April 2022 - gerrit-log - lists.osmocom.org

Change in osmo-e1d[master]: e1oip: Rename e1oip:overflow to e1oip:e1o_overflow

by tnt

Attention is currently required from: manawyrm, roox, laforge. tnt has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-e1d/+/27840 ) Change subject: e1oip: Rename e1oip:overflow to e1oip:e1o_overflow ...................................................................... Patch Set 2: Code-Review+1 -- To view, visit https://gerrit.osmocom.org/c/osmo-e1d/+/27840 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-e1d Gerrit-Branch: master Gerrit-Change-Id: I0af6a2847c8356b011bcff5cc4f5d909c574ea21 Gerrit-Change-Number: 27840 Gerrit-PatchSet: 2 Gerrit-Owner: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: manawyrm <osmocom.account(a)tbspace.de> Gerrit-Reviewer: roox <mardnh(a)gmx.de> Gerrit-Reviewer: tnt <tnt(a)246tNt.com> Gerrit-Attention: manawyrm <osmocom.account(a)tbspace.de> Gerrit-Attention: roox <mardnh(a)gmx.de> Gerrit-Attention: laforge <laforge(a)osmocom.org> Gerrit-Comment-Date: Wed, 20 Apr 2022 12:35:50 +0000 Gerrit-HasComments: No Gerrit-Has-Labels: Yes Gerrit-MessageType: comment

3 years, 2 months

1
0
0 0

Change in osmo-e1d[master]: octoi: differentiate UNDERRUN from SUBSTITUTED in counters

by tnt

Attention is currently required from: manawyrm, roox, laforge. tnt has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-e1d/+/27845 ) Change subject: octoi: differentiate UNDERRUN from SUBSTITUTED in counters ...................................................................... Patch Set 1: (1 comment) File src/octoi/frame_rifo.c: https://gerrit.osmocom.org/c/osmo-e1d/+/27845/comment/757e53ef_2a772371 PS1, Line 141: if (frame_rifo_depth(rifo) == 0) { I think technically the current frame_rifo_depth is "off-by-one" in the sense that it will return 0 if the frame that's about to be read is the last one. -- To view, visit https://gerrit.osmocom.org/c/osmo-e1d/+/27845 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-e1d Gerrit-Branch: master Gerrit-Change-Id: I5448430b09ec10a16decdfd0a4a40850fe2ed1a6 Gerrit-Change-Number: 27845 Gerrit-PatchSet: 1 Gerrit-Owner: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: manawyrm <osmocom.account(a)tbspace.de> Gerrit-Reviewer: roox <mardnh(a)gmx.de> Gerrit-Reviewer: tnt <tnt(a)246tNt.com> Gerrit-Attention: manawyrm <osmocom.account(a)tbspace.de> Gerrit-Attention: roox <mardnh(a)gmx.de> Gerrit-Attention: laforge <laforge(a)osmocom.org> Gerrit-Comment-Date: Wed, 20 Apr 2022 12:34:06 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: No Gerrit-MessageType: comment

3 years, 2 months

1
0
0 0

Change in osmo-e1d[master]: octoi: Disconnect the link when >= 7500 underruns/s

by tnt

Attention is currently required from: manawyrm, roox, laforge. tnt has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-e1d/+/27839 ) Change subject: octoi: Disconnect the link when >= 7500 underruns/s ...................................................................... Patch Set 1: Code-Review+1 -- To view, visit https://gerrit.osmocom.org/c/osmo-e1d/+/27839 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-e1d Gerrit-Branch: master Gerrit-Change-Id: Ie3fffa1c1c20962b40320c8cc088c140b8d64e77 Gerrit-Change-Number: 27839 Gerrit-PatchSet: 1 Gerrit-Owner: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: manawyrm <osmocom.account(a)tbspace.de> Gerrit-Reviewer: roox <mardnh(a)gmx.de> Gerrit-Reviewer: tnt <tnt(a)246tNt.com> Gerrit-Attention: manawyrm <osmocom.account(a)tbspace.de> Gerrit-Attention: roox <mardnh(a)gmx.de> Gerrit-Attention: laforge <laforge(a)osmocom.org> Gerrit-Comment-Date: Wed, 20 Apr 2022 12:30:51 +0000 Gerrit-HasComments: No Gerrit-Has-Labels: Yes Gerrit-MessageType: comment

