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

[M] Change in osmo-bts[master]: csd_v110: handle empty/incomplete Uplink frames gracefully

by laforge

Attention is currently required from: fixeria. laforge has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bts/+/34219 ) Change subject: csd_v110: handle empty/incomplete Uplink frames gracefully ...................................................................... Patch Set 1: Code-Review+1 -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/34219 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: I7cbf868df3ba5d390cea3d529eef26d18dbe55ab Gerrit-Change-Number: 34219 Gerrit-PatchSet: 1 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Attention: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Comment-Date: Sat, 26 Aug 2023 06:04:17 +0000 Gerrit-HasComments: No Gerrit-Has-Labels: Yes Gerrit-MessageType: comment

1 year, 11 months

1
0
0 0

[S] Change in osmo-bts[master]: csd_v110: properly set E1/E2/E3 for non-transparent data

by laforge

Attention is currently required from: fixeria. laforge has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bts/+/34218 ) Change subject: csd_v110: properly set E1/E2/E3 for non-transparent data ...................................................................... Patch Set 1: Code-Review+1 -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/34218 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ie38c12e462654cd9fe83a0420bc8ea8b476214b8 Gerrit-Change-Number: 34218 Gerrit-PatchSet: 1 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Attention: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Comment-Date: Sat, 26 Aug 2023 06:03:56 +0000 Gerrit-HasComments: No Gerrit-Has-Labels: Yes Gerrit-MessageType: comment

1 year, 11 months

1
0
0 0

[S] Change in osmo-bts[master]: csd_v110: fix comments in csd_v110_rtp_{en,de}code()

by laforge

Attention is currently required from: fixeria. laforge has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bts/+/34217 ) Change subject: csd_v110: fix comments in csd_v110_rtp_{en,de}code() ...................................................................... Patch Set 1: Code-Review+1 -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/34217 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: I6fee665e587bfa76058187e13f0cdafaea991940 Gerrit-Change-Number: 34217 Gerrit-PatchSet: 1 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Attention: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Comment-Date: Sat, 26 Aug 2023 06:03:33 +0000 Gerrit-HasComments: No Gerrit-Has-Labels: Yes Gerrit-MessageType: comment

1 year, 11 months

1
0
0 0

[S] Change in osmo-bts[master]: csd_v110: properly set E1/E2/E3 for non-transparent data

by fixeria

fixeria has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-bts/+/34218 ) Change subject: csd_v110: properly set E1/E2/E3 for non-transparent data ...................................................................... csd_v110: properly set E1/E2/E3 for non-transparent data Change-Id: Ie38c12e462654cd9fe83a0420bc8ea8b476214b8 Related: OS#1572 --- M src/common/csd_v110.c 1 file changed, 21 insertions(+), 3 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-bts refs/changes/18/34218/1 diff --git a/src/common/csd_v110.c b/src/common/csd_v110.c index 8fa4229..230a6a7 100644 --- a/src/common/csd_v110.c +++ b/src/common/csd_v110.c @@ -102,9 +102,17 @@ osmo_csd_3k6_decode_frame(&df, &data[i * 36], 36); /* E1 .. E3 must set by out-of-band knowledge */ - df.e_bits[0] = e1e2e3_map[lchan->csd_mode][0]; - df.e_bits[1] = e1e2e3_map[lchan->csd_mode][1]; - df.e_bits[2] = e1e2e3_map[lchan->csd_mode][2]; + if (lchan->csd_mode == LCHAN_CSD_M_NT) { + /* non-transparent: as per 3GPP TS 48.020, Table 7 */ + df.e_bits[0] = 0; /* E1: as per 15.1.2, shall be set to 0 (for BSS-MSC) */ + df.e_bits[1] = (i >> 1) & 0x01; /* E2: 0 for Q1/Q2, 1 for Q3/Q4 */ + df.e_bits[2] = (i >> 0) & 0x01; /* E3: 0 for Q1/Q3, 1 for Q2/Q4 */ + } else { + /* transparent: as per 3GPP TS 44.021, Figure 4 */ + df.e_bits[0] = e1e2e3_map[lchan->csd_mode][0]; /* E1 */ + df.e_bits[1] = e1e2e3_map[lchan->csd_mode][1]; /* E2 */ + df.e_bits[2] = e1e2e3_map[lchan->csd_mode][2]; /* E3 */ + } osmo_v110_encode_frame(&ra_bits[i * 80], 80, &df); } -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/34218 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ie38c12e462654cd9fe83a0420bc8ea8b476214b8 Gerrit-Change-Number: 34218 Gerrit-PatchSet: 1 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-MessageType: newchange

