osmo-msc/tests/msc_vlr/msc_vlr_tests.c

1240 lines
32 KiB
C
Raw Normal View History

/* Osmocom MSC+VLR end-to-end tests */
/* (C) 2017 by sysmocom s.f.m.c. GmbH <info@sysmocom.de>
*
* All Rights Reserved
*
* Author: Neels Hofmeyr <nhofmeyr@sysmocom.de>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation; either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*
*/
#include <getopt.h>
#include <stdlib.h>
#include <osmocom/core/logging.h>
#include <osmocom/core/utils.h>
#include <osmocom/core/msgb.h>
#include <osmocom/core/application.h>
#include <osmocom/gsm/protocol/gsm_04_11.h>
#include <osmocom/gsm/gsup.h>
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
#include <osmocom/msc/gsup_client_mux.h>
#include <osmocom/msc/gsm_04_11.h>
#include <osmocom/msc/debug.h>
#include <osmocom/msc/gsm_04_08.h>
#include <osmocom/msc/transaction.h>
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
#include <osmocom/msc/ran_msg.h>
#include <osmocom/msc/msc_a.h>
#include <osmocom/msc/msc_i.h>
#include <osmocom/msc/msc_t.h>
#include <osmocom/msc/call_leg.h>
#include <osmocom/msc/rtp_stream.h>
#include "msc_vlr_tests.h"
void *msc_vlr_tests_ctx = NULL;
void *msgb_ctx = NULL;
bool _log_lines = false;
struct gsm_network *net = NULL;
const char *gsup_tx_expected = NULL;
bool gsup_tx_confirmed;
struct msgb *dtap_tx_expected = NULL;
bool dtap_tx_confirmed;
enum result_sent lu_result_sent;
enum result_sent cm_service_result_sent;
bool auth_request_sent;
const char *auth_request_expect_rand;
const char *auth_request_expect_autn;
bool cipher_mode_cmd_sent;
bool cipher_mode_cmd_sent_with_imeisv;
const char *cipher_mode_expect_kc;
bool security_mode_ctrl_sent;
const char *security_mode_expect_ck;
const char *security_mode_expect_ik;
bool iu_release_expected = false;
bool iu_release_sent = false;
bool bssap_clear_expected = false;
bool bssap_clear_sent = false;
bool bssap_assignment_expected = false;
bool bssap_assignment_sent = false;
bool iu_rab_assignment_expected = false;
bool iu_rab_assignment_sent = false;
uint32_t cc_to_mncc_tx_expected_msg_type = 0;
const char *cc_to_mncc_tx_expected_imsi = NULL;
bool cc_to_mncc_tx_confirmed = false;
uint32_t cc_to_mncc_tx_got_callref = 0;
enum rtp_direction expecting_crcx = -1;
bool got_crcx = false;
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
extern int ran_dec_dtap_undup_pdisc_ctr_bin(uint8_t pdisc);
/* static state variables for the L3 send sequence numbers */
static uint8_t n_sd[4];
/* patch a correct send sequence number into the given message */
static void patch_l3_seq_nr(struct msgb *msg)
{
struct gsm48_hdr *gh = msgb_l3(msg);
uint8_t pdisc = gsm48_hdr_pdisc(gh);
uint8_t *msg_type_oct = &msg->l3h[1];
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
int bin = ran_dec_dtap_undup_pdisc_ctr_bin(pdisc);
if (bin >= 0 && bin < ARRAY_SIZE(n_sd)) {
/* patch in n_sd into the msg_type octet */
*msg_type_oct = (*msg_type_oct & 0x3f) | ((n_sd[bin] & 0x3) << 6);
//fprintf(stderr, "pdisc=0x%02x bin=%d, patched n_sd=%u\n\n", pdisc, bin, n_sd[bin] & 3);
/* increment N(SD) */
n_sd[bin] = (n_sd[bin] + 1) % 4;
} else {
//fprintf(stderr, "pdisc=0x%02x NO SEQ\n\n", pdisc);
}
}
/* reset L3 sequence numbers (e.g. new RR connection) */
static void reset_l3_seq_nr()
{
memset(n_sd, 0, sizeof(n_sd));
}
struct msgb *msgb_from_hex(const char *label, uint16_t size, const char *hex)
{
A5/n Ciph: request Classmark Update if missing When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we need a ciph algo flag from a Classmark information that is not yet known (usually CM 2 during LU), send a BSSMAP Classmark Request to get it. To manage the intermission of the Classmark Request, add - msc_classmark_request_then_cipher_mode_cmd(), - state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE, - event SUBSCR_CONN_E_CLASSMARK_UPDATE. From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and re-initiate Ciphering Mode. To be able to re-enter the Ciphering Mode algo decision, factor it out into msc_geran_set_cipher_mode(). Rationale: In the following commit, essentially we stopped supporting A5/3 ciphering: commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670 "MSC: Intersect configured A5 algorithms with MS-supported ones" Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3 A5/3 was no longer supported because from that commit on, we strictly checked the MS-supported ciphers, but we did not have Classmark 2 available during Location Updating. This patch changes that: when Classmark 2 is missing, actively request it by a BSSMAP Classmark Request; continue Ciphering only after the Response. Always request missing Classmark, even if a lesser cipher were configured available. If the Classmark Update response fails to come in, cause an attach failure. Instead, we could attempt to use a lesser cipher that is also enabled. That is left as a future feature, should that become relevant. I think it's unlikely. Technically, we could now end up requesting a Classmark Updating both during LU (vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the only time we lack a Classmark is: during Location Updating with A5/3 enabled. A5/1 support is indicated in CM1 which is always available, and A5/3 support is indicated in CM2, which is always available during CM Service Request as well as Paging Response. So this patch has practical relevance only for Location Updating. For networks that permit only A5/3, this patch fixes Location Updating. For networks that support A5/3 and A5/1, so far we always used A5/1 during LU, and after this patch we request CM2 and likely use A5/3 instead. In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works during LU. Also verify that the lack of a Classmark Response results in attach failure. In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is missing during proc_arq_fsm and proves that that code path works, even though the practical relevance is currently zero. It would only become interesting if ciphering algorithms A5/4 and higher became relevant, because support of those would be indicated in Classmark 3, which would always require a Classmark Request. Related: OS#3043 Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore) Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
struct msgb *msg = msgb_alloc_headroom(size, 4, label);
unsigned char *rc;