3 years, 2 months

1
0
0 0

Change in osmo-bts[master]: osmo-bts-trx: amr_loop: do not miss C/I samples

by fixeria

Attention is currently required from: laforge. fixeria has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bts/+/27778 ) Change subject: osmo-bts-trx: amr_loop: do not miss C/I samples ...................................................................... Patch Set 2: (1 comment) File src/osmo-bts-trx/amr_loop.c: https://gerrit.osmocom.org/c/osmo-bts/+/27778/comment/fbb624fa_a65b29b1 PS1, Line 53: if (mi != chan_state->dl_cmr) > I would expect the C/I ratio to stay more or less the same, no matter which codec mode is used. […] This is a perfectly valid point. I'll add this to the commit message. Indeed, C/I is calculated by checking the training sequence bits, i.e. the burst midamble, which remains the same regardless of the current codec rate. -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/27778 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ieb5473ead7200f652b5d0e339e4e252d6567482d Gerrit-Change-Number: 27778 Gerrit-PatchSet: 2 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: pespin <pespin(a)sysmocom.de> Gerrit-Attention: laforge <laforge(a)osmocom.org> Gerrit-Comment-Date: Wed, 20 Apr 2022 12:23:23 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: No Comment-In-Reply-To: laforge <laforge(a)osmocom.org> Comment-In-Reply-To: fixeria <vyanitskiy(a)sysmocom.de> Comment-In-Reply-To: pespin <pespin(a)sysmocom.de> Gerrit-MessageType: comment

3 years, 2 months

1
0
0 0

Change in osmo-e1d[master]: octoi: differentiate UNDERRUN from SUBSTITUTED in counters