1 year, 11 months

1
0
0 0

[M] Change in osmo-bts[master]: csd_v110: handle empty/incomplete Uplink frames gracefully

by fixeria

fixeria has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-bts/+/34219 ) Change subject: csd_v110: handle empty/incomplete Uplink frames gracefully ...................................................................... csd_v110: handle empty/incomplete Uplink frames gracefully Change-Id: I7cbf868df3ba5d390cea3d529eef26d18dbe55ab Related: OS#1572 --- M src/common/csd_v110.c M tests/csd/csd_test.c M tests/csd/csd_test.err 3 files changed, 90 insertions(+), 3 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-bts refs/changes/19/34219/1 diff --git a/src/common/csd_v110.c b/src/common/csd_v110.c index 230a6a7..5046d6e 100644 --- a/src/common/csd_v110.c +++ b/src/common/csd_v110.c @@ -87,9 +87,14 @@ if (OSMO_UNLIKELY(desc->num_blocks == 0)) return -ENOTSUP; - /* handle empty/incomplete frames gracefully */ - if (OSMO_UNLIKELY(data_len < (desc->num_blocks * desc->num_bits))) - return -ENODATA; + /* handle empty/incomplete Uplink frames gracefully */ + if (OSMO_UNLIKELY(data_len < (desc->num_blocks * desc->num_bits))) { + /* encode N idle frames as per 3GPP TS 44.021, section 8.1.6 */ + memset(&ra_bits[0], 0x01, sizeof(ra_bits)); + for (unsigned int i = 0; i < desc->num_blocks; i++) + memset(&ra_bits[i * 80], 0x00, 8); /* alignment pattern */ + goto ra1_ra2; + } /* RA1'/RA1: convert from radio rate to an intermediate data rate */ for (unsigned int i = 0; i < desc->num_blocks; i++) { @@ -117,6 +122,7 @@ osmo_v110_encode_frame(&ra_bits[i * 80], 80, &df); } +ra1_ra2: /* RA1/RA2: convert from an intermediate rate to 64 kbit/s */ if (desc->num_blocks == 4) { /* 4 * 80 bits (16 kbit/s) => 2 bits per octet */ diff --git a/tests/csd/csd_test.c b/tests/csd/csd_test.c index f24331d..466ff88 100644 --- a/tests/csd/csd_test.c +++ b/tests/csd/csd_test.c @@ -135,6 +135,17 @@ fprintf(stderr, "[!] Data mismatch @ %03u: D%u vs E%u\n", i, data_dec[i], data_enc[i]); } + + fprintf(stderr, "[i] Testing '%s' (IDLE)\n", tc->name); + + /* encode an idle RTP frame and print it */ + rc = csd_v110_rtp_encode(&lchan, &rtp[0], &data_enc[0], 0); + fprintf(stderr, "[i] csd_v110_rtp_encode() returns %d\n", rc); + if (rc != RFC4040_RTP_PLEN) + return; + /* print the encoded RTP frame (16 bytes per row) */ + for (unsigned int i = 0; i < sizeof(rtp) / 16; i++) + fprintf(stderr, " %s\n", osmo_hexdump(&rtp[i * 16], 16)); } int main(int argc, char **argv) diff --git a/tests/csd/csd_test.err b/tests/csd/csd_test.err index 94505dc..b862324 100644 --- a/tests/csd/csd_test.err +++ b/tests/csd/csd_test.err @@ -12,6 +12,18 @@ ff 7f 7f 7f bf bf bf bf ff 7f 7f 7f bf ff 7f 7f bf bf bf bf ff 7f 7f 7f bf bf bf bf ff 7f 7f 7f [i] csd_v110_rtp_decode() returns 240 +[i] Testing 