A5/n Ciph: request Classmark Update if missing When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we need a ciph algo flag from a Classmark information that is not yet known (usually CM 2 during LU), send a BSSMAP Classmark Request to get it. To manage the intermission of the Classmark Request, add - msc_classmark_request_then_cipher_mode_cmd(), - state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE, - event SUBSCR_CONN_E_CLASSMARK_UPDATE. From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and re-initiate Ciphering Mode. To be able to re-enter the Ciphering Mode algo decision, factor it out into msc_geran_set_cipher_mode(). Rationale: In the following commit, essentially we stopped supporting A5/3 ciphering: commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670 "MSC: Intersect configured A5 algorithms with MS-supported ones" Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3 A5/3 was no longer supported because from that commit on, we strictly checked the MS-supported ciphers, but we did not have Classmark 2 available during Location Updating. This patch changes that: when Classmark 2 is missing, actively request it by a BSSMAP Classmark Request; continue Ciphering only after the Response. Always request missing Classmark, even if a lesser cipher were configured available. If the Classmark Update response fails to come in, cause an attach failure. Instead, we could attempt to use a lesser cipher that is also enabled. That is left as a future feature, should that become relevant. I think it's unlikely. Technically, we could now end up requesting a Classmark Updating both during LU (vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the only time we lack a Classmark is: during Location Updating with A5/3 enabled. A5/1 support is indicated in CM1 which is always available, and A5/3 support is indicated in CM2, which is always available during CM Service Request as well as Paging Response. So this patch has practical relevance only for Location Updating. For networks that permit only A5/3, this patch fixes Location Updating. For networks that support A5/3 and A5/1, so far we always used A5/1 during LU, and after this patch we request CM2 and likely use A5/3 instead. In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works during LU. Also verify that the lack of a Classmark Response results in attach failure. In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is missing during proc_arq_fsm and proves that that code path works, even though the practical relevance is currently zero. It would only become interesting if ciphering algorithms A5/4 and higher became relevant, because support of those would be indicated in Classmark 3, which would always require a Classmark Request. Related: OS#3043 Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore) Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
msg->l2h = msg->data;
rc = msgb_put(msg, osmo_hexparse(hex, msg->data, msgb_tailroom(msg)));
OSMO_ASSERT(rc == msg->l2h);
return msg;
}
static const char *gh_type_name(struct gsm48_hdr *gh)
{
return gsm48_pdisc_msgtype_name(gsm48_hdr_pdisc(gh),
gsm48_hdr_msg_type(gh));
}
void gsup_rx(const char *rx_hex, const char *expect_tx_hex)
{
int rc;
struct msgb *msg;
const char *label;
gsup_expect_tx(expect_tx_hex);
msg = msgb_from_hex("gsup", 1024, rx_hex);
label = osmo_gsup_message_type_name(msg->l2h[0]);
fprintf(stderr, "<-- GSUP rx %s: %s\n", label,
osmo_hexdump_nospc(msgb_l2(msg), msgb_l2len(msg)));
/* GSUP read cb takes ownership of msgb */
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
rc = gsup_client_mux_rx(net->gcm->gsup_client, msg);
fprintf(stderr, "<-- GSUP rx %s: vlr_gsupc_read_cb() returns %d\n",
label, rc);
if (expect_tx_hex)
OSMO_ASSERT(gsup_tx_confirmed);
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
bool conn_exists(const struct msub *msub)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct msub *i;
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
if (!msub)
return false;
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
llist_for_each_entry(i, &msub_list, entry) {
if (i == msub)
return true;
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
btw("msub gone");
return false;
}
/* Simplified version of the cm_service_request_concludes() */
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
void conn_conclude_cm_service_req(struct msub *msub, const char *cm_service_use)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
int32_t count;
struct msc_a *msc_a = msub_msc_a(msub);
btw("Concluding CM Service Request");
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
OSMO_ASSERT(conn_exists(msub));
count = osmo_use_count_by(&msc_a->use_count, cm_service_use);
OSMO_ASSERT(count > 0);
OSMO_ASSERT(osmo_use_count_get_put(&msc_a->use_count, cm_service_use, -count) == 0);
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
ASSERT_RELEASE_CLEAR(msc_a->c.ran->type);
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
void dummy_msc_i_action(struct osmo_fsm_inst *fi, uint32_t event, void *data)
{
}
static const struct osmo_fsm_state dummy_msc_i_states[] = {
{
.name = "0",
.in_event_mask = 0xffffffff,
.action = dummy_msc_i_action,
},
};
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct osmo_fsm dummy_msc_i_fsm = {
.name = "dummy_msc_i",
.states = dummy_msc_i_states,
.num_states = ARRAY_SIZE(dummy_msc_i_states),
.log_subsys = DMSC,
.event_names = msc_i_fsm_event_names,
};
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct msc_i *dummy_msc_i_alloc(struct msub *msub, struct ran_infra *ran)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
return msub_role_alloc(g_msub, MSC_ROLE_I, &dummy_msc_i_fsm, struct msc_i, ran);
}
enum osmo_rat_type rx_from_ran = OSMO_RAT_GERAN_A;
struct msub *g_msub = NULL;
void dtap_expect_tx(const char *hex)
{
/* Has the previously expected dtap been received? */
OSMO_ASSERT(!dtap_tx_expected);
if (!hex)
return;
dtap_tx_expected = msgb_from_hex("dtap_tx_expected", 1024, hex);
/* Mask the sequence number out */
if (msgb_length(dtap_tx_expected) >= 2)
dtap_tx_expected->data[1] &= 0x3f;
dtap_tx_confirmed = false;
}
static int _validate_dtap(struct msgb *msg, enum osmo_rat_type to_ran)
{
struct gsm48_hdr *gh = (void*)msg->data;
uint8_t pdisc = gsm48_hdr_pdisc(gh);
uint8_t msgt = gsm48_hdr_msg_type(gh);
btw("DTAP --%s--> MS: %s: %s",
osmo_rat_type_name(to_ran), gh_type_name((void*)msg->data),
osmo_hexdump_nospc(msg->data, msg->len));
if (pdisc == GSM48_PDISC_MM
&& msgt == GSM48_MT_MM_CM_SERV_ACC) {
cm_service_result_sent |= RES_ACCEPT;
talloc_free(msg);
return 0;
}
if (pdisc == GSM48_PDISC_MM
&& msgt == GSM48_MT_MM_CM_SERV_REJ) {
cm_service_result_sent |= RES_REJECT;
talloc_free(msg);
return 0;
}
OSMO_ASSERT(dtap_tx_expected);
/* Mask the sequence number out before comparing */
msg->data[1] &= 0x3f;
if (!msgb_eq_data_print(msg, dtap_tx_expected->data, dtap_tx_expected->len)) {
btw("Expected %s", osmo_hexdump(dtap_tx_expected->data, dtap_tx_expected->len));
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
abort();
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
btw("DTAP matches expected message");
talloc_free(msg);