by laforge

laforge has uploaded this change for review. ( https://gerrit.osmocom.org/c/osmo-e1d/+/27845 ) Change subject: octoi: differentiate UNDERRUN from SUBSTITUTED in counters ...................................................................... octoi: differentiate UNDERRUN from SUBSTITUTED in counters A real _underrun_ happens if the RIFO is empty (depth == 0) and should be counted different from a mere frame substition due to packet loss. Change-Id: I5448430b09ec10a16decdfd0a4a40850fe2ed1a6 --- M src/octoi/e1oip.c M src/octoi/e1oip.h M src/octoi/frame_rifo.c M src/octoi/octoi.c 4 files changed, 16 insertions(+), 5 deletions(-) git pull ssh://gerrit.osmocom.org:29418/osmo-e1d refs/changes/45/27845/1 diff --git a/src/octoi/e1oip.c b/src/octoi/e1oip.c index 6a5a240..38a22b5 100644 --- a/src/octoi/e1oip.c +++ b/src/octoi/e1oip.c @@ -44,7 +44,8 @@ #include "e1oip.h" static const struct rate_ctr_desc iline_ctr_description[] = { - [LINE_CTR_E1oIP_UNDERRUN] = { "e1oip:underrun", "Frames missing/substituted in IP->E1 direction"}, + [LINE_CTR_E1oIP_UNDERRUN] = { "e1oip:underrun", "Frames underrun / slipped in IP->E1 direction"}, + [LINE_CTR_E1oIP_SUBSTITUTED] = { "e1oip:substituted", "Frames substituted in IP->E1 direction"}, [LINE_CTR_E1oIP_OVERFLOW] = { "e1oip:overflow", "Frames overflowed in IP->E1 direction"}, [LINE_CTR_E1oIP_RX_OUT_OF_ORDER] = { "e1oip:rx:pkt_out_of_order", "Packets out-of-order in IP->E1 direction"}, [LINE_CTR_E1oIP_RX_OUT_OF_WIN] = { "e1oip:rx:pkt_out_of_win", "Packets out-of-rx-window in IP->E1 direction"}, diff --git a/src/octoi/e1oip.h b/src/octoi/e1oip.h index 99b88f0..c992622 100644 --- a/src/octoi/e1oip.h +++ b/src/octoi/e1oip.h @@ -15,6 +15,7 @@ enum e1oip_line_ctr { LINE_CTR_E1oIP_UNDERRUN, + LINE_CTR_E1oIP_SUBSTITUTED, LINE_CTR_E1oIP_OVERFLOW, LINE_CTR_E1oIP_RX_OUT_OF_ORDER, LINE_CTR_E1oIP_RX_OUT_OF_WIN, diff --git a/src/octoi/frame_rifo.c b/src/octoi/frame_rifo.c index 94758e9..19cdb54 100644 --- a/src/octoi/frame_rifo.c +++ b/src/octoi/frame_rifo.c @@ -131,14 +131,19 @@ /*! pull one frames out of the RIFO. * \param rifo The RIFO from which we want to pull frames * \param out Caller-allocated output buffer - * \returns 0 on success; -1 on error (no frame available) */ + * \returns 0 on success; -1 if no frame available; -2 if RIFO depth == 0 */ int frame_rifo_out(struct frame_rifo *rifo, uint8_t *out) { uint32_t next_out_bucket = (rifo->next_out - rifo->buf) / BYTES_PER_FRAME; bool bucket_bit = bucket_bit_get(rifo, next_out_bucket); int rc = 0; - if (!bucket_bit) { + if (frame_rifo_depth(rifo) == 0) { + /* if we don't have any RIFO depth at all, our jitter buffer has + * run empty and most likely there is some fundamental clock sync problem + * somewhere. */ + rc = -2; + } else if (!bucket_bit) { /* caller is supposed to copy/duplicate previous frame */ rc = -1; } else { diff --git a/src/octoi/octoi.c b/src/octoi/octoi.c index 8585f0d..5f4c2f9 100644 --- a/src/octoi/octoi.c +++ b/src/octoi/octoi.c @@ -127,10 +127,14 @@ for (int i = 0; i < fts; i++) { uint8_t *cur = buf + BYTES_PER_FRAME*i; rc = frame_rifo_out(&iline->e1t.rifo, cur); - if (rc < 0) { - iline_ctr_add(iline, LINE_CTR_E1oIP_UNDERRUN, 1); + if (rc == -1) { + iline_ctr_add(iline, LINE_CTR_E1oIP_SUBSTITUTED, 1); /* substitute with last received frame */ memcpy(cur, iline->e1t.last_frame, BYTES_PER_FRAME); + } else if (rc == -2) { + iline_ctr_add(iline, LINE_CTR_E1oIP_UNDERRUN, 1); + /* substitute with all-FF frame */ + memset(cur, 0xff, BYTES_PER_FRAME); } } iline_stat_set(iline, LINE_STAT_E1oIP_E1T_FIFO, frame_rifo_depth(&iline->e1t.rifo)); -- To view, visit https://gerrit.osmocom.org/c/osmo-e1d/+/27845 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-e1d Gerrit-Branch: master Gerrit-Change-Id: I5448430b09ec10a16decdfd0a4a40850fe2ed1a6 Gerrit-Change-Number: 27845 Gerrit-PatchSet: 1 Gerrit-Owner: laforge <laforge(a)osmocom.org> Gerrit-MessageType: newchange

3 years, 2 months

1
0
0 0

Change in osmo-bts[master]: osmo-bts-trx: use C/I in the AMR link adaptation loop

