb *l1p_msg) +{ + GsmL1_Prim_t *l1p = msgb_l1prim(l1p_msg); + GsmL1_PhDataInd_t *data_ind = &l1p->u.phDataInd; + struct msgb *rmsg = NULL; + struct gsm_lchan *lchan = &trx->ts[L1SAP_CHAN2TS(chan_nr)].lchan[l1sap_chan2ss(chan_nr)]; + + if (is_recv_only(lchan->abis_ip.speech_mode)) + return -EAGAIN; + + LOGPFN(DL1P, LOGL_DEBUG, data_ind->u32Fn, "chan_nr %d Rx FACCH\n", chan_nr); + /* Push empty payload to upper layers */ + rmsg = msgb_alloc_headroom(256, 128, "L1P-to-RTP"); + return add_l1sap_header(trx, rmsg, lchan, chan_nr, data_ind->u32Fn, + data_ind->measParam.fBer * 10000, + data_ind->measParam.fLinkQuality * 10, + 0, /* suppress RSSI like in osmo-bts-trx */ + data_ind->measParam.i16BurstTiming * 64, + 0); +} + struct msgb *gen_empty_tch_msg(struct gsm_lchan *lchan, uint32_t fn) { struct msgb *msg; -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/32100 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: I39d15faade28fb7d670493a99a0e0bdb654e2a4a Gerrit-Change-Number: 32100 Gerrit-PatchSet: 1 Gerrit-Owner: falconia <falcon(a)freecalypso.org> Gerrit-MessageType: newchange

1 0

[S] Change in gapk[master]: fmt_rtp_amr,fmt_rtp_efr: replace damaged packets with silence
by Matan Perelman 28 Mar '23

28 Mar '23

Attention is currently required from: Matan Perelman. Hello Jenkins Builder, I'd like you to reexamine a change. Please visit https://gerrit.osmocom.org/c/gapk/+/32094 to look at the new patch set (#2). Change subject: fmt_rtp_amr,fmt_rtp_efr: replace damaged packets with silence ...................................................................... fmt_rtp_amr,fmt_rtp_efr: replace damaged packets with silence Change-Id: I7245aa0bc0955cc8b94d5401a15e694f50498093 --- M src/fmt_rtp_amr.c M src/fmt_rtp_efr.c 2 files changed, 29 insertions(+), 0 deletions(-) git pull ssh://gerrit.osmocom.org:29418/gapk refs/changes/94/32094/2 -- To view, visit https://gerrit.osmocom.org/c/gapk/+/32094 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: gapk Gerrit-Branch: master Gerrit-Change-Id: I7245aa0bc0955cc8b94d5401a15e694f50498093 Gerrit-Change-Number: 32094 Gerrit-PatchSet: 2 Gerrit-Owner: Matan Perelman <matan1008(a)gmail.com> Gerrit-Reviewer: Jenkins Builder Gerrit-Attention: Matan Perelman <matan1008(a)gmail.com> Gerrit-MessageType: newpatchset

1 0

[S] Change in gapk[master]: fmt_rtp_amr,fmt_rtp_efr: replace damaged packets with silence
by Jenkins Builder 28 Mar '23

28 Mar '23

Jenkins Builder has posted comments on this change. ( https://gerrit.osmocom.org/c/gapk/+/32094 ) Change subject: fmt_rtp_amr,fmt_rtp_efr: replace damaged packets with silence ...................................................................... Patch Set 1: (1 comment) File src/fmt_rtp_efr.c: Robot Comment from checkpatch (run ID jenkins-gerrit-lint-5493): https://gerrit.osmocom.org/c/gapk/+/32094/comment/f83be597_88627b6c PS1, Line 70: if (!memcmp(DAMAGED_PACKET, src, EFR_LEN)) { braces {} are not necessary for single statement blocks -- To view, visit https://gerrit.osmocom.org/c/gapk/+/32094 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: gapk Gerrit-Branch: master Gerrit-Change-Id: I7245aa0bc0955cc8b94d5401a15e694f50498093 Gerrit-Change-Number: 32094 Gerrit-PatchSet: 1 Gerrit-Owner: Matan Perelman <matan1008(a)gmail.com> Gerrit-CC: Jenkins Builder Gerrit-Comment-Date: Tue, 28 Mar 2023 05:33:38 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: No Gerrit-MessageType: comment

1 0