dtap_tx_confirmed = true;
talloc_free(dtap_tx_expected);
dtap_tx_expected = NULL;
return 0;
}
static void bssap_validate_clear_cmd()
{
OSMO_ASSERT(bssap_clear_expected);
bssap_clear_expected = false;
bssap_clear_sent = true;
}
static void iucs_validate_clear_cmd()
{
OSMO_ASSERT(iu_release_expected);
iu_release_expected = false;
iu_release_sent = true;
}
static int bssap_validate_cipher_mode_cmd(const struct ran_cipher_mode_command *cmd)
{
int i;
const char *got_key;
cipher_mode_cmd_sent = true;
cipher_mode_cmd_sent_with_imeisv = cmd->geran.retrieve_imeisv;
btw("sending Ciphering Mode Command: retrieve_imeisv=%d", cipher_mode_cmd_sent_with_imeisv);
for (i = 0; i < 7; i++) {
if (!(cmd->geran.a5_encryption_mask & (1 << i)))
continue;
btw("...perm algo: A5/%d", i);
}
got_key = osmo_hexdump_nospc(cmd->vec->kc, sizeof(cmd->vec->kc));
btw("...key: %s", got_key);
if (!cipher_mode_expect_kc
|| strcmp(cipher_mode_expect_kc, got_key)) {
log("FAILURE: expected kc=%s", cipher_mode_expect_kc ? : "NULL");
OSMO_ASSERT(false);
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
return 0;
}
static void bssap_validate_assignment_cmd()
{
OSMO_ASSERT(bssap_assignment_expected);
bssap_assignment_expected = false;
bssap_assignment_sent = true;
}
static void iucs_validate_assignment_cmd()
{
OSMO_ASSERT(iu_rab_assignment_expected);
iu_rab_assignment_expected = false;
iu_rab_assignment_sent = true;
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
static int iucs_validate_security_mode_ctrl(const struct ran_cipher_mode_command *cmd)
{
const char *got_ik;
got_ik = osmo_hexdump_nospc(cmd->vec->ik, sizeof(cmd->vec->ik));
btw("sending SecurityModeControl: ik=%s", got_ik);
security_mode_ctrl_sent = true;
if (!security_mode_expect_ik
|| strcmp(security_mode_expect_ik, got_ik)) {
log("FAILURE: expected ik=%s", security_mode_expect_ik ? : "NULL");
OSMO_ASSERT(false);
}
return 0;
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct msgb *dont_ran_encode(struct osmo_fsm_inst *caller_fi, const struct ran_msg *ran_enc_msg)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct msc_role_common *c = caller_fi->priv;
enum osmo_rat_type ran_type = c->ran->type;
const char *ran_name = osmo_rat_type_name(ran_type);
LOG_RAN_ENC(caller_fi, DMSC, LOGL_INFO, "%s on %s\n", ran_msg_type_name(ran_enc_msg->msg_type),
ran_name);
switch (ran_enc_msg->msg_type) {
case RAN_MSG_DTAP:
_validate_dtap(ran_enc_msg->dtap, ran_type);
break;
case RAN_MSG_CLEAR_COMMAND:
switch (ran_type) {
case OSMO_RAT_GERAN_A:
bssap_validate_clear_cmd();
break;
case OSMO_RAT_UTRAN_IU:
iucs_validate_clear_cmd();
break;
default:
OSMO_ASSERT(false);
}
break;
case RAN_MSG_CIPHER_MODE_COMMAND:
switch (ran_type) {
case OSMO_RAT_GERAN_A:
bssap_validate_cipher_mode_cmd(&ran_enc_msg->cipher_mode_command);
break;
case OSMO_RAT_UTRAN_IU:
iucs_validate_security_mode_ctrl(&ran_enc_msg->cipher_mode_command);
break;
default:
OSMO_ASSERT(false);
}
break;
case RAN_MSG_ASSIGNMENT_COMMAND:
switch (ran_type) {
case OSMO_RAT_GERAN_A:
bssap_validate_assignment_cmd();
break;
case OSMO_RAT_UTRAN_IU:
iucs_validate_assignment_cmd();
break;
default:
OSMO_ASSERT(false);
}
break;
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
default:
break;
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
/* We're testing MSC and VLR interaction, not message encoding.
* Return whatever. The test msc_i instance is a dummy and drops these.
* But it must be msg_free()-able.
*/
return msgb_alloc(1, "unused dummy msg");
}
struct ran_infra test_ran_infra[] = {
[OSMO_RAT_GERAN_A] = {
.type = OSMO_RAT_GERAN_A,
.an_proto = OSMO_GSUP_ACCESS_NETWORK_PROTOCOL_TS3G_48006,
.log_subsys = DBSSAP,
.tdefs = msc_tdefs_geran,
.ran_encode = dont_ran_encode,
},
[OSMO_RAT_UTRAN_IU] = {
.type = OSMO_RAT_UTRAN_IU,
.an_proto = OSMO_GSUP_ACCESS_NETWORK_PROTOCOL_TS3G_25413,
.log_subsys = DIUCS,
.tdefs = msc_tdefs_utran,
.ran_encode = dont_ran_encode,
},
};
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
static int fake_msc_a_ran_dec(const struct ran_msg *ran_dec_msg)
{
struct msc_a_ran_dec_data d = {
.from_role = MSC_ROLE_I,
};
return msc_a_ran_decode_cb(g_msub->role[MSC_ROLE_A], &d, ran_dec_msg);
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
void rx_from_ms(struct msgb *msg)
{
struct gsm48_hdr *gh = msgb_l3(msg);
struct ran_msg ran_dec_msg;
struct gsm0808_cell_id cell_id = {
.id_discr = CELL_IDENT_LAI_AND_LAC,
.id.lai_and_lac = {
.plmn = {
.mcc = 1,
.mnc = 2,
},
.lac = 23,
},
};
struct msc_a *msc_a;
log("MSC <--%s-- MS: %s", osmo_rat_type_name(rx_from_ran), gh_type_name(gh));
if (!conn_exists(g_msub))
g_msub = NULL;
if (!g_msub) {
log("new conn");
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
g_msub = msub_alloc(net);
msc_a_alloc(g_msub, &test_ran_infra[rx_from_ran]);
dummy_msc_i_alloc(g_msub, &test_ran_infra[rx_from_ran]);
reset_l3_seq_nr();
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
ran_dec_msg = (struct ran_msg){
.msg_type = RAN_MSG_COMPL_L3,
.compl_l3 = {
.cell_id = &cell_id,
.msg = msg,
},
};
} else {
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
ran_dec_msg = (struct ran_msg){
.msg_type = RAN_MSG_DTAP,
.dtap = msg,
};
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
msc_a = msub_msc_a(g_msub);
msc_a_get(msc_a, __func__);
patch_l3_seq_nr(msg);
fake_msc_a_ran_dec(&ran_dec_msg);
msc_a_put(msc_a, __func__);
if (!conn_exists(g_msub))
g_msub = NULL;
}
void ms_sends_msg(const char *hex)
{
struct msgb *msg;
msg = msgb_from_hex("ms_sends_msg", 1024, hex);
msg->l1h = msg->l2h = msg->l3h = msg->data;
rx_from_ms(msg);
A5/n Ciph: request Classmark Update if missing When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we need a ciph algo flag from a Classmark information that is not yet known (usually CM 2 during LU), send a BSSMAP Classmark Request to get it. To manage the intermission of the Classmark Request, add - msc_classmark_request_then_cipher_mode_cmd(), - state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE, - event SUBSCR_CONN_E_CLASSMARK_UPDATE. From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and re-initiate Ciphering Mode. To be able to re-enter the Ciphering Mode algo decision, factor it out into msc_geran_set_cipher_mode(). Rationale: In the following commit, essentially we stopped supporting A5/3 ciphering: commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670 "MSC: Intersect configured A5 algorithms with MS-supported ones" Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3 A5/3 was no longer supported because from that commit on, we strictly checked the MS-supported ciphers, but we did not have Classmark 2 available during Location Updating. This patch changes that: when Classmark 2 is missing, actively request it by a BSSMAP Classmark Request; continue Ciphering