Cellular Network Infrastructure » Radio Access Network » OsmoBTS

I have made some experiments with TC_pcu_ptcch. As far as I can see the mac block is received via the PCUIF and then scheduled into l1ts->dl_prims. I can also see the MAC block dequeued and encoded. That is as far as I can follow so far.

The trxcon/faketrx installation I use is somewhat older, so I think there is no regression in here. Also checking out a slightly earlier version of osmo-bts / libosmocore / osmo-ttcn3-hacks does not help.

I think I know why it's failing for you dexter and also failing with the recent master on Jenkins. In your setup trxcon is too old and does not have GPRS related logic, which is required for recent ttcn3-bts-test. In case of Jenkins, it's a regression caused by the recent trxcon/l1gprs patches. This is what I found in the logging:

20230827103746451 DSCH NOTICE trxcon(0)[0x6120000000a0]: Add a new TDMA timeslot #7 (sched_trx.c:207)
20230827103746451 DSCH NOTICE trxcon(0)[0x6120000000a0]: (Re)configure TDMA timeslot #7 as PDCH (sched_trx.c:274)
20230827103746451 DSCH NOTICE trxcon(0)[0x6120000000a0]: TS7-PDTCH activating (sched_trx.c:476)
20230827103746451 DSCH NOTICE trxcon(0)[0x6120000000a0]: TS7-PTCCH activating (sched_trx.c:476)
20230827103746452 DGPRS ERROR trxcon(0)[0x6120000000a0]: Rx malformed Downlink TBF config (len=4 < 8) (l1gprs.c:568)
20230827103746472 DSCHD ERROR trxcon(0)[0x6120000000a0]: TS7-PDTCH Received bad frame (rc=-1, ber=247/676) at fn=784 (sched_lchan_pdtch.c:90)
20230827103746472 DGPRS DEBUG trxcon(0)[0x6120000000a0]: (PDCH-7) Rx DL BLOCK.ind (PDTCH, fn=784, len=0):  (l1gprs.c:693)
20230827103746472 DGPRS ERROR trxcon(0)[0x6120000000a0]: (PDCH-7) Rx DL BLOCK.ind (fn=784), but this PDCH has no configured TBFs (l1gprs.c:706)
20230827103746490 DGPRS DEBUG trxcon(0)[0x6120000000a0]: (PDCH-7) Rx DL BLOCK.ind (PDTCH, fn=788, len=23): 00 00 0b 6d d3 14 ad 37 1e 88 6f bd af a2 c3 20 ec 6f 02 cc a1 1a c7  (l1gprs.c:693)
20230827103746490 DGPRS ERROR trxcon(0)[0x6120000000a0]: (PDCH-7) Rx DL BLOCK.ind (fn=788), but this PDCH has no configured TBFs (l1gprs.c:706)

The problem is likely caused by my patch:

commit 8b4640ec19d95735363c635b84ca814d9c35ab76
Author: Vadim Yanitskiy <vyanitskiy@sysmocom.de>
Date:   Wed Jun 7 23:06:24 2023 +0700

    l1ctl_proto: add 'start_fn' field to UL/DL TBF CFG.req messages

    Change-Id: Ibb6a05165fe1c81268fb0e3674adae4065e78171
    Related: OS#5500

which was merged on Aug 20th by Pau. The problem is that we did not update the TTCN-3 definitions!

This patch fixes the problem:

https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/34220 library: L1CTL: add 'start_fn' field to UL/DL TBF CFG.req messages [NEW]

pass BTS_Tests.TC_pcu_data_req_pdtch
pass BTS_Tests.TC_pcu_data_req_ptcch
pass BTS_Tests.TC_pcu_ptcch

The patch has been merged.