by fixeria

fixeria has submitted this change. ( https://gerrit.osmocom.org/c/osmo-bts/+/27758 ) Change subject: osmo-bts-trx: use C/I in the AMR link adaptation loop ...................................................................... osmo-bts-trx: use C/I in the AMR link adaptation loop Change-Id: Ide84bf864f56020c0265cfb9731615d4f7bad7f5 Related: SYS#5917, OS#4984 --- M include/osmo-bts/scheduler.h M src/common/scheduler.c M src/osmo-bts-trx/amr_loop.c M src/osmo-bts-trx/amr_loop.h M src/osmo-bts-trx/sched_lchan_tchf.c M src/osmo-bts-trx/sched_lchan_tchh.c 6 files changed, 53 insertions(+), 49 deletions(-) Approvals: laforge: Looks good to me, but someone else must approve fixeria: Looks good to me, approved osmith: Looks good to me, but someone else must approve Jenkins Builder: Verified diff --git a/include/osmo-bts/scheduler.h b/include/osmo-bts/scheduler.h index e39315f..eab904c 100644 --- a/include/osmo-bts/scheduler.h +++ b/include/osmo-bts/scheduler.h @@ -114,8 +114,8 @@ /* AMR */ uint8_t codec[4]; /* 4 possible codecs for amr */ int codecs; /* number of possible codecs */ - float ber_sum; /* sum of bit error rates */ - int ber_num; /* number of bit error rates */ + int lqual_cb_sum; /* sum of link quality samples (in cB) */ + int lqual_cb_num; /* number of link quality samples */ uint8_t ul_ft; /* current uplink FT index */ uint8_t dl_ft; /* current downlink FT index */ uint8_t ul_cmr; /* current uplink CMR index */ diff --git a/src/common/scheduler.c b/src/common/scheduler.c index 96189ef..8ec9ccf 100644 --- a/src/common/scheduler.c +++ b/src/common/scheduler.c @@ -1185,8 +1185,8 @@ chan_state->dl_ft = initial_id; chan_state->ul_cmr = initial_id; chan_state->dl_cmr = initial_id; - chan_state->ber_sum = 0; - chan_state->ber_num = 0; + chan_state->lqual_cb_sum = 0; + chan_state->lqual_cb_num = 0; } rc = 0; } diff --git a/src/osmo-bts-trx/amr_loop.c b/src/osmo-bts-trx/amr_loop.c index e5c7fec..5793d03 100644 --- a/src/osmo-bts-trx/amr_loop.c +++ b/src/osmo-bts-trx/amr_loop.c @@ -1,6 +1,7 @@ /* AMR link adaptation loop (see 3GPP TS 45.009, section 3) */ /* (C) 2013 by Andreas Eversberg <jolly(a)eversberg.eu> + * (C) 2022 by sysmocom - s.m.f.c. GmbH <info(a)sysmocom.de> * * All Rights Reserved * @@ -32,16 +33,11 @@ #include "amr_loop.h" void trx_loop_amr_input(struct l1sched_chan_state *chan_state, - int n_errors, int n_bits_total) + const struct l1sched_meas_set *meas_set) { - struct gsm_lchan *lchan = chan_state->lchan; - float ber; - - /* calculate BER (Bit Error Ratio) */ - if (n_bits_total == 0) - ber = 1.0; /* 100% BER */ - else - ber = (float) n_errors / (float) n_bits_total; + const struct gsm_lchan *lchan = chan_state->lchan; + const struct amr_multirate_conf *cfg = &lchan->tch.amr_mr; + int lqual_cb = meas_set->ci_cb; /* cB (centibel) */ /* check if loop is enabled */ if (!chan_state->amr_loop) @@ -51,47 +47,54 @@ if (chan_state->ul_ft != chan_state->dl_cmr) return; - /* count bit errors */ + /* count per-block C/I samples for further averaging */ if (lchan->type == GSM_LCHAN_TCH_H) { - chan_state->ber_num += 2; - chan_state->ber_sum += (ber + ber); + chan_state->lqual_cb_num += 2; + chan_state->lqual_cb_sum += (lqual_cb + lqual_cb); } else { - chan_state->ber_num++; - chan_state->ber_sum += ber; + chan_state->lqual_cb_num++; + chan_state->lqual_cb_sum += lqual_cb; } /* count frames */ - if (chan_state->ber_num < 48) + if (chan_state->lqual_cb_num < 48) return; - /* calculate average (reuse ber variable) */ - ber = chan_state->ber_sum / chan_state->ber_num; + /* calculate average (reuse lqual_cb variable) */ + lqual_cb = chan_state->lqual_cb_sum / chan_state->lqual_cb_num; - /* reset bit errors */ - chan_state->ber_num = 0; - chan_state->ber_sum = 0; + LOGPLCHAN(lchan, DLOOP, LOGL_DEBUG, "AMR link quality (C/I) is %d cB, " + "codec mode=%d\n", lqual_cb, chan_state->ul_ft); - LOGPLCHAN(lchan, DLOOP, LOGL_DEBUG, "Current bit error rate (BER) %.6f " - "codec id %d\n", ber, chan_state->ul_ft); + /* reset the link quality measurements */ + chan_state->lqual_cb_num = 0; + chan_state->lqual_cb_sum = 0; - /* degrade */ if (chan_state->dl_cmr > 0) { - /* degrade, if ber is above threshold FIXME: C/I */ - if (ber > - lchan->tch.amr_mr.mode[chan_state->dl_cmr-1].threshold) { - LOGPLCHAN(lchan, DLOOP, LOGL_DEBUG, "Degrading due to BER %.6f " - "from codec id %d to %d\n", ber, chan_state->dl_cmr, - chan_state->dl_cmr - 1); + /* The threshold/hysteresis is in 0.5 dB steps, convert to cB: + * 1dB is 10cB, so 0.5dB is 5cB - this is why we multiply by 5. */ + const int thresh_lower_cb = cfg->mode[chan_state->dl_cmr - 1].threshold * 5; + + /* Degrade if the link quality is below THR_MX_Dn(i - 1) */ + if (lqual_cb < thresh_lower_cb) { + LOGPLCHAN(lchan, DLOOP, LOGL_INFO, "Degrading AMR codec mode: " + "%d -> %d due to link quality %d cB < THR_MX_Dn=%d cB\n", + chan_state->dl_cmr, chan_state->dl_cmr - 1, + lqual_cb, thresh_lower_cb); chan_state->dl_cmr--; } } else if (chan_state->dl_cmr < chan_state->codecs - 1) { - /* degrade, if ber is above threshold FIXME: C/I*/ - if (ber < - lchan->tch.amr_mr.mode[chan_state->dl_cmr].threshold - - lchan->tch.amr_mr.mode[chan_state->dl_cmr].hysteresis) { - LOGPLCHAN(lchan, DLOOP, LOGL_DEBUG, "Upgrading due to BER %.6f " - "from codec id %d to %d\n", ber, chan_state->dl_cmr, - chan_state->dl_cmr + 1); + /* The threshold/hysteresis is in 0.5 dB steps, convert to cB: + * 1dB is 10cB, so 0.5dB is 5cB - this is why we multiply by 5. */ + const int thresh_upper_cb = cfg->mode[chan_state->dl_cmr].threshold * 5 \ + + cfg->mode[chan_state->dl_cmr].hysteresis * 5; + + /* Upgrade if the link quality is above THR_MX_Up(i) */ + if (lqual_cb > thresh_upper_cb) { + LOGPLCHAN(lchan, DLOOP, LOGL_INFO, "Upgrading