only after the Response. Always request missing Classmark, even if a lesser cipher were configured available. If the Classmark Update response fails to come in, cause an attach failure. Instead, we could attempt to use a lesser cipher that is also enabled. That is left as a future feature, should that become relevant. I think it's unlikely. Technically, we could now end up requesting a Classmark Updating both during LU (vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the only time we lack a Classmark is: during Location Updating with A5/3 enabled. A5/1 support is indicated in CM1 which is always available, and A5/3 support is indicated in CM2, which is always available during CM Service Request as well as Paging Response. So this patch has practical relevance only for Location Updating. For networks that permit only A5/3, this patch fixes Location Updating. For networks that support A5/3 and A5/1, so far we always used A5/1 during LU, and after this patch we request CM2 and likely use A5/3 instead. In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works during LU. Also verify that the lack of a Classmark Response results in attach failure. In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is missing during proc_arq_fsm and proves that that code path works, even though the practical relevance is currently zero. It would only become interesting if ciphering algorithms A5/4 and higher became relevant, because support of those would be indicated in Classmark 3, which would always require a Classmark Request. Related: OS#3043 Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore) Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
msgb_free(msg);
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
void ms_sends_classmark_update(const struct osmo_gsm48_classmark *classmark)
A5/n Ciph: request Classmark Update if missing When the VLR requests a Ciphering Mode with vlr_ops.set_ciph_mode(), and if we need a ciph algo flag from a Classmark information that is not yet known (usually CM 2 during LU), send a BSSMAP Classmark Request to get it. To manage the intermission of the Classmark Request, add - msc_classmark_request_then_cipher_mode_cmd(), - state SUBSCR_CONN_S_WAIT_CLASSMARK_UPDATE, - event SUBSCR_CONN_E_CLASSMARK_UPDATE. From state AUTH_CIPH, switch to state WAIT_CLASSMARK_UPDATE. Once the BSSMAP Classmark Response, is received, switch back to SUBSCR_CONN_S_AUTH_CIPH and re-initiate Ciphering Mode. To be able to re-enter the Ciphering Mode algo decision, factor it out into msc_geran_set_cipher_mode(). Rationale: In the following commit, essentially we stopped supporting A5/3 ciphering: commit 71330720b6efdda2fcfd3e9c0cb45f89e32e5670 "MSC: Intersect configured A5 algorithms with MS-supported ones" Change-Id: Id124923ee52a357cb7d3e04d33f585214774f3a3 A5/3 was no longer supported because from that commit on, we strictly checked the MS-supported ciphers, but we did not have Classmark 2 available during Location Updating. This patch changes that: when Classmark 2 is missing, actively request it by a BSSMAP Classmark Request; continue Ciphering only after the Response. Always request missing Classmark, even if a lesser cipher were configured available. If the Classmark Update response fails to come in, cause an attach failure. Instead, we could attempt to use a lesser cipher that is also enabled. That is left as a future feature, should that become relevant. I think it's unlikely. Technically, we could now end up requesting a Classmark Updating both during LU (vlr_lu_fsm) and CM Service/Paging Response (proc_arq_fsm), but in practice the only time we lack a Classmark is: during Location Updating with A5/3 enabled. A5/1 support is indicated in CM1 which is always available, and A5/3 support is indicated in CM2, which is always available during CM Service Request as well as Paging Response. So this patch has practical relevance only for Location Updating. For networks that permit only A5/3, this patch fixes Location Updating. For networks that support A5/3 and A5/1, so far we always used A5/1 during LU, and after this patch we request CM2 and likely use A5/3 instead. In msc_vlr_test_gsm_ciph, verify that requesting Classmark 2 for A5/3 works during LU. Also verify that the lack of a Classmark Response results in attach failure. In msc_vlr_test_gsm_ciph, a hacky unit test fakes a situation where a CM2 is missing during proc_arq_fsm and proves that that code path works, even though the practical relevance is currently zero. It would only become interesting if ciphering algorithms A5/4 and higher became relevant, because support of those would be indicated in Classmark 3, which would always require a Classmark Request. Related: OS#3043 Depends: I4a2e1d3923e33912579c4180aa1ff8e8f5abb7e7 (libosmocore) Change-Id: I73c7cb6a86624695bd9c0f59abb72e2fdc655131
2018-09-13 01:23:07 +00:00
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct ran_msg ran_dec = {
.msg_type = RAN_MSG_CLASSMARK_UPDATE,
.classmark_update = {
.classmark = classmark,
},
};
fake_msc_a_ran_dec(&ran_dec);
}
static int ms_sends_msg_fake(uint8_t pdisc, uint8_t msg_type)
{
int rc;
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct ran_msg ran_dec;
struct msgb *msg;
struct gsm48_hdr *gh;
msg = msgb_alloc(1024, "ms_sends_msg_fake");
msg->l1h = msg->l2h = msg->l3h = msg->data;
gh = (struct gsm48_hdr*)msgb_put(msg, sizeof(*gh));
gh->proto_discr = pdisc;
gh->msg_type = msg_type;
/* some amount of data, whatever */
msgb_put(msg, 123);
patch_l3_seq_nr(msg);
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
ran_dec = (struct ran_msg){
.msg_type = RAN_MSG_DTAP,
.dtap = msg,
};
rc = fake_msc_a_ran_dec(&ran_dec);
talloc_free(msg);
return rc;
}
static inline void ms_msg_log_err(uint8_t val, uint8_t msgtype)
{
int rc = ms_sends_msg_fake(val, msgtype);
if (rc != -EACCES)
log("Unexpected return value %u != %u for %s/%s",
-rc, -EACCES, gsm48_pdisc_name(val), gsm48_cc_msg_name(msgtype));
}
void thwart_rx_non_initial_requests()
{
log("requests shall be thwarted");
ms_msg_log_err(GSM48_PDISC_CC, GSM48_MT_CC_SETUP);
ms_msg_log_err(GSM48_PDISC_MM, 0x33); /* nonexistent */
ms_msg_log_err(GSM48_PDISC_RR, GSM48_MT_RR_SYSINFO_1);
ms_msg_log_err(GSM48_PDISC_SMS, GSM411_MT_CP_DATA);
}
void send_sms(struct vlr_subscr *receiver,
struct vlr_subscr *sender,
char *str)
{
struct gsm_sms *sms = sms_from_text(receiver, sender->msisdn, 0, str);
gsm411_send_sms(net, receiver, sms);
}
unsigned char next_rand_byte = 0;
/* override, requires '-Wl,--wrap=osmo_get_rand_id' */
int __real_osmo_get_rand_id(uint8_t *buf, size_t num);
int __wrap_osmo_get_rand_id(uint8_t *buf, size_t num)
{
size_t i;
for (i = 0; i < num; i++)
buf[i] = next_rand_byte++;
return 1;
}