'TCH/F9.6' (IDLE) +[i] csd_v110_rtp_encode() returns 160 + 3f 3f 3f 3f ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff 3f 3f 3f 3f ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + 3f 3f 3f 3f ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff 3f 3f 3f 3f ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff [i] Testing 'TCH/F4.8' (bitnum=120) [i] csd_v110_rtp_encode() returns 160 7f 7f 7f 7f 7f 7f 7f 7f ff 7f ff 7f ff 7f ff 7f @@ -25,6 +37,18 @@ ff 7f ff 7f ff 7f ff 7f ff ff 7f ff 7f ff 7f ff ff 7f ff 7f ff 7f ff 7f ff ff 7f ff 7f ff 7f ff [i] csd_v110_rtp_decode() returns 120 +[i] Testing 'TCH/F4.8' (IDLE) +[i] csd_v110_rtp_encode() returns 160 + 7f 7f 7f 7f 7f 7f 7f 7f ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + 7f 7f 7f 7f 7f 7f 7f 7f ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff [i] Testing 'TCH/H4.8' (bitnum=240) [i] csd_v110_rtp_encode() returns 160 3f 3f 3f 3f bf bf bf bf ff 7f 7f 7f bf bf bf bf @@ -38,6 +62,18 @@ ff 7f 7f 7f bf bf bf bf ff 7f 7f 7f bf ff 7f 7f bf bf bf bf ff 7f 7f 7f bf bf bf bf ff 7f 7f 7f [i] csd_v110_rtp_decode() returns 240 +[i] Testing 'TCH/H4.8' (IDLE) +[i] csd_v110_rtp_encode() returns 160 + 3f 3f 3f 3f ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff 3f 3f 3f 3f ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + 3f 3f 3f 3f ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff 3f 3f 3f 3f ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff [i] Testing 'TCH/F2.4' (bitnum=72) [i] csd_v110_rtp_encode() returns 160 7f 7f 7f 7f 7f 7f 7f 7f ff 7f 7f ff ff 7f 7f ff @@ -51,6 +87,18 @@ ff 7f 7f ff ff 7f 7f ff ff 7f 7f ff ff 7f 7f ff ff 7f 7f ff ff 7f 7f ff ff 7f 7f ff ff 7f 7f ff [i] csd_v110_rtp_decode() returns 72 +[i] Testing 'TCH/F2.4' (IDLE) +[i] csd_v110_rtp_encode() returns 160 + 7f 7f 7f 7f 7f 7f 7f 7f ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + 7f 7f 7f 7f 7f 7f 7f 7f ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff [i] Testing 'TCH/H2.4' (bitnum=144) [i] csd_v110_rtp_encode() returns 160 3f 3f 3f 3f bf 7f bf 7f bf 7f bf 7f bf 7f bf 7f @@ -64,3 +112,15 @@ bf 7f bf 7f bf 7f bf 7f bf 7f bf 7f ff 3f 7f 7f bf 7f bf 7f bf 7f bf 7f bf 7f bf 7f bf 7f bf 7f [i] csd_v110_rtp_decode() returns 144 +[i] Testing 'TCH/H2.4' (IDLE) +[i] csd_v110_rtp_encode() returns 160 + 3f 3f 3f 3f ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff 3f 3f 3f 3f ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + 3f 3f 3f 3f ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff 3f 3f 3f 3f ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff + ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/34219 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: I7cbf868df3ba5d390cea3d529eef26d18dbe55ab Gerrit-Change-Number: 34219 Gerrit-PatchSet: 1 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-MessageType: newchange