AMR codec mode: " + "%d -> %d due to link quality %d cB > THR_MX_Up=%d cB\n", + chan_state->dl_cmr, chan_state->dl_cmr + 1, + lqual_cb, thresh_upper_cb); chan_state->dl_cmr++; } } @@ -102,9 +105,9 @@ if (chan_state->amr_loop == loop) return; if (!chan_state->amr_loop) { - /* reset bit errors */ - chan_state->ber_num = 0; - chan_state->ber_sum = 0; + /* reset the link quality measurements */ + chan_state->lqual_cb_num = 0; + chan_state->lqual_cb_sum = 0; } chan_state->amr_loop = loop; diff --git a/src/osmo-bts-trx/amr_loop.h b/src/osmo-bts-trx/amr_loop.h index 85f73fd..efff76c 100644 --- a/src/osmo-bts-trx/amr_loop.h +++ b/src/osmo-bts-trx/amr_loop.h @@ -11,6 +11,6 @@ */ void trx_loop_amr_input(struct l1sched_chan_state *chan_state, - int n_errors, int n_bits_total); + const struct l1sched_meas_set *meas_set); void trx_loop_amr_set(struct l1sched_chan_state *chan_state, int loop); diff --git a/src/osmo-bts-trx/sched_lchan_tchf.c b/src/osmo-bts-trx/sched_lchan_tchf.c index 8ca0961..fb19dd2 100644 --- a/src/osmo-bts-trx/sched_lchan_tchf.c +++ b/src/osmo-bts-trx/sched_lchan_tchf.c @@ -197,8 +197,6 @@ break; } - if (rc) - trx_loop_amr_input(chan_state, n_errors, n_bits_total); /* only good speech frames get rtp header */ if (rc != GSM_MACBLOCK_LEN && rc >= 4) { if (chan_state->amr_last_dtx == AMR_OTHER) { @@ -225,6 +223,9 @@ /* meas_avg.fn now contains TDMA frame number of the first burst */ fn_begin = meas_avg.fn; + if (tch_mode == GSM48_CMODE_SPEECH_AMR) + trx_loop_amr_input(chan_state, &meas_avg); + /* Check if the frame is bad */ if (rc < 0) { LOGL1SB(DL1P, LOGL_NOTICE, l1ts, bi, "Received bad data (%u/%u)\n", diff --git a/src/osmo-bts-trx/sched_lchan_tchh.c b/src/osmo-bts-trx/sched_lchan_tchh.c index e3ce600..1b4aa40 100644 --- a/src/osmo-bts-trx/sched_lchan_tchh.c +++ b/src/osmo-bts-trx/sched_lchan_tchh.c @@ -233,9 +233,6 @@ break; } - if (rc) - trx_loop_amr_input(chan_state, n_errors, n_bits_total); - /* only good speech frames get rtp header */ if (rc != GSM_MACBLOCK_LEN && rc >= 4) { if (chan_state->amr_last_dtx == AMR_OTHER) { @@ -264,6 +261,9 @@ /* meas_avg.fn now contains TDMA frame number of the first burst */ fn_begin = meas_avg.fn; + if (tch_mode == GSM48_CMODE_SPEECH_AMR) + trx_loop_amr_input(chan_state, &meas_avg); + /* Check if the frame is bad */ if (rc < 0) { LOGL1SB(DL1P, LOGL_NOTICE, l1ts, bi, "Received bad data (%u/%u)\n", -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/27758 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ide84bf864f56020c0265cfb9731615d4f7bad7f5 Gerrit-Change-Number: 27758 Gerrit-PatchSet: 6 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: dexter <pmaier(a)sysmocom.de> Gerrit-Reviewer: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: osmith <osmith(a)sysmocom.de> Gerrit-Reviewer: pespin <pespin(a)sysmocom.de> Gerrit-MessageType: merged

