cellular-infrastructure
/
osmo-pcu
mirror of https://gerrit.osmocom.org/osmo-pcu
9
0
Fork 1
Code Issues Releases Wiki Activity
osmo-pcu/src/bts.h

342 lines
8.8 KiB
C
Raw Normal View History

bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
/* bts.h
*
* Copyright (C) 2012 Ivan Klyuchnikov
* Copyright (C) 2013 by Holger Hans Peter Freyther
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
*/
#pragma once
#ifdef __cplusplus
extern "C" {
#include <osmocom/core/linuxlist.h>
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
#include <osmocom/core/rate_ctr.h>
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
#include <osmocom/core/timer.h>
}
bts: Introduce a PollController that has the responsibility to poll For each frame indication received by the BTS the poll controller is asked to expire timedout entries.
2013-10-17 18:12:37 +00:00
#include "poll_controller.h"
sba: Create a SBAController that will manage the sbas for a BTS The PollController is a friend of the SBAController and is allowed to access the internal list. The list is hidden from everyone else. This is done because the calculation of timeout should belong into the PollController and not into the SBAController.
2013-10-19 18:47:12 +00:00
#include "sba.h"
ta: Create TimingAdvance class and make it belong to the BTS This allows us to easily flush the state in case a PCU/BTS connection is lost.
2013-10-19 18:04:57 +00:00
#include "ta.h"
bts/tbf: Move the lists into the BTS and do the look-up from the BTS The list belongs to the BTS. This makes cleaning this up more easy and establishes a hierachy of resources that start from the BTS. The debug_diagram code is now broken.
2013-10-19 19:10:38 +00:00
#include "tbf.h"
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
#endif
#include <stdint.h>
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
struct BTS;
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
/*
* PDCH instance
*/
struct gprs_rlcmac_pdch {
pdch: Move enable/disable into the PDCH code When a PDCH is disabled all resources should be freed. This is currently not possible as the PDCH does not know where it belongs to. On top of that the list (and other resources) should be properly initialized on construction so that disable() is idempotent and does not check if it was disabled. During the re-factoring I noticed that during a sysmobts re-start some resources are not freed. I left a warning in the code to resolve this issue later.
2013-10-19 15:37:48 +00:00
#ifdef __cplusplus
pdch: Move paging dequeue into the PDCH object Rely on packet_paging_request returning NULL in case the queue is empty. We should move the write_packet_paging_request into a separate file/object as well.
2013-10-19 16:15:44 +00:00
struct gprs_rlcmac_paging *dequeue_paging();
struct msgb *packet_paging_request();
bts/pdch: Move the adding of paging to the BTS/PDCH objects Only the gprs_rlcmac_pdch will manipulate the paging list now. There can be various more refactorings of the code but they can be done later. E.g. on memory allocation failure we can continue instead of leaving the code, we should also set any_tbf only after things have been paged.
2013-10-19 16:28:59 +00:00
void add_paging(struct gprs_rlcmac_paging *pag);
pdch: Move paging dequeue into the PDCH object Rely on packet_paging_request returning NULL in case the queue is empty. We should move the write_packet_paging_request into a separate file/object as well.
2013-10-19 16:15:44 +00:00
pdch: Simplify the reset code, rename variables to XYZ_no Simplify the reset code now that the PDCH can know where it is located. Rename the variables in the sba to trx_no and ts_no as it stores the number and not the actual thing.
2013-10-20 14:37:05 +00:00
void free_resources();
pdch: Move enable/disable into the PDCH code When a PDCH is disabled all resources should be freed. This is currently not possible as the PDCH does not know where it belongs to. On top of that the list (and other resources) should be properly initialized on construction so that disable() is idempotent and does not check if it was disabled. During the re-factoring I noticed that during a sysmobts re-start some resources are not freed. I left a warning in the code to resolve this issue later.
2013-10-19 15:37:48 +00:00
bool is_enabled() const;
void enable();
void disable();
pdch: Instead of passing bts, trx, ts use the pdch All dispatching will go through the PDCH. This will clean a lot of the look-ups inside the gprs_rlcmac_data.c and continue with adding structure to the pcu code.