/* override, requires '-Wl,--wrap=gsm340_gen_scts' */
void __real_gsm340_gen_scts(uint8_t *scts, time_t time);
void __wrap_gsm340_gen_scts(uint8_t *scts, time_t time)
{
/* Write fixed time bytes for deterministic test results */
osmo_hexparse("07101000000000", scts, 7);
}
const char *paging_expecting_imsi = NULL;
uint32_t paging_expecting_tmsi;
bool paging_sent;
void paging_expect_imsi(const char *imsi)
{
paging_expecting_imsi = imsi;
paging_expecting_tmsi = GSM_RESERVED_TMSI;
}
void paging_expect_tmsi(uint32_t tmsi)
{
paging_expecting_tmsi = tmsi;
paging_expecting_imsi = NULL;
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
/* override, requires '-Wl,--wrap=ran_peers_down_paging' */
int __real_ran_peers_down_paging(struct sccp_ran_inst *sri, enum CELL_IDENT page_where, struct vlr_subscr *vsub,
enum paging_cause cause);
int __wrap_ran_peers_down_paging(struct sccp_ran_inst *sri, enum CELL_IDENT page_where, struct vlr_subscr *vsub,
enum paging_cause cause)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
log("paging request (%s) to %s on %s", paging_cause_name(cause), vlr_subscr_name(vsub),
osmo_rat_type_name(sri->ran->type));
OSMO_ASSERT(paging_expecting_imsi || (paging_expecting_tmsi != GSM_RESERVED_TMSI));
if (paging_expecting_imsi)
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
VERBOSE_ASSERT(strcmp(paging_expecting_imsi, vsub->imsi), == 0, "%d");
if (paging_expecting_tmsi != GSM_RESERVED_TMSI) {
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
VERBOSE_ASSERT(paging_expecting_tmsi, == vsub->tmsi, "0x%08x");
}
paging_sent = true;
return 1;
}
void clear_vlr()
{
struct vlr_subscr *vsub, *n;
llist_for_each_entry_safe(vsub, n, &net->vlr->subscribers, list) {
vlr_subscr_free(vsub);
}
net->authentication_required = false;
net->a5_encryption_mask = (1 << 0);
net->vlr->cfg.check_imei_rqd = false;
net->vlr->cfg.assign_tmsi = false;
net->vlr->cfg.retrieve_imeisv_early = false;
net->vlr->cfg.retrieve_imeisv_ciphered = false;
net->vlr->cfg.auth_tuple_max_reuse_count = 0;
net->vlr->cfg.auth_reuse_old_sets_on_error = false;
rx_from_ran = OSMO_RAT_GERAN_A;
auth_request_sent = false;
auth_request_expect_rand = NULL;
auth_request_expect_autn = NULL;
cipher_mode_cmd_sent = false;
cipher_mode_cmd_sent_with_imeisv = false;
cipher_mode_expect_kc = NULL;
security_mode_ctrl_sent = false;
security_mode_expect_ck = NULL;
security_mode_expect_ik = NULL;
next_rand_byte = 0;
iu_release_expected = false;
iu_release_sent = false;
bssap_clear_expected = false;
bssap_clear_sent = false;
osmo_gettimeofday_override = false;
expecting_crcx = -1;
got_crcx = false;
bssap_assignment_expected = false;
bssap_assignment_sent = false;
iu_rab_assignment_expected = false;
iu_rab_assignment_sent = false;
}
static struct log_info_cat test_categories[] = {
[DMSC] = {
.name = "DMSC",
.description = "Mobile Switching Center",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DRLL] = {
.name = "DRLL",
.description = "A-bis Radio Link Layer (RLL)",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DMM] = {
.name = "DMM",
.description = "Layer3 Mobility Management (MM)",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DRR] = {
.name = "DRR",
.description = "Layer3 Radio Resource (RR)",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DCC] = {
.name = "DCC",
.description = "Layer3 Call Control (CC)",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DVLR] = {
.name = "DVLR",
.description = "Visitor Location Register",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DREF] = {
.name = "DREF",
.description = "Reference Counting",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DPAG] = {
.name = "DPAG",
.description = "Paging Subsystem",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DIUCS] = {
.name = "DIUCS",
.description = "Iu-CS Protocol",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DMNCC] = {
.name = "DMNCC",
.description = "MNCC API for Call Control application",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DBSSAP] = {
.name = "DBSSAP",
.description = "BSSAP Protocol (A Interface)",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
[DSS] = {
.name = "DSS",
.description = "Supplementary Services",
.enabled = 1, .loglevel = LOGL_DEBUG,
},
};
static struct log_info info = {
.cat = test_categories,
.num_cat = ARRAY_SIZE(test_categories),
};
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct gsm_mncc *on_call_release_mncc_sends_to_cc_data = NULL;
int mncc_recv(struct gsm_network *net, struct msgb *msg)
{
struct gsm_mncc *mncc = (void*)msg->data;
log("MSC --> MNCC: callref 0x%x: %s", mncc->callref,
get_mncc_name(mncc->msg_type));
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
if (mncc->msg_type == MNCC_REL_IND && on_call_release_mncc_sends_to_cc_data) {
log("MNCC: callref 0x%x: Call Release triggering %s", mncc->callref,
get_mncc_name(on_call_release_mncc_sends_to_cc_data->msg_type));
mncc_tx_to_cc(net,
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
on_call_release_mncc_sends_to_cc_data);
on_call_release_mncc_sends_to_cc_data = NULL;
return 0;
}
OSMO_ASSERT(cc_to_mncc_tx_expected_msg_type);
if (cc_to_mncc_tx_expected_msg_type != mncc->msg_type) {
log("Mismatch! Expected MNCC msg type: %s",
get_mncc_name(cc_to_mncc_tx_expected_msg_type));
abort();
}
if (strcmp(cc_to_mncc_tx_expected_imsi, mncc->imsi)) {
log("Mismatch! Expected MNCC msg IMSI: '%s', got '%s'",
cc_to_mncc_tx_expected_imsi,
mncc->imsi);
abort();
}
cc_to_mncc_tx_confirmed = true;
cc_to_mncc_tx_got_callref = mncc->callref;
cc_to_mncc_tx_expected_imsi = NULL;
cc_to_mncc_tx_expected_msg_type = 0;
talloc_free(msg);
return 0;
}
struct osmo_gsup_client *
__real_osmo_gsup_client_create2(struct ipaccess_unit *ipa_dev, const char *ip_addr,
unsigned int tcp_port, osmo_gsup_client_read_cb_t read_cb,
struct osmo_oap_client_config *oap_config);
struct osmo_gsup_client *
__wrap_osmo_gsup_client_create2(struct ipaccess_unit *ipa_dev, const char *ip_addr,
unsigned int tcp_port, osmo_gsup_client_read_cb_t read_cb,
struct osmo_oap_client_config *oap_config)
{
struct osmo_gsup_client *gsupc;
gsupc = talloc_zero(msc_vlr_tests_ctx, struct osmo_gsup_client);
OSMO_ASSERT(gsupc);
return gsupc;
}
/* override, requires '-Wl,--wrap=gsup_client_send' */
int __real_osmo_gsup_client_send(struct osmo_gsup_client *gsupc, struct msgb *msg);
int __wrap_osmo_gsup_client_send(struct osmo_gsup_client *gsupc, struct msgb *msg)
{
uint8_t buf[512];
int len;
fprintf(stderr, "GSUP --> HLR: %s: %s\n",
osmo_gsup_message_type_name(msg->data[0]), osmo_hexdump_nospc(msg->data, msg->len));
OSMO_ASSERT(gsup_tx_expected);
OSMO_ASSERT(strlen(gsup_tx_expected) <= (sizeof(buf) * 2));
len = osmo_hexparse(gsup_tx_expected, buf, sizeof(buf));
if (len < 1)
abort();