3 years, 2 months

1
0
0 0

Change in osmo-bts[master]: osmo-bts-trx: amr_loop: improve logging in trx_loop_amr_input()

by fixeria

fixeria has submitted this change. ( https://gerrit.osmocom.org/c/osmo-bts/+/27774 ) Change subject: osmo-bts-trx: amr_loop: improve logging in trx_loop_amr_input() ...................................................................... osmo-bts-trx: amr_loop: improve logging in trx_loop_amr_input() Currently we're logging AMR mode *indexes*, not the actual modes. Let's make it cleaner for the user by logging both mode and index. Change-Id: I4feb3a3823799a290a50a48275e45f3331874d1a Related: SYS#5917, OS#4984 --- M src/osmo-bts-trx/amr_loop.c 1 file changed, 12 insertions(+), 11 deletions(-) Approvals: laforge: Looks good to me, approved pespin: Looks good to me, but someone else must approve Jenkins Builder: Verified diff --git a/src/osmo-bts-trx/amr_loop.c b/src/osmo-bts-trx/amr_loop.c index 5793d03..90b156c 100644 --- a/src/osmo-bts-trx/amr_loop.c +++ b/src/osmo-bts-trx/amr_loop.c @@ -37,6 +37,7 @@ { const struct gsm_lchan *lchan = chan_state->lchan; const struct amr_multirate_conf *cfg = &lchan->tch.amr_mr; + const uint8_t mi = chan_state->ul_ft; /* mode index 0..3 */ int lqual_cb = meas_set->ci_cb; /* cB (centibel) */ /* check if loop is enabled */ @@ -44,7 +45,7 @@ return; /* wait for MS to use the requested codec */ - if (chan_state->ul_ft != chan_state->dl_cmr) + if (mi != chan_state->dl_cmr) return; /* count per-block C/I samples for further averaging */ @@ -64,36 +65,36 @@ lqual_cb = chan_state->lqual_cb_sum / chan_state->lqual_cb_num; LOGPLCHAN(lchan, DLOOP, LOGL_DEBUG, "AMR link quality (C/I) is %d cB, " - "codec mode=%d\n", lqual_cb, chan_state->ul_ft); + "codec mode[%u]=%u\n", lqual_cb, mi, cfg->mode[mi].mode); /* reset the link quality measurements */ chan_state->lqual_cb_num = 0; chan_state->lqual_cb_sum = 0; - if (chan_state->dl_cmr > 0) { + if (mi > 0) { /* The threshold/hysteresis is in 0.5 dB steps, convert to cB: * 1dB is 10cB, so 0.5dB is 5cB - this is why we multiply by 5. */ - const int thresh_lower_cb = cfg->mode[chan_state->dl_cmr - 1].threshold * 5; + const int thresh_lower_cb = cfg->mode[mi - 1].threshold * 5; /* Degrade if the link quality is below THR_MX_Dn(i - 1) */ if (lqual_cb < thresh_lower_cb) { LOGPLCHAN(lchan, DLOOP, LOGL_INFO, "Degrading AMR codec mode: " - "%d -> %d due to link quality %d cB < THR_MX_Dn=%d cB\n", - chan_state->dl_cmr, chan_state->dl_cmr - 1, + "[%u]=%u -> [%u]=%u due to link quality %d cB < THR_MX_Dn=%d cB\n", + mi, cfg->mode[mi].mode, mi - 1, cfg->mode[mi - 1].mode, lqual_cb, thresh_lower_cb); chan_state->dl_cmr--; } - } else if (chan_state->dl_cmr < chan_state->codecs - 1) { + } else if (mi < chan_state->codecs - 1) { /* The threshold/hysteresis is in 0.5 dB steps, convert to cB: * 1dB is 10cB, so 0.5dB is 5cB - this is why we multiply by 5. */ - const int thresh_upper_cb = cfg->mode[chan_state->dl_cmr].threshold * 5 \ - + cfg->mode[chan_state->dl_cmr].hysteresis * 5; + const int thresh_upper_cb = cfg->mode[mi].threshold * 5 \ + + cfg->mode[mi].hysteresis * 5; /* Upgrade if the link quality is above THR_MX_Up(i) */ if (lqual_cb > thresh_upper_cb) { LOGPLCHAN(lchan, DLOOP, LOGL_INFO, "Upgrading AMR codec mode: " - "%d -> %d due to link quality %d cB > THR_MX_Up=%d cB\n", - chan_state->dl_cmr, chan_state->dl_cmr + 1, + "[%u]=%u -> [%u]=%u due to link quality %d cB > THR_MX_Up=%d cB\n", + mi, cfg->mode[mi].mode, mi + 1, cfg->mode[mi + 1].mode, lqual_cb, thresh_upper_cb); chan_state->dl_cmr++; } -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/27774 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: I4feb3a3823799a290a50a48275e45f3331874d1a Gerrit-Change-Number: 27774 Gerrit-PatchSet: 3 Gerrit-Owner: fixeria <vyanitskiy(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

3 years, 2 months

1
0
0 0

Change in osmo-bts[master]: osmo-bts-trx: amr_loop: simplify trx_loop_amr_set()