1 year, 11 months

1
0
0 0

[S] Change in osmo-bts[master]: csd_v110: fix comments in csd_v110_rtp_{en,de}code()

by fixeria

fixeria has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-bts/+/34217 ) Change subject: csd_v110: fix comments in csd_v110_rtp_{en,de}code() ...................................................................... csd_v110: fix comments in csd_v110_rtp_{en,de}code() Change-Id: I6fee665e587bfa76058187e13f0cdafaea991940 Related: OS#1572 --- M src/common/csd_v110.c 1 file changed, 14 insertions(+), 4 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-bts refs/changes/17/34217/1 diff --git a/src/common/csd_v110.c b/src/common/csd_v110.c index 9fc7ad3..8fa4229 100644 --- a/src/common/csd_v110.c +++ b/src/common/csd_v110.c @@ -91,11 +91,11 @@ if (OSMO_UNLIKELY(data_len < (desc->num_blocks * desc->num_bits))) return -ENODATA; - /* RA1'/RA1: convert to an intermediate data rate */ + /* RA1'/RA1: convert from radio rate to an intermediate data rate */ for (unsigned int i = 0; i < desc->num_blocks; i++) { struct osmo_v110_decoded_frame df; - /* convert modified V.110 frames to normal V.110 frames */ + /* convert a V.110 36-/60-bit frame to a V.110 80-bit frame */ if (desc->num_bits == 60) osmo_csd_12k_6k_decode_frame(&df, &data[i * 60], 60); else /* desc->num_bits == 36 */ @@ -158,11 +158,11 @@ ra_bits[i] = (rtp[i] >> 7); } - /* RA1'/RA1: convert to an intermediate data rate */ + /* RA1'/RA1: convert from an intermediate rate to radio rate */ for (unsigned int i = 0; i < desc->num_blocks; i++) { struct osmo_v110_decoded_frame df; - /* convert modified V.110 frames to normal V.110 frames */ + /* convert a V.110 80-bit frame to a V.110 36-/60-bit frame */ osmo_v110_decode_frame(&df, &ra_bits[i * 80], 80); if (desc->num_bits == 60) osmo_csd_12k_6k_encode_frame(&data[i * 60], 60, &df); -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/34217 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: I6fee665e587bfa76058187e13f0cdafaea991940 Gerrit-Change-Number: 34217 Gerrit-PatchSet: 1 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-MessageType: newchange

1 year, 11 months

1
0
0 0

[S] Change in libosmo-netif[master]: stream test: Fix test output check

by daniel

Attention is currently required from: arehbein. daniel has posted comments on this change. ( https://gerrit.osmocom.org/c/libosmo-netif/+/34210 ) Change subject: stream test: Fix test output check ...................................................................... Patch Set 1: Code-Review+2 -- To view, visit https://gerrit.osmocom.org/c/libosmo-netif/+/34210 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: libosmo-netif Gerrit-Branch: master Gerrit-Change-Id: Ib8f01c3651c13814c212328946415094da39d6a8 Gerrit-Change-Number: 34210 Gerrit-PatchSet: 1 Gerrit-Owner: arehbein <arehbein(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: daniel <dwillmann(a)sysmocom.de> Gerrit-Reviewer: pespin <pespin(a)sysmocom.de> Gerrit-Attention: arehbein <arehbein(a)sysmocom.de> Gerrit-Comment-Date: Fri, 25 Aug 2023 18:48:40 +0000 Gerrit-HasComments: No Gerrit-Has-Labels: Yes Gerrit-MessageType: comment

1 year, 11 months

1
0
0 0

[M] Change in osmo-trx[master]: transceiver: add experimental viterbi equalizer support