/* Compare only the length expected. Extra data is fine, to not care about new GSUP IEs invented later. */
if (msg->len < len) {
fprintf(stderr, "ERROR: GSUP message too short, expected '%s'\n", gsup_tx_expected);
abort();
}
if (memcmp(msg->data, buf, len)) {
fprintf(stderr, "ERROR: GSUP message mismatch, expected it to start with '%s'\n", gsup_tx_expected);
abort();
}
talloc_free(msg);
gsup_tx_confirmed = true;
gsup_tx_expected = NULL;
return 0;
}
struct rtp_stream fake_rtp[2] = {
{
.dir = RTP_TO_RAN,
.local = {
.ip = "10.23.42.1",
.port = 99,
},
.remote = {
.ip = "10.23.42.2",
.port = 100,
},
},
{
.dir = RTP_TO_CN,
.local = {
.ip = "10.23.42.1",
.port = 23,
},
.remote = {
.ip = "10.23.42.2",
.port = 42,
},
},
};
void expect_crcx(enum rtp_direction towards)
{
OSMO_ASSERT(expecting_crcx == -1);
expecting_crcx = towards;
got_crcx = false;
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
/* override, requires '-Wl,--wrap=call_leg_ensure_ci' */
int __real_call_leg_ensure_ci(struct call_leg *cl, enum rtp_direction dir, uint32_t call_id, struct gsm_trans *for_trans,
const enum mgcp_codecs *codec_if_known, const struct osmo_sockaddr_str *remote_addr_if_known);
int __wrap_call_leg_ensure_ci(struct call_leg *cl, enum rtp_direction dir, uint32_t call_id, struct gsm_trans *for_trans,
const enum mgcp_codecs *codec_if_known, const struct osmo_sockaddr_str *remote_addr_if_known)
{
if (!cl->rtp[dir]) {
log("MGW <--CRCX to %s-- MSC: callref=0x%x", rtp_direction_name(dir), call_id);
OSMO_ASSERT(expecting_crcx == dir);
expecting_crcx = -1;
got_crcx = true;
call_leg_ensure_rtp_alloc(cl, dir, call_id, for_trans);
if (codec_if_known)
rtp_stream_set_codec(cl->rtp[dir], *codec_if_known);
if (remote_addr_if_known && osmo_sockaddr_str_is_nonzero(remote_addr_if_known))
rtp_stream_set_remote_addr(cl->rtp[dir], remote_addr_if_known);
}
return 0;
}
void crcx_ok(enum rtp_direction dir)
{
struct msc_a *msc_a = msub_msc_a(g_msub);
struct call_leg *cl = msc_a->cc.call_leg;
OSMO_ASSERT(cl);
OSMO_ASSERT(cl->rtp[dir]);
osmo_sockaddr_str_from_str(&cl->rtp[dir]->local, "10.23.23.1", 23);
//osmo_sockaddr_str_from_str(&cl->rtp[dir].remote, "10.42.42.1", 42);
log("MGW --CRCX OK to %s--> MSC", rtp_direction_name(dir));
osmo_fsm_inst_dispatch(cl->fi, CALL_LEG_EV_RTP_STREAM_ADDR_AVAILABLE, cl->rtp[dir]);
}
static int fake_vlr_tx_lu_acc(void *msc_conn_ref, uint32_t send_tmsi)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct msc_a *msc_a = msc_conn_ref;
if (send_tmsi == GSM_RESERVED_TMSI)
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
btw("sending LU Accept for %s", msc_a->c.fi->id);
else
btw("sending LU Accept for %s, with TMSI 0x%08x",
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
msc_a->c.fi->id, send_tmsi);
lu_result_sent |= RES_ACCEPT;
return 0;
}
static int fake_vlr_tx_lu_rej(void *msc_conn_ref, enum gsm48_reject_value cause)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct msc_a *msc_a = msc_conn_ref;
btw("sending LU Reject for %s, cause %u", msc_a->c.fi->id, cause);
lu_result_sent |= RES_REJECT;
return 0;
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
static int fake_vlr_tx_cm_serv_rej(void *msc_conn_ref, enum osmo_cm_service_type cm_service_type,
enum gsm48_reject_value cause)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct msc_a *msc_a = msc_conn_ref;
btw("sending CM Service Reject (%s) for %s, cause: %s",
osmo_cm_service_type_name(cm_service_type), msc_a->c.fi->id, gsm48_reject_value_name(cause));
cm_service_result_sent |= RES_REJECT;
return 0;
}
static int fake_vlr_tx_auth_req(void *msc_conn_ref, struct vlr_auth_tuple *at,
bool send_autn)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct msc_a *msc_a = msc_conn_ref;
char *hex;
bool ok = true;
btw("sending %s Auth Request for %s: tuple use_count=%d key_seq=%d auth_types=0x%x and...",
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
send_autn? "UMTS" : "GSM", msc_a->c.fi->id,
at->use_count, at->key_seq, at->vec.auth_types);
hex = osmo_hexdump_nospc((void*)&at->vec.rand, sizeof(at->vec.rand));
btw("...rand=%s", hex);
if (!auth_request_expect_rand
|| strcmp(hex, auth_request_expect_rand) != 0) {
ok = false;
log("FAILURE: expected rand=%s",
auth_request_expect_rand ? auth_request_expect_rand : "-");
}
if (send_autn) {
hex = osmo_hexdump_nospc((void*)&at->vec.autn, sizeof(at->vec.autn));
btw("...autn=%s", hex);
if (!auth_request_expect_autn
|| strcmp(hex, auth_request_expect_autn) != 0) {
ok = false;
log("FAILURE: expected autn=%s",
auth_request_expect_autn ? auth_request_expect_autn : "-");
}
} else if (auth_request_expect_autn) {
ok = false;
log("FAILURE: no AUTN sent, expected AUTN = %s",
auth_request_expect_autn);
}
if (send_autn)
btw("...expecting res=%s",
osmo_hexdump_nospc((void*)&at->vec.res, at->vec.res_len));
else
btw("...expecting sres=%s",
osmo_hexdump_nospc((void*)&at->vec.sres, sizeof(at->vec.sres)));
auth_request_sent = ok;
return 0;
}
static int fake_vlr_tx_auth_rej(void *msc_conn_ref)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct msc_a *msc_a = msc_conn_ref;
btw("sending Auth Reject for %s", msc_a->c.fi->id);
return 0;
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
void ms_sends_ciphering_mode_complete(const char *inner_ran_msg)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct ran_msg ran_dec;
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
msc_a_get(msub_msc_a(g_msub), __func__);
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
ran_dec = (struct ran_msg){
.msg_type = RAN_MSG_CIPHER_MODE_COMPLETE,
};
fake_msc_a_ran_dec(&ran_dec);
if (inner_ran_msg) {
struct msgb *msg = msgb_from_hex("cipher_mode_complete_ran", 1024, inner_ran_msg);
msg->l1h = msg->l2h = msg->l3h = msg->data;
ran_dec = (struct ran_msg){
.msg_type = RAN_MSG_DTAP,
.dtap = msg,
};
patch_l3_seq_nr(msg);
fake_msc_a_ran_dec(&ran_dec);
msgb_free(msg);
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
msc_a_put(msub_msc_a(g_msub), __func__);
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
if (!conn_exists(g_msub))
g_msub = NULL;
}
void ms_sends_security_mode_complete(uint8_t utran_encryption)
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct ran_msg ran_dec;
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
ran_dec = (struct ran_msg){
.msg_type = RAN_MSG_CIPHER_MODE_COMPLETE,
.cipher_mode_complete.utran_encryption = utran_encryption,
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
};
fake_msc_a_ran_dec(&ran_dec);
if (!conn_exists(g_msub))
g_msub = NULL;
}
void ms_sends_assignment_complete(enum mgcp_codecs assigned_codec)
{
struct ran_msg ran_dec;
ran_dec = (struct ran_msg){
.msg_type = RAN_MSG_ASSIGNMENT_COMPLETE,
.assignment_complete = {
.codec = assigned_codec,
},
};
fake_msc_a_ran_dec(&ran_dec);
if (!conn_exists(g_msub))
g_msub = NULL;
}