2013-10-26 14:42:38 +00:00
/* dispatching of messages */
int rcv_block(uint8_t *data, uint8_t len, uint32_t fn, int8_t rssi);
pdch: Remove the the bts parameter from rcv_data_block_acknowledged
2013-10-26 16:39:36 +00:00
gprs_rlcmac_bts *bts_data() const;
BTS *bts() const;
pdch: Remove the trx_no/ts_no parameter and use/caclulate it on demand Simplify the depedencies and use the inline functions when we need to figure out the numbers.
2013-10-26 16:53:16 +00:00
uint8_t trx_no() const;
gprs_rlcmac_pdch: Don't access private members This patch introduces methods to get ul and dl tbf by tfi and uses them in gprs_rlcmac_sched. Sponsored by: On-Waves ehf
2014-06-04 12:56:09 +00:00
bts: Return the special type for {ul,dl}_tbf_by_* functions Start returning the special type instead of the base gprs_rlcmac_tbf when retrieving a TBF. Ticket: SYS#389 Sponsored-by: On-Waves ehf
2014-07-10 15:44:06 +00:00
struct gprs_rlcmac_ul_tbf *ul_tbf_by_tfi(uint8_t tfi);
struct gprs_rlcmac_dl_tbf *dl_tbf_by_tfi(uint8_t tfi);
pdch: Move enable/disable into the PDCH code When a PDCH is disabled all resources should be freed. This is currently not possible as the PDCH does not know where it belongs to. On top of that the list (and other resources) should be properly initialized on construction so that disable() is idempotent and does not check if it was disabled. During the re-factoring I noticed that during a sysmobts re-start some resources are not freed. I left a warning in the code to resolve this issue later.
2013-10-19 15:37:48 +00:00
#endif
uint8_t m_is_enabled; /* TS is enabled */
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
uint8_t tsc; /* TSC of this slot */
uint8_t next_ul_tfi; /* next uplink TBF/TFI to schedule (0..31) */
uint8_t next_dl_tfi; /* next downlink TBF/TFI to schedule (0..31) */
sched: Attempt to improve the fairness and schedule UL/AL ACK/ASS It is possible that certain UL ACK messages are not sent when there are many many uplink and downlink assignments. Try to be more fair and schedule them round-robin. This way no starvation should occur.
2013-11-26 15:39:28 +00:00
uint8_t next_ctrl_prio; /* next kind of ctrl message to schedule */
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
struct llist_head paging_list; /* list of paging messages */
uint32_t last_rts_fn; /* store last frame number of RTS */
bts: Add backpointers to the PDCH and TRX structures This will allow to kill various parameters from all the functions as we can walk back.
2013-10-20 08:14:03 +00:00
/* back pointers */
struct gprs_rlcmac_trx *trx;
uint8_t ts_no;
pdch/tbf: Move gprs_rlcmac_rcv_data_block_acknowledged into the pdch Move the method into the PDCH. Extract the finding of TLLI into a new class called Decoding. Move the assemble and forward LLC frames into the TBF as it is poking in the internals of the TBF.
2013-10-26 15:32:04 +00:00
#ifdef __cplusplus
private:
pdch: Remove the trx_no/ts_no parameter and use/caclulate it on demand Simplify the depedencies and use the inline functions when we need to figure out the numbers.
2013-10-26 16:53:16 +00:00
int rcv_data_block_acknowledged(uint8_t *data, uint8_t len, int8_t rssi);
int rcv_control_block(bitvec *rlc_block, uint32_t fn);
pdch/tbf: Move gprs_rlcmac_rcv_data_block_acknowledged into the pdch Move the method into the PDCH. Extract the finding of TLLI into a new class called Decoding. Move the assemble and forward LLC frames into the TBF as it is poking in the internals of the TBF.