by Hoernchen

Hoernchen has submitted this change. ( https://gerrit.osmocom.org/c/osmo-trx/+/33930 ) ( 14 is the latest approved patch-set. No files were changed between the latest approved patch-set and the submitted one. )Change subject: transceiver: add experimental viterbi equalizer support ...................................................................... transceiver: add experimental viterbi equalizer support The VA is already being used by the ms side and is part of the original gsm design. It only works for gmsk, 4sps, and needs a bit of rx burst scaling and burst shifting. Change-Id: I9d7a4ff72e323832a94d885d5714fcde01ceeb3d --- M CommonLibs/config_defs.h M CommonLibs/trx_vty.c M Transceiver52M/Makefile.am M Transceiver52M/Transceiver.cpp M Transceiver52M/osmo-trx.cpp M Transceiver52M/radioInterface.h 6 files changed, 131 insertions(+), 12 deletions(-) Approvals: pespin: Looks good to me, but someone else must approve laforge: Looks good to me, approved Jenkins Builder: Verified diff --git a/CommonLibs/config_defs.h b/CommonLibs/config_defs.h index bf49bcf..07a3981 100644 --- a/CommonLibs/config_defs.h +++ b/CommonLibs/config_defs.h @@ -67,4 +67,5 @@ double ul_gain; double dl_gain; } overrides; + bool use_va; }; diff --git a/CommonLibs/trx_vty.c b/CommonLibs/trx_vty.c index 55142ef..bd1b0ce 100644 --- a/CommonLibs/trx_vty.c +++ b/CommonLibs/trx_vty.c @@ -339,6 +339,25 @@ return CMD_SUCCESS; } +DEFUN_ATTR(cfg_use_viterbi, cfg_use_viterbi_cmd, + "viterbi-eq (disable|enable)", + "Use viterbi equalizer for gmsk (default=disable)\n" + "Disable VA\n" + "Enable VA\n", + CMD_ATTR_HIDDEN) +{ + struct trx_ctx *trx = trx_from_vty(vty); + + if (strcmp("disable", argv[0]) == 0) + trx->cfg.use_va = false; + else if (strcmp("enable", argv[0]) == 0) + trx->cfg.use_va = true; + else + return CMD_WARNING; + + return CMD_SUCCESS; +} + DEFUN(cfg_swap_channels, cfg_swap_channels_cmd, "swap-channels (disable|enable)", "Swap primary and secondary channels of the PHY (if any)\n" @@ -700,6 +719,8 @@ vty_out(vty, " dl-gain-override %f%s", trx->cfg.overrides.dl_gain, VTY_NEWLINE); if (trx->cfg.overrides.ul_gain_override) vty_out(vty, " ul-gain-override %f%s", trx->cfg.overrides.ul_gain, VTY_NEWLINE); + if (trx->cfg.use_va) + vty_out(vty, " viterbi-eq %s%s", trx->cfg.use_va ? "enable" : "disable", VTY_NEWLINE); trx_rate_ctr_threshold_write_config(vty, " "); for (i = 0; i < trx->cfg.num_chans; i++) { @@ -869,6 +890,7 @@ install_element(TRX_NODE, &cfg_dl_freq_override_cmd); install_element(TRX_NODE, &cfg_ul_gain_override_cmd); install_element(TRX_NODE, &cfg_dl_gain_override_cmd); + install_element(TRX_NODE, &cfg_use_viterbi_cmd); install_node(&chan_node, dummy_config_write); install_element(CHAN_NODE, &cfg_chan_rx_path_cmd); install_element(CHAN_NODE, &cfg_chan_tx_path_cmd); diff --git a/Transceiver52M/Makefile.am b/Transceiver52M/Makefile.am index f8556f0..e82111e 100644 --- a/Transceiver52M/Makefile.am +++ b/Transceiver52M/Makefile.am @@ -40,7 +40,9 @@ ChannelizerBase.cpp \ Channelizer.cpp \ Synthesis.cpp \ - proto_trxd.c + proto_trxd.c \ + grgsm_vitac/grgsm_vitac.cpp \ + grgsm_vitac/viterbi_detector.cc libtransceiver_common_la_SOURCES = \ $(COMMON_SOURCES) \ diff --git a/Transceiver52M/Transceiver.cpp