Changes

testenv: Dockerfile: rebar3 version via apt-cache

When building for Debian Unstable, we need a higher rebar3 version so it
is compatible with the Erlang OTP version in the distribution. Get the
version of rebar3 from the distribution via apt-cache and download it
instead of hardcoding a version number.

Related: OS#7025#note-10
Change-Id: Ie3c8812c41648dd5ca655114dec00ae7d48eaba8

testenv: support Debian Unstable

Related: SYS#7025#note-10
Change-Id: Ib8f04ee1e8b33f6e724ea2a74b1a96f26746e878

testenv: init: check distro argument early

Check the --distro argument and print a proper error, instead of failing
later on with a stack trace if it is unsupported.

Change-Id: I9d51c2b236f10f4c10e991873df47b4f3297e6e1

CCID: Check if reader times out while expecting procedure byte

A case 2 APDU (according to ISO 7816) is sent toward the reader. The
reader expects a procedure byte or a status word. If none of these are
sent by the SIM the reader must timeout and send an error message back
to the host.

Change-Id: Iacd6aacaf8220e69b9b7038e354d54c788d1eb05

CCID: Check if reader handles Abort correctly

The tests aborts an ongoing case 2 APDU (according to ISO 7816). The
SIM responds with a NULL procedure byte so that the reader does not
time out.

The abort messages are sent on both, the control pipe and the bulk
pipe with the same sequence number.

This test fails with osmo-ccid-firmware, because the abort
message/handling is not yet implemented.

Change-Id: Iebe97e73497b8468ebf08faf2c4db700fc76997f

deps: use a fork of titan.ProtocolEmulations.SCCP

The fork adds support for sending UDTS via
ASP_SCCP_N_NOTICE_req.

Change-Id: I8c0de4dab088930d5d7a9febcbebcb5e1f9411e8

stp: add ASP_SCCP_N_NOTICE_req support and f_asp_tx_notice()

Wire ASP_SCCP_N_NOTICE_req as a simple passthrough in TCAP_CODEC_PT,
so test components can send UDTS via the existing SCCP_TCAP port.
Add f_asp_tx_notice() to STP_Tests_TCAP as a convenience wrapper.

Change-Id: I74c05405f1c99d43b4bf863f959275ee51301f2e

stp: Add TCAP_N_NOTICE_req to send a UDTS with TCAP payload

Change-Id: Ie13cecc2f2fc8a20d9646f0fb010b91c2ecf7766

stp: Test that UDTS containing tcap is routed correctly

Related: SYS#8090
Depends: libosmo-sigtran.git I2051aa9e84e553feb17daabb93ba3ce82358e0e4
Change-Id: I264ab424fbc3652249535bbe4f30029defa64c09

CCID: Check if reader times out after first status word

A case 1 APDU is sent towards the reader. The reader expects two status
words. If only the first word is sent by the SIM, the reader must
timeout and send an error message back to the host.

Change-Id: Ic5b892e356c13808555d75746c48c6d8c96ec462

CCID: Send a wrong procedure byte towards the reader

A case 2 request (according to ISO 7816) is sent and a response with
data is expected, but the first byte replied by the SIM is not a
procedure byte, nor a valid status byte.

This text expects the reader to return an error that states an incorrect
received procedure byte.

Change-Id: Iaa0bd8845b3408fba309874fe41c855d8e7efccc

STP_Tests_TCAP: Fix ipa_to_ipa testcase

tcap routing doesn't work the way the test thinks. E.g. a Begin can't be
routed to a special destination because it does't contain a dtid which
is needed to route to the correct ASP.

Change-Id: Ie274232b706348b5ab6d0ac248883d86c53ebb4e

STP_Tests_TCAP: Test routing of fragments in TC_tcap_loadshare_ipa_tcap_range_success_pc

A TCAP Begin will cause the stp to create a cache entry and route
according to that. Send only a TCAP Continue so we check the tcap range
routing logic.

Change-Id: I1cb47174b8f0e69970e75109b9d15b2012e150c3

eIM, IPAd: migrate from SGP.32 V.1.0.1 to SGP.32 V.1.2

SGP.32 V.1.2 is not backward compatible to SGP.32 V.1.0.1. This patch updates
the ASN.1 specification and the TTCN3 templates and types accordingly.

It also applies the bare minimum of changes required to compile the eIM and IPAd
testcases.

Related: SYS#8100
Related: SYS#8101
Change-Id: Id9e112669d9ed8caaec0893c55189eb566fe59be
Co-authored-by: Andreas Eversberg <aeversberg@sysmocom.de>

library: add f_gsm_fn_{sum,sub,diff}()

Change-Id: Ie91ccef5d3f170f4a341a6cf9d3323a896b299e6

bts: TC_pcu_interf_ind: check reporting interval instead of absolute FN

Recent osmo-bts commit 53f0ec29 broke TC_pcu_interf_ind:

  Verdict: fail reason: Odd TDMA frame number := 935

There's no requirement that periodic interference reports must land
on a TDMA frame number satisfying 'fn mod 104 == 0' - only that they
keep arriving every configured averaging period.  Instead, verify
that the interval between consecutive reports equals 104 TDMA frames
(1 SACCH period), using the new f_gsm_fn_sub() helper.