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
void ran_sends_clear_complete()
{
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
struct ran_msg ran_dec;
ran_dec = (struct ran_msg){
.msg_type = RAN_MSG_CLEAR_COMPLETE,
};
fake_msc_a_ran_dec(&ran_dec);
if (!conn_exists(g_msub))
g_msub = NULL;
}
const struct timeval fake_time_start_time = { 123, 456 };
void fake_time_start()
{
struct timespec *clock_override;
osmo_gettimeofday_override_time = fake_time_start_time;
osmo_gettimeofday_override = true;
clock_override = osmo_clock_override_gettimespec(CLOCK_MONOTONIC);
OSMO_ASSERT(clock_override);
clock_override->tv_sec = fake_time_start_time.tv_sec;
clock_override->tv_nsec = fake_time_start_time.tv_usec * 1000;
osmo_clock_override_enable(CLOCK_MONOTONIC, true);
fake_time_passes(0, 0);
}
static struct {
bool verbose;
int run_test_nr;
} cmdline_opts = {
.verbose = false,
.run_test_nr = -1,
};
static void print_help(const char *program)
{
printf("Usage:\n"
" %s [-v] [N [N...]]\n"
"Options:\n"
" -h --help show this text.\n"
" -v --verbose print source file and line numbers\n"
" N run only the Nth test (first test is N=1)\n",
program
);
}
static void handle_options(int argc, char **argv)
{
while (1) {
int option_index = 0, c;
static struct option long_options[] = {
{"help", 0, 0, 'h'},
{"verbose", 1, 0, 'v'},
{0, 0, 0, 0}
};
c = getopt_long(argc, argv, "hv",
long_options, &option_index);
if (c == -1)
break;
switch (c) {
case 'h':
print_help(argv[0]);
exit(0);
case 'v':
cmdline_opts.verbose = true;
break;
default:
/* catch unknown options *as well as* missing arguments. */
fprintf(stderr, "Error in command line options. Exiting.\n");
exit(-1);
break;
}
}
}
static void run_tests(int nr)
{
int test_nr;
nr--; /* arg's first test is 1, in here it's 0 */
for (test_nr = 0; msc_vlr_tests[test_nr]; test_nr++) {
size_t talloc_blocks_before_test;
if (nr >= 0 && test_nr != nr)
continue;
if (cmdline_opts.verbose)
fprintf(stderr, "(test nr %d)\n", test_nr + 1);
talloc_blocks_before_test = talloc_total_blocks(msc_vlr_tests_ctx);
msc_vlr_tests[test_nr]();
if (talloc_total_blocks(msc_vlr_tests_ctx) != talloc_blocks_before_test) {
fprintf(stderr, "ERROR: talloc leak: %zu blocks\n",
talloc_total_blocks(msc_vlr_tests_ctx) - talloc_blocks_before_test);
talloc_report_full(msc_vlr_tests_ctx, stderr);
fprintf(stderr, "\n");
}
if (talloc_total_blocks(msgb_ctx) > 1) {
fprintf(stderr, "ERROR: msgb leak:\n");
talloc_report_full(msgb_ctx, stderr);
fprintf(stderr, "\n");
}
if (cmdline_opts.verbose)
fprintf(stderr, "(test nr %d)\n", test_nr + 1);
}
}
struct gsm_network *test_net(void *ctx)
{
struct gsm_network *net = gsm_network_init(ctx, mncc_recv);
net->gsup_server_addr_str = talloc_strdup(net, "no_gsup_server");
net->gsup_server_port = 0;
OSMO_ASSERT(msc_vlr_alloc(net) == 0);
OSMO_ASSERT(net->vlr);
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
OSMO_ASSERT(msc_gsup_client_start(net) == 0);
OSMO_ASSERT(net->gcm);
OSMO_ASSERT(msc_vlr_start(net) == 0);
net->vlr->ops.tx_lu_acc = fake_vlr_tx_lu_acc;
net->vlr->ops.tx_lu_rej = fake_vlr_tx_lu_rej;
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
net->vlr->ops.tx_cm_serv_acc = msc_vlr_tx_cm_serv_acc;
net->vlr->ops.tx_cm_serv_rej = fake_vlr_tx_cm_serv_rej;
net->vlr->ops.tx_auth_req = fake_vlr_tx_auth_req;
net->vlr->ops.tx_auth_rej = fake_vlr_tx_auth_rej;
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
net->vlr->ops.set_ciph_mode = msc_a_vlr_set_cipher_mode;
/* Allocate fake SCCP Ran Instances */
net->a.sri = talloc_zero(net, struct sccp_ran_inst);
*net->a.sri = (struct sccp_ran_inst){
.ran = &test_ran_infra[OSMO_RAT_GERAN_A],
};
INIT_LLIST_HEAD(&net->a.sri->ran_peers);
INIT_LLIST_HEAD(&net->a.sri->ran_conns);
net->iu.sri = talloc_zero(net, struct sccp_ran_inst);
*net->iu.sri = (struct sccp_ran_inst){
.ran = &test_ran_infra[OSMO_RAT_UTRAN_IU],
};
INIT_LLIST_HEAD(&net->iu.sri->ran_peers);
INIT_LLIST_HEAD(&net->iu.sri->ran_conns);
net->mgw.tdefs = g_mgw_tdefs;
mgcp_client_conf_init(&net->mgw.conf);
net->mgw.tdefs = g_mgw_tdefs;
net->mgw.client = mgcp_client_init(net, &net->mgw.conf);
return net;
}
int main(int argc, char **argv)
{
handle_options(argc, argv);
osmo_fsm_term_safely(true);
msc_vlr_tests_ctx = talloc_named_const(NULL, 0, "msc_vlr_tests_ctx");
msgb_ctx = msgb_talloc_ctx_init(msc_vlr_tests_ctx, 0);
osmo_init_logging2(msc_vlr_tests_ctx, &info);
_log_lines = cmdline_opts.verbose;
OSMO_ASSERT(osmo_stderr_target);
log_set_use_color(osmo_stderr_target, 0);
log_set_print_timestamp(osmo_stderr_target, 0);
log_set_print_filename2(osmo_stderr_target, LOG_FILENAME_NONE);
log_set_print_category(osmo_stderr_target, 1);
log_set_print_category_hex(osmo_stderr_target, 0);
log_set_category_filter(osmo_stderr_target, DLSMS, 1, LOGL_DEBUG);
log_set_category_filter(osmo_stderr_target, DLMGCP, 0, LOGL_NOTICE);
if (cmdline_opts.verbose) {
log_set_print_filename2(osmo_stderr_target, LOG_FILENAME_BASENAME);
log_set_print_filename_pos(osmo_stderr_target, LOG_FILENAME_POS_LINE_END);
log_set_use_color(osmo_stderr_target, 1);
log_set_print_level(osmo_stderr_target, 1);
}
net = test_net(msc_vlr_tests_ctx);
osmo_fsm_log_addr(false);
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
osmo_fsm_log_timeouts(cmdline_opts.verbose);
call_leg_init(net);
large refactoring: support inter-BSC and inter-MSC Handover 3GPP TS 49.008 '4.3 Roles of MSC-A, MSC-I and MSC-T' defines distinct roles: - MSC-A is responsible for managing subscribers, - MSC-I is the gateway to the RAN. - MSC-T is a second transitory gateway to another RAN during Handover. After inter-MSC Handover, the MSC-I is handled by a remote MSC instance, while the original MSC-A retains the responsibility of subscriber management. MSC-T exists in this patch but is not yet used, since Handover is only prepared for, not yet implemented. Facilitate Inter-MSC and inter-BSC Handover by the same internal split of MSC roles. Compared to inter-MSC Handover, mere inter-BSC has the obvious simplifications: - all of MSC-A, MSC-I and MSC-T roles will be served by the same osmo-msc instance, - messages between MSC-A and MSC-{I,T} don't need to be routed via E-interface (GSUP), - no call routing between MSC-A and -I via MNCC necessary. This is the largest code bomb I have submitted, ever. Out of principle, I apologize to everyone trying to read this as a whole. Unfortunately, I see no sense in trying to split this patch into smaller bits. It would be a huge amount of work to introduce these changes in separate chunks, especially if each should in turn be useful and pass all test suites. So, unfortunately, we are stuck with this code bomb. The following are some details and rationale for this rather huge refactoring: * separate MSC subscriber management from ran_conn struct ran_conn is