2013-10-26 15:32:04 +00:00
bts: Pass the Packet_Control_Acknowledgement_t into the recv method
2013-10-27 12:35:30 +00:00
void rcv_control_ack(Packet_Control_Acknowledgement_t *, uint32_t fn);
bts: Use Packet_Downlink_Ack_Nack_t as parameter
2013-10-27 10:41:01 +00:00
void rcv_control_dl_ack_nack(Packet_Downlink_Ack_Nack_t *, uint32_t fn);
bts: Simplify the code and use Packet_Resource_Request_t*
2013-10-27 10:37:53 +00:00
void rcv_resource_request(Packet_Resource_Request_t *t, uint32_t fn);
bts: Work with the Packet_Measurement_Report_t
2013-10-27 10:19:13 +00:00
void rcv_measurement_report(Packet_Measurement_Report_t *t, uint32_t fn);
bts: Ensure tbf direction with OSMO_ASSERT() In the lookup functions make sure that we are actually returning tbfs with the expected direction. Ticket: SYS#389 Sponsored-by: On-Waves ehf
2014-07-10 15:44:06 +00:00
gprs_rlcmac_tbf *tbf_from_list_by_tfi(struct llist_head *tbf_list, uint8_t tfi,
enum gprs_rlcmac_tbf_direction dir);
pdch/tbf: Move gprs_rlcmac_rcv_data_block_acknowledged into the pdch Move the method into the PDCH. Extract the finding of TLLI into a new class called Decoding. Move the assemble and forward LLC frames into the TBF as it is poking in the internals of the TBF.
2013-10-26 15:32:04 +00:00
#endif
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
};
struct gprs_rlcmac_trx {
void *fl1h;
uint16_t arfcn;
struct gprs_rlcmac_pdch pdch[8];
struct gprs_rlcmac_tbf *ul_tbf[32]; /* array of UL TBF, by UL TFI */
struct gprs_rlcmac_tbf *dl_tbf[32]; /* array of DL TBF, by DL TFI */
bts: Add backpointers to the PDCH and TRX structures This will allow to kill various parameters from all the functions as we can walk back.
2013-10-20 08:14:03 +00:00
/* back pointers */
struct BTS *bts;
uint8_t trx_no;
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
};
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
/**
* This is the data from C. As soon as our minimal compiler is gcc 4.7
* we can start to compile pcu_vty.c with c++ and remove the split.
*/
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
struct gprs_rlcmac_bts {
uint8_t bsic;
uint8_t fc_interval;
bssgp: Add VTY command to Limit the bucket size by time Currently the bucket size is by default being computed based on the leak rate and the expected life time of LLC frames. The latter is either taken from 'queue lifetime' (if given) or a fixed value is used. Using 'queue lifetime' has the drawback that it sets a 'hard' limit, since frames will be dropped if they stay in the queue for a longer time. This commit adds a VTY command to specifically set the time used for the computation of the bucket size advertised to the SGSN. It does not affect the PCUs queue handling in any way. If the bucket time is not set (or if the 'no' command has been used), the old behaviour (see above) is applied. The following VTY commands are added (config-pcu node): - flow-control bucket-time <1-65534> Sets the time in centisecs - no flow-control bucket-time Don't use this time Ticket: OW#1432 Sponsored-by: On-Waves ehf
2015-05-06 08:47:30 +00:00
uint16_t fc_bucket_time;
bssgp: Make BVC bucket size / leak rate configurable Currently the FLOW-CONTROL_BVC message contains fixed values: The tag is 1, the BVC bucket size is 6MB, the BVC bucket leak rate is 820kbit/s, the MS bucket size is 50kB, and the MS leak rate is 50kbit/s. This commit makes the BVC parameters configurable and adds the following VTY commands: - flow-control force-bvc-bucket-size <1-6553500> - no flow-control force-bvc-bucket-size - flow-control force-bvc-leak-rate <1-6553500> - no flow-control force-bvc-leak-rate - flow-control force-ms-bucket-size <1-6553500> - no flow-control force-ms-bucket-size - flow-control force-ms-leak-rate <1-6553500> - no flow-control force-ms-leak-rate The 'no' variants restore the default behaviour. Sponsored-by: On-Waves ehf
2015-04-21 10:56:48 +00:00
uint32_t fc_bvc_bucket_size;
uint32_t fc_bvc_leak_rate;
uint32_t fc_ms_bucket_size;
uint32_t fc_ms_leak_rate;
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
uint8_t cs1;
uint8_t cs2;
uint8_t cs3;
uint8_t cs4;
uint8_t initial_cs_dl, initial_cs_ul;
uint8_t force_cs; /* 0=use from BTS 1=use from VTY */
uint16_t force_llc_lifetime; /* overrides lifetime from SGSN */