Change-Id: Ia96f2926400bae8c0a6fb8f81eb56b48ca5f2de9
Related: 53f0ec29 ("l1sap: fix duplicate RF RESOURCE INDICATION on clock bootstrap")

stp: for TCAP loadshare: use NI == national

Use a non-zero NI. The TCAP tests don't check it
on the receive path, but the osmo-stp is checking
it if it matches the configuration.

Related: SYS#8061
Change-Id: I51ef302ad72ff3c434fddb39006ce50106a5918f

5gc: disable TC_handover_inter_ngran_xn for latest

Temporarily disable the test for latest as it is known to result in a
crash with open5gs v2.8.0.

Related: OS#7023
Change-Id: Ibf3683683093c41a85bb443e14eab23d23fefa24

Revert "bts: limit stderr logging to ERROR to avoid long write() to ext4 fs"

This reverts commit 3c798765f04dd7428c76e3c5cbf3f2ee27a0af4d.

Reducing logging verbosity did not help: we're still seeing the clock
skew errors and restarts.  This makes debugging harder, and is not
relevant anymore because we've switched to tmpfs [1].

Related: [1] docker-playground.git Id9a93f7149ef7e9bfde1f4fe3a8299ba46645d50
Change-Id: Id234394fda1ce752f93448e2fc03887a097f6b9d

Revert "bts: limit stderr logging to NOTICE to avoid long write() to ext4 fs"

This reverts commit c3ab9ec33f37c6837fb8bcb0fa0bbd2e98781fb7.

Reducing logging verbosity did not help: we're still seeing the clock
skew errors and restarts.  This makes debugging harder, and is not
relevant anymore because we've switched to tmpfs [1].

Related: [1] docker-playground.git Id9a93f7149ef7e9bfde1f4fe3a8299ba46645d50
Change-Id: Ief6c3a6ca3053ccf745a186f7bb440e8540336f4

stp: tcap-loadshare: add TC_tcap_loadshare_m3ua_to_ipa_udts

If a TCAP message arrives which is:
* not a TCAP Begin or Abort (e.g. a TCAP Continue)
* not in the TCAP session cache/tracking
* not have a dTID for a registered TCAP Add Range

The tcap load-share will reject this message with a UDTS

Related: SYS#8061
Change-Id: I181b25aedfd70d156c08197d361560b6d055e65a

bts: f_init(): print testcasename() in the IUT's logs

It's useful to see testcase names in the IUT's logs in some
situations.  For instance, when some testcase crashes the IUT.

Change-Id: I16136f32d7143edfc1638e359ef5b1066c248732
Related: OS#6794

bsc: fix race condition in RSLEm

It may happen that the RSLEm component receives a DChan message
earlier than we suspend the DChan processing.  We saw this once:

  RSL for unknown Dchan
      BSC_Tests.ttcn:12769 BSC_Tests control part
      BSC_Tests.ttcn:5888 TC_ho_int_a5_4 testcase

Let's guard against this by calling f_rslem_suspend() earlier.

Change-Id: I012d4db05538d294a6b4aa85be4a40bb0bad28a3

testenv: run "make deps" every time

The previous logic of not always running "make deps" was useful when
"make deps" would take a long time to complete even if there were no
changes. But this has been fixed in c7e40d6a ("Makefile: avoid fetching
deps if not needed"). Run it every time now in order to fix that deps
are not updated after deps/Makefile gets changed.

Change-Id: I02668530ce36c51a1c641fb816464c9c8e6ded9e

5gc: Initial NAS receiving support

With this patch it is already possible to receive the NAS Auth Request
in the test ConnHdlr.

nas.git dep needs updating to include a new patch fixing decoding of
NG_AUTHENTICATION_REQUEST.

Change-Id: I6697f30f95777fc9aed16a529a669d10f42c57be

5gc: Initial NAS transmitting support

With this patch it is already possible to transmit the NAS Auth Response
in the test ConnHdlr.
We still lack generating a proper RES to have the response properly
accepted.

Change-Id: I951e98217b0150eed58fabb8e50a4782c1a3b510

testenv: podman: Install newer libcurl from bookworm-backports

It was already spotted a few days ago that libcurl4 7.88.1-10+deb12u12,
currently being shipped by debian12, contains some sort of bug where it
doesn't properly format/read HTTP2 requests when used in open5gs.
It was found out that installing a newer version of libcurl4 fixes the
problem.
This patch also fixes the problems spotted in the 5gc testsuite when
running within podman (they didn't show up outside podman because I'm
using Archlinux and hence using newer libcurl 8.13.0 too).

Change-Id: I799eeac73c49b9596502ad98acaa11dba3b88e97

5gc: Fix upfd failing to start outside podman

tuns need to be recreated beforehand with sudo, as already done in
ttcn3-ggsn-tests-ogs, otherwise upfd would need root access to create
them.

Change-Id: I5d3e0c3b7391d312304190acf07b347320481275