b/Transceiver52M/Transceiver.cpp index c9894f0..fed4289 100644 --- a/Transceiver52M/Transceiver.cpp +++ b/Transceiver52M/Transceiver.cpp @@ -29,6 +29,7 @@ #include <fstream> #include "Transceiver.h" #include <Logger.h> +#include <grgsm_vitac/grgsm_vitac.h> extern "C" { #include "osmo_signal.h" @@ -208,6 +209,8 @@ return false; } + initvita(); + mDataSockets.resize(mChans, -1); @@ -614,6 +617,44 @@ return mRadioInterface->rssiOffset(chan) + cfg->rssi_offset; } +static SoftVector *demodAnyBurst_va(const signalVector &burst, CorrType type, int sps, int rach_max_toa, int tsc) +{ + auto conved_beg = reinterpret_cast<const std::complex<float> *>(&burst.begin()[0]); + std::complex<float> chan_imp_resp[CHAN_IMP_RESP_LENGTH * d_OSR]; + float ncmax; + const unsigned burst_len_bits = 148 + 8; + char demodded_softbits[burst_len_bits]; + SoftVector *bits = new SoftVector(burst_len_bits); + + if (type == CorrType::TSC) { + auto rach_burst_start = get_norm_chan_imp_resp(conved_beg, chan_imp_resp, &ncmax, tsc); + rach_burst_start = std::max(rach_burst_start, 0); + detect_burst_nb(conved_beg, chan_imp_resp, rach_burst_start, demodded_softbits); + } else { + auto normal_burst_start = get_access_imp_resp(conved_beg, chan_imp_resp, &ncmax, 0); + normal_burst_start = std::max(normal_burst_start, 0); + detect_burst_ab(conved_beg, chan_imp_resp, normal_burst_start, demodded_softbits, rach_max_toa); + } + + float *s = &bits->begin()[0]; + for (unsigned int i = 0; i < 148; i++) + s[i] = demodded_softbits[i] * -1; + for (unsigned int i = 148; i < burst_len_bits; i++) + s[i] = 0; + return bits; +} + +#define USE_VA + +#ifdef USE_VA +// signalvector is owning despite claiming not to, but we can pretend, too.. +static void dummy_free(void *wData){}; +static void *dummy_alloc(size_t newSize) +{ + return 0; +}; +#endif + /* * Pull bursts from the FIFO and handle according to the slot * and burst correlation type. Equalzation is currently disabled. @@ -634,6 +675,9 @@ TransceiverState *state = &mStates[chan]; bool ctr_changed = false; double rssi_offset; + static complex burst_shift_buffer[625]; + static signalVector shift_vec(burst_shift_buffer, 0, 625, dummy_alloc, dummy_free); + signalVector *shvec_ptr = &shift_vec; /* Blocking FIFO read */ radioVector *radio_burst = mReceiveFIFO[chan]->read(); @@ -713,8 +757,15 @@ max_toa = (type == RACH || type == EXT_RACH) ? mMaxExpectedDelayAB : mMaxExpectedDelayNB; + if (cfg->use_va) { + // shifted burst copy to make the old demod and detection happy + std::copy(burst->begin() + 20, burst->end() - 20, shift_vec.begin()); + } else { + shvec_ptr = burst; + } + /* Detect normal or RACH bursts */ - rc = detectAnyBurst(*burst, mTSC, BURST_THRESH, cfg->rx_sps, type, max_toa, &ebp); + rc = detectAnyBurst(*shvec_ptr, mTSC, BURST_THRESH, cfg->rx_sps, type, max_toa, &ebp); if (rc <= 0) { if (rc == -SIGERR_CLIP) { LOGCHAN(chan, DTRXDUL, INFO) << "Clipping detected on received RACH or Normal Burst"; @@ -728,7 +779,13 @@ goto ret_idle; } - rxBurst = demodAnyBurst(*burst, (CorrType) rc, cfg->rx_sps, &ebp); + if (cfg->use_va) { + scaleVector(*burst, { (1. / (float)((1 << 14) - 1)), 0 }); + rxBurst = demodAnyBurst_va(*burst, (CorrType)rc, cfg->rx_sps, max_toa, mTSC); + } else { + rxBurst = demodAnyBurst(*shvec_ptr, (CorrType)rc, cfg->rx_sps, &ebp); + } + bi->toa = ebp.toa; bi->tsc = ebp.tsc; bi->ci = ebp.ci; diff --git a/Transceiver52M/osmo-trx.cpp b/Transceiver52M/osmo-trx.cpp index 89a577b..8f5cc85 100644 --- a/Transceiver52M/osmo-trx.cpp +++ b/Transceiver52M/osmo-trx.cpp @@ -77,6 +77,24 @@ static RadioDevice *usrp; static RadioInterface *radio; +/* adjusts read timestamp offset to make the viterbi equalizer happy by including the start tail bits */ +template <typename B> +class rif_va_wrapper : public B { + bool use_va; + + public: + template <typename... Args> + rif_va_wrapper(bool use_va, Args &&...args) : B(std::forward<Args>(args)...), use_va(use_va) + { + } + bool start() override + { + auto rv = B::start(); + B::readTimestamp -= use_va ? 20 : 0; + return rv; + }; +}; + /* Create radio interface * The interface consists of sample rate changes, frequency shifts, * channel multiplexing, and other conversions. The transceiver core @@ -91,17 +109,17 @@ switch (type) { case RadioDevice::NORMAL: - radio = new RadioInterface(usrp, trx->cfg.tx_sps, - trx->cfg.rx_sps, trx->cfg.num_chans); + radio = new rif_va_wrapper<RadioInterface>(trx->cfg.use_va, usrp, trx->cfg.tx_sps, trx->cfg.rx_sps, + trx->cfg.num_chans); break; case RadioDevice::RESAMP_64M: case RadioDevice::RESAMP_100M: - radio = new RadioInterfaceResamp(usrp, trx->cfg.tx_sps, - trx->cfg.rx_sps); + radio = new rif_va_wrapper<RadioInterfaceResamp>(trx->cfg.use_va, usrp, trx->cfg.tx_sps, + trx->cfg.rx_sps); break; case RadioDevice::MULTI_ARFCN: - radio = new RadioInterfaceMulti(usrp, trx->cfg.tx_sps, - trx->cfg.rx_sps, trx->cfg.num_chans); + radio = new rif_va_wrapper<RadioInterfaceMulti>(trx->cfg.use_va, usrp, trx->cfg.tx_sps, trx->cfg.rx_sps, + trx->cfg.num_chans); break; default: LOG(ALERT) << "Unsupported radio interface configuration"; @@ -465,6 +483,12 @@ return -1; } + if (trx->cfg.use_va && + (trx->cfg.egprs || trx->cfg.multi_arfcn || trx->cfg.tx_sps != 4 || trx->cfg.rx_sps != 4)) { + LOG(ERROR) << "Viterbi equalizer only works for gmsk with 4 tx/rx samples per symbol!"; + return -1; + } + return 0; } diff --git a/Transceiver52M/radioInterface.h b/Transceiver52M/radioInterface.h index cd40ddf..b05af78 100644 --- a/Transceiver52M/radioInterface.h +++ b/Transceiver52M/radioInterface.h @@ -75,7 +75,7 @@ public: /** start the interface */ - bool start(); + virtual bool start(); bool stop(); /** initialization */ @@ -151,7 +151,7 @@ public: RadioInterfaceResamp(RadioDevice* wDevice, size_t tx_sps, size_t rx_sps); - ~RadioInterfaceResamp(); + virtual ~RadioInterfaceResamp(); bool init(int type); void close(); @@ -184,7 +184,7 @@ public: RadioInterfaceMulti(RadioDevice* radio, size_t tx_sps, size_t rx_sps, size_t chans = 1); - ~RadioInterfaceMulti(); + virtual ~RadioInterfaceMulti(); bool init(int type); void close(); -- To view, visit https://gerrit.osmocom.org/c/osmo-trx/+/33930 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-trx Gerrit-Branch: master Gerrit-Change-Id: I9d7a4ff72e323832a94d885d5714fcde01ceeb3d Gerrit-Change-Number: 33930 Gerrit-PatchSet: 15 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: pespin <pespin(a)sysmocom.de> Gerrit-MessageType: merged