tbf: Use a hysteresis when discarding DL LLC frames Currently single LLC blocks are discarded when the PDU lifetime expires. If an IP packet has been fragmented either on the IP or on the LLC layer and is therefore distributed over several LLC frames, the kept fragments are transmitted and then discarded by the MS because of the missing PDU. This can cause massive IP packet loss when there are many fragmented packets (e.g. when trying 'ping -s1800' or if the GGSN chops downlink IP packets into several SNDCP packets). On the other hand, discarding too many packets might disturb the congestion handling of TCP. Dropping plain TCP ACKs might also hinder flow control and congestion avoidance. This commit adds a hysteresis algorithm to the LLC discard loop. If an LLC message's age reaches the high water mark, further message's with an age above the low water mark are discarded, too. This is aborted, if a GMM, a non-UI, or a small message is detected. In these cases, that message is kept. The following VTY commands are added (pcu config node): - queue hysteresis <1-65535> set the difference between high (lifetime) and low watermark in centiseconds - no queue hysteresis disable this feature (default) Since the SGSN will most probably send all fragments of a single N-PDU without much delay between them, a value slightly above the average transmission delay jitter between SGSN and PCU is probably a sensible value to discard all fragments of a single IP packet. This is an experimental feature that might be replaced by more advanced means of active queue management in the future. Sponsored-by: On-Waves ehf
2015-03-20 11:02:42 +00:00
uint32_t llc_discard_csec;
tbf: Force ACK after the last DL LCC frame has been received If the protocol layers above LLC (e.g. TCP) need an acknowledgement to continue, it can take up to 400ms (single TS) until the MS is polled for Ack/Nack which it can use to request an uplink TBF quickly. The 400ms result from requesting an DL Ack/Nack every 20 RLC blocks until all pending LLC frames have been sent. Especially TCP's slow start mechanism can lead to a high delay at the start of the connection, since the sender will eventually stop after having sent the first packets (up to 4 (RFC2581) or 10 (RFC6928)). This commit modifies append_data() to (re-)start a timer every time it handles an LLC packet and to request an Ack/Nack every time it expires. So if the server ceases to send IP packets, the MS is polled in the assumption, that the server is waiting for an ACK. The following VTY commands are added (pcu node): - queue idle-ack-delay <1-65535> timeout in centiseconds - no queue idle-ack-delay disable this feature (default) A sensible value is 10 (100ms) that at gave promising results when testing locally. Sponsored-by: On-Waves ehf
2015-04-02 11:58:09 +00:00
uint32_t llc_idle_ack_csec;
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
uint8_t t3142;
uint8_t t3169;
uint8_t t3191;
uint16_t t3193_msec;
uint8_t t3195;
uint8_t n3101;
uint8_t n3103;
uint8_t n3105;
struct gprs_rlcmac_trx trx[8];
int (*alloc_algorithm)(struct gprs_rlcmac_bts *bts,
struct gprs_rlcmac_tbf *old_tbf,
struct gprs_rlcmac_tbf *tbf, uint32_t cust, uint8_t single);
uint32_t alloc_algorithm_curst; /* options to customize algorithm */
uint8_t force_two_phase;
uint8_t alpha, gamma;
tbf: Implement delayed release of a downlink TBF Currently a DL TBF is immediately closed, when the LLC queue is drained. This will lead to a new DL assignment if data is received afterwards. In addition, it is not possible to keep the PACCH open to poll the MS for UL establishment requests there. GSM 44.060, 9.3.1a suggests to delay the release of an inactive TBF for some time (max 5s). This commit mainly changes create_new_bsn() to send LLC dummy commands as filler if no LLC data is available until keep_open() returns false. The keep_open() functions returns true unless a configurable time has passed after the LLC data store drained. By default, that time is not set which causes keep_open() to always return false, so that delayed release is effectively disabled. The following VTY commands are added: - dl-tbf-idle-time <1-5000> to set the delay in ms - no dl-tbf-idle-time to disable delayed release Ticket: #556 Sponsored-by: On-Waves ehf
2015-03-19 10:22:38 +00:00
uint32_t dl_tbf_idle_msec; /* hold time for idle DL TBFs */
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
bts/tbf: Move the lists into the BTS and do the look-up from the BTS The list belongs to the BTS. This makes cleaning this up more easy and establishes a hierachy of resources that start from the BTS. The debug_diagram code is now broken.