by fixeria

fixeria has submitted this change. ( https://gerrit.osmocom.org/c/osmo-bts/+/27773 ) Change subject: osmo-bts-trx: amr_loop: simplify trx_loop_amr_set() ...................................................................... osmo-bts-trx: amr_loop: simplify trx_loop_amr_set() Change-Id: Ibeed00ef48450edfe1e0d4e179fb4c777b11beb9 Related: SYS#5917, OS#4984 --- M src/osmo-bts-trx/amr_loop.c 1 file changed, 4 insertions(+), 9 deletions(-) Approvals: laforge: Looks good to me, approved pespin: Looks good to me, but someone else must approve Jenkins Builder: Verified diff --git a/src/osmo-bts-trx/amr_loop.c b/src/osmo-bts-trx/amr_loop.c index d6e6aac..e5c7fec 100644 --- a/src/osmo-bts-trx/amr_loop.c +++ b/src/osmo-bts-trx/amr_loop.c @@ -99,18 +99,13 @@ void trx_loop_amr_set(struct l1sched_chan_state *chan_state, int loop) { - if (chan_state->amr_loop && !loop) { - chan_state->amr_loop = 0; + if (chan_state->amr_loop == loop) return; - } - - if (!chan_state->amr_loop && loop) { - chan_state->amr_loop = 1; - + if (!chan_state->amr_loop) { /* reset bit errors */ chan_state->ber_num = 0; chan_state->ber_sum = 0; - - return; } + + chan_state->amr_loop = loop; } -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/27773 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ibeed00ef48450edfe1e0d4e179fb4c777b11beb9 Gerrit-Change-Number: 27773 Gerrit-PatchSet: 3 Gerrit-Owner: fixeria <vyanitskiy(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

3 years, 2 months

1
0
0 0

Change in osmo-e1d[master]: octoi: slip frames if RIFO runs empty in IP->E1 direction

by tnt

Attention is currently required from: laforge. tnt has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-e1d/+/27837 ) Change subject: octoi: slip frames if RIFO runs empty in IP->E1 direction ...................................................................... Patch Set 2: (1 comment) This change is ready for review. Patchset: PS2: I'd just drop this TBH ... Disconnect seems like a better approach as you re-prime everything. And really this should also never happen. Why is still does is probably some bug or mis-config we haven't traced yet and fixing that should be the proper solution. -- To view, visit https://gerrit.osmocom.org/c/osmo-e1d/+/27837 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-e1d Gerrit-Branch: master Gerrit-Change-Id: Ic2a714fedb9a0698c17ef22447904a803c26ebfc Gerrit-Change-Number: 27837 Gerrit-PatchSet: 2 Gerrit-Owner: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: manawyrm <osmocom.account(a)tbspace.de> Gerrit-Reviewer: roox <mardnh(a)gmx.de> Gerrit-Reviewer: tnt <tnt(a)246tNt.com> Gerrit-Attention: laforge <laforge(a)osmocom.org> Gerrit-Comment-Date: Wed, 20 Apr 2022 11:52:47 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: No Gerrit-MessageType: comment

3 years, 2 months

1
0
0 0

Change in osmo-bts[master]: osmo-bts-trx: use C/I in the AMR link adaptation loop

by fixeria

Attention is currently required from: pespin, dexter. fixeria has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bts/+/27758 ) Change subject: osmo-bts-trx: use C/I in the AMR link adaptation loop ...................................................................... Patch Set 5: Code-Review+2 (1 comment) Patchset: PS5: 1+1 -- To view, visit https://gerrit.osmocom.org/c/osmo-bts/+/27758 To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings Gerrit-Project: osmo-bts Gerrit-Branch: master Gerrit-Change-Id: Ide84bf864f56020c0265cfb9731615d4f7bad7f5 Gerrit-Change-Number: 27758 Gerrit-PatchSet: 5 Gerrit-Owner: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: dexter <pmaier(a)sysmocom.de> Gerrit-Reviewer: fixeria <vyanitskiy(a)sysmocom.de> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: osmith <osmith(a)sysmocom.de> Gerrit-Reviewer: pespin <pespin(a)sysmocom.de> Gerrit-Attention: pespin <pespin(a)sysmocom.de> Gerrit-Attention: dexter <pmaier(a)sysmocom.de> Gerrit-Comment-Date: Wed, 20 Apr 2022 11:49:03 +0000 Gerrit-HasComments: Yes Gerrit-Has-Labels: Yes Gerrit-MessageType: comment

3 years, 2 months

1
0
0 0

2025

2024

2023

2022

gerrit-log April 2022