1 year, 11 months

1
0
0 0

[S] Change in osmocom-bb[master]: trxcon/l1sched: refactor l1sched_mframe_layout()

by fixeria

fixeria has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmocom-bb/+/34216 ) Change subject: trxcon/l1sched: refactor l1sched_mframe_layout() ...................................................................... trxcon/l1sched: refactor l1sched_mframe_layout() Change-Id: Ie1c3679879ef97d42237d6723d1619ac0ef97f6b --- M src/host/trxcon/include/osmocom/bb/l1sched/l1sched.h M src/host/trxcon/src/sched_mframe.c 2 files changed, 19 insertions(+), 10 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmocom-bb refs/changes/16/34216/1 diff --git a/src/host/trxcon/include/osmocom/bb/l1sched/l1sched.h b/src/host/trxcon/include/osmocom/bb/l1sched/l1sched.h index 98282c1..aa5886a 100644 --- a/src/host/trxcon/include/osmocom/bb/l1sched/l1sched.h +++ b/src/host/trxcon/include/osmocom/bb/l1sched/l1sched.h @@ -327,8 +327,8 @@ }; extern const struct l1sched_lchan_desc l1sched_lchan_desc[_L1SCHED_CHAN_MAX]; -const struct l1sched_tdma_multiframe *l1sched_mframe_layout( - enum gsm_phys_chan_config config, int tn); +const struct l1sched_tdma_multiframe * +l1sched_mframe_layout(enum gsm_phys_chan_config config, uint8_t tn); /* Scheduler management functions */ struct l1sched_state *l1sched_alloc(void *ctx, const struct l1sched_cfg *cfg, void *priv); diff --git a/src/host/trxcon/src/sched_mframe.c b/src/host/trxcon/src/sched_mframe.c index 90b7763..0d95e0a 100644 --- a/src/host/trxcon/src/sched_mframe.c +++ b/src/host/trxcon/src/sched_mframe.c @@ -2087,15 +2087,15 @@ }, }; -const struct l1sched_tdma_multiframe *l1sched_mframe_layout( - enum gsm_phys_chan_config config, int tn) +const struct l1sched_tdma_multiframe * +l1sched_mframe_layout(enum gsm_phys_chan_config config, uint8_t tn) { - int i, ts_allowed; - - for (i = 0; i < ARRAY_SIZE(layouts); i++) { - ts_allowed = layouts[i].slotmask & (0x01 << tn); - if (layouts[i].chan_config == config && ts_allowed) - return &layouts[i]; + for (unsigned int i = 0; i < ARRAY_SIZE(layouts); i++) { + if (layouts[i].chan_config != config) + continue; + if (~layouts[i].slotmask & (1 << tn)) + continue; + return &layouts[i]; } return NULL; -- To view, visit https://gerrit.osmocom.org/c/osmocom-bb/+/34216 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmocom-bb Gerrit-Branch: master Gerrit-Change-Id: Ie1c3679879ef97d42237d6723d1619ac0ef97f6b Gerrit-Change-Number: 34216 Gerrit-PatchSet: 1 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-MessageType: newchange

1 year, 11 months

1
0
0 0

[M] Change in docker-playground[master]: Introduce osmo-dia2gsup-master, ttcn3-dia2gsup-test

by fixeria

Attention is currently required from: pespin. fixeria has posted comments on this change. ( https://gerrit.osmocom.org/c/docker-playground/+/34215 ) Change subject: Introduce osmo-dia2gsup-master, ttcn3-dia2gsup-test ...................................................................... Patch Set 2: (5 comments) File osmo_dia2gsup-master/Dockerfile: https://gerrit.osmocom.org/c/docker-playground/+/34215/comment/da478e61_f7d… PS2, Line 19: rebar3 escriptize Maybe copy the resulting `osmo-dia2gsup` to `/usr/local/bin` here, so that it's easier to call it? https://gerrit.osmocom.org/c/docker-playground/+/34215/comment/56786f55_618… PS2, Line 26: --config /data/osmo_dia2gsup.config You're passing the config file twice. Isn't the `ERL_FLAGS` enough? File ttcn3-dia2gsup-test/Makefile: https://gerrit.osmocom.org/c/docker-playground/+/34215/comment/afabce0e_653… PS2, Line 1: ws File ttcn3-dia2gsup-test/jenkins.sh: https://gerrit.osmocom.org/c/docker-playground/+/34215/comment/d225df93_13f… PS2, Line 6: debian-bookworm-erlang This looks like a dependency of `osmo_dia2gsup-$IMAGE_SUFFIX`, and should not be listed here if so. This script is not using this image directly. https://gerrit.osmocom.org/c/docker-playground/+/34215/comment/ee96b35a_1d8… PS2, Line 31: /bin/sh -c "ERL_FLAGS='-config /data/osmo_dia2gsup.config' /tmp/osmo_dia2gsup/_build/default/bin/osmo-dia2gsup --config /data/osmo_dia2gsup.config >/data/osmo_dia2gsup.log 2>&1" This line is already specified in `CMD` of the respective Dockerfile, why duplicating it here? -- To view, visit https://gerrit.osmocom.org/c/docker-playground/+/34215 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: docker-playground Gerrit-Branch: master Gerrit-Change-Id: I646096d5a010df02af75aefa33b023b55997a6d7 Gerrit-Change-Number: 34215 Gerrit-PatchSet: 2 Gerrit-Owner: pespin <pespin(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-CC: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Attention: pespin <pespin(a)sysmocom.de> Gerrit-Comment-Date: Fri, 25 Aug 2023 18:12:20 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: No Gerrit-MessageType: comment

1 year, 11 months

1
0
0 0

2025

2024

2023

2022

gerrit-log August 2023