reduced from the pivotal subscriber management entity it has been so far to a mere storage for an SCCP connection ID and an MSC subscriber reference. The new pivotal subscriber management entity is struct msc_a -- struct msub lists the msc_a, msc_i, msc_t roles, the vast majority of code paths however use msc_a, since MSC-A is where all the interesting stuff happens. Before handover, msc_i is an FSM implementation that encodes to the local ran_conn. After inter-MSC Handover, msc_i is a compatible but different FSM implementation that instead forwards via/from GSUP. Same goes for the msc_a struct: if osmo-msc is the MSC-I "RAN proxy" for a remote MSC-A role, the msc_a->fi is an FSM implementation that merely forwards via/from GSUP. * New SCCP implementation for RAN access To be able to forward BSSAP and RANAP messages via the GSUP interface, the individual message layers need to be cleanly separated. The IuCS implementation used until now (iu_client from libosmo-ranap) did not provide this level of separation, and needed a complete rewrite. It was trivial to implement this in such a way that both BSSAP and RANAP can be handled by the same SCCP code, hence the new SCCP-RAN layer also replaces BSSAP handling. sccp_ran.h: struct sccp_ran_inst provides an abstract handler for incoming RAN connections. A set of callback functions provides implementation specific details. * RAN Abstraction (BSSAP vs. RANAP) The common SCCP implementation did set the theme for the remaining refactoring: make all other MSC code paths entirely RAN-implementation-agnostic. ran_infra.c provides data structures that list RAN implementation specifics, from logging to RAN de-/encoding to SCCP callbacks and timers. A ran_infra pointer hence allows complete abstraction of RAN implementations: - managing connected RAN peers (BSC, RNC) in ran_peer.c, - classifying and de-/encoding RAN PDUs, - recording connected LACs and cell IDs and sending out Paging requests to matching RAN peers. * RAN RESET now also for RANAP ran_peer.c absorbs the reset_fsm from a_reset.c; in consequence, RANAP also supports proper RESET semantics now. Hence osmo-hnbgw now also needs to provide proper RESET handling, which it so far duly ignores. (TODO) * RAN de-/encoding abstraction The RAN abstraction mentioned above serves not only to separate RANAP and BSSAP implementations transparently, but also to be able to optionally handle RAN on distinct levels. Before Handover, all RAN messages are handled by the MSC-A role. However, after an inter-MSC Handover, a standalone MSC-I will need to decode RAN PDUs, at least in order to manage Assignment of RTP streams between BSS/RNC and MNCC call forwarding. ran_msg.h provides a common API with abstraction for: - receiving events from RAN, i.e. passing RAN decode from the BSC/RNC and MS/UE: struct ran_dec_msg represents RAN messages decoded from either BSSMAP or RANAP; - sending RAN events: ran_enc_msg is the counterpart to compose RAN messages that should be encoded to either BSSMAP or RANAP and passed down to the BSC/RNC and MS/UE. The RAN-specific implementations are completely contained by ran_msg_a.c and ran_msg_iu.c. In particular, Assignment and Ciphering have so far been distinct code paths for BSSAP and RANAP, with switch(via_ran){...} statements all over the place. Using RAN_DEC_* and RAN_ENC_* abstractions, these are now completely unified. Note that SGs does not qualify for RAN abstraction: the SGs interface always remains with the MSC-A role, and SGs messages follow quite distinct semantics from the fairly similar GERAN and UTRAN. * MGW and RTP stream management So far, managing MGW endpoints via MGCP was tightly glued in-between GSM-04.08-CC on the one and MNCC on the other side. Prepare for switching RTP streams between different RAN peers by moving to object-oriented implementations: implement struct call_leg and struct rtp_stream with distinct FSMs each. For MGW communication, use the osmo_mgcpc_ep API that has originated from osmo-bsc and recently moved to libosmo-mgcp-client for this purpose. Instead of implementing a sequence of events with code duplication for the RAN and CN sides, the idea is to manage each RTP stream separately by firing and receiving events as soon as codecs and RTP ports are negotiated, and letting the individual FSMs take care of the MGW management "asynchronously". The caller provides event IDs and an FSM instance that should be notified of RTP stream setup progress. Hence it becomes possible to reconnect RTP streams from one GSM-04.08-CC to another (inter-BSC Handover) or between CC and MNCC RTP peers (inter-MSC Handover) without duplicating the MGCP code for each transition. The number of FSM implementations used for MGCP handling may seem a bit of an overkill. But in fact, the number of perspectives on RTP forwarding are far from trivial: - an MGW endpoint is an entity with N connections, and MGCP "sessions" for configuring them by talking to the MGW; - an RTP stream is a remote peer connected to one of the endpoint's connections, which is asynchronously notified of codec and RTP port choices; - a call leg is the higher level view on either an MT or MO side of a voice call, a combination of two RTP streams to forward between two remote peers. BSC MGW PBX CI CI [MGW-endpoint] [--rtp_stream--] [--rtp_stream--] [----------------call_leg----------------] * Use counts Introduce using the new osmo_use_count API added to libosmocore for this purpose. Each use token has a distinct name in the logging, which can be a globally constant name or ad-hoc, like the local __func__ string constant. Use in the new struct msc_a, as well as change vlr_subscr to the new osmo_use_count API. * FSM Timeouts Introduce using the new osmo_tdef API, which provides a common VTY implementation for all timer numbers, and FSM state transitions with the correct timeout. Originated in osmo-bsc, recently moved to libosmocore. Depends: Ife31e6798b4e728a23913179e346552a7dd338c0 (libosmocore) Ib9af67b100c4583342a2103669732dab2e577b04 (libosmocore) Id617265337f09dfb6ddfe111ef5e578cd3dc9f63 (libosmocore) Ie9e2add7bbfae651c04e230d62e37cebeb91b0f5 (libosmo-sccp) I26be5c4b06a680f25f19797407ab56a5a4880ddc (osmo-mgw) Ida0e59f9a1f2dd18efea0a51680a67b69f141efa (osmo-mgw) I9a3effd38e72841529df6c135c077116981dea36 (osmo-mgw) Change-Id: I27e4988e0371808b512c757d2b52ada1615067bd
2018-12-07 13:47:34 +00:00
OSMO_ASSERT(osmo_fsm_register(&dummy_msc_i_fsm) == 0);
clear_vlr();
if (optind >= argc)
run_tests(-1);
else {
int arg;
long int nr;
for (arg = optind; arg < argc; arg++) {
errno = 0;
nr = strtol(argv[arg], NULL, 10);
if (errno) {
fprintf(stderr, "Invalid argument: %s\n",
argv[arg]);
exit(1);
}
run_tests(nr);
}
}
printf("Done\n");
return 0;
}