2013-10-19 19:10:38 +00:00
/* TBF handling, make private or move into TBFController */
/* list of uplink TBFs */
struct llist_head ul_tbfs;
/* list of downlink TBFs */
struct llist_head dl_tbfs;
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
/**
* Point back to the C++ object. This is used during the transition
* period.
*/
struct BTS *bts;
};
#ifdef __cplusplus
/**
* I represent a GSM BTS. I have one or more TRX, I know the current
* GSM time and I have controllers that help with allocating resources
* on my TRXs.
*/
struct BTS {
public:
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
enum {
CTR_TBF_DL_ALLOCATED,
CTR_TBF_DL_FREED,
CTR_TBF_UL_ALLOCATED,
CTR_TBF_UL_FREED,
tbf: Count how often we re-use a TBF that was already being deactivated
2013-11-13 14:02:50 +00:00
CTR_TBF_REUSED,
rlc: Count the sent and resent RLC blocks
2013-11-13 18:36:57 +00:00
CTR_RLC_SENT,
CTR_RLC_RESENT,
tbf: Count how often we re-start a BSN in the send routine There appears to be a scheduling issue. Even without any NACKs we are re-transmitting a lot of frames. It might be because of this.
2013-11-23 23:01:50 +00:00
CTR_RLC_RESTARTED,
rlc: Count the window stalls on the RLC level
2013-11-19 16:09:37 +00:00
CTR_RLC_STALLED,
rlc: Count nacked frames in the statistics too
2013-11-23 00:01:19 +00:00
CTR_RLC_NACKED,
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
CTR_DECODE_ERRORS,
sba: Count SBA allocation, frees and timeouts Add a warning about the receive message poking in the internal of the sba. This will be cleaned up in a follow up commit
2013-10-27 09:00:47 +00:00
CTR_SBA_ALLOCATED,
CTR_SBA_FREED,
CTR_SBA_TIMEDOUT,
llc: Count timedout and silently dropped frames A DL tbf can be discarded and then the already queued LLCs will be silently dropped. Count this event.
2013-10-27 09:34:31 +00:00
CTR_LLC_FRAME_TIMEDOUT,
CTR_LLC_FRAME_DROPPED,
bts: Count the number of llc frames that were "scheduled" to be sent This does not mean that they have been successfully transferred to the SGSN/MS but at least that they have reached a certain point in the message flow.
2013-11-13 15:43:26 +00:00
CTR_LLC_FRAME_SCHED,
bts: Count the rach frames we receive
2013-10-27 09:50:35 +00:00
CTR_RACH_REQUESTS,
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
};
tbf: Poll MS on idle DL TBFs If an MS wants to open a new UL TBF, it can either use (P)RACH or request one in a Ack/Nack message for a DL TBF (PACCH). When a TBF becomes idle (LCC queue is empty but the TBF is kept open), there aren't any Ack/Nack requests that can be used by the MS to ask for an UL TBF, therefore it has to use the RACH. This leads to many RACH requests even for a single HTTP transaction, so it takes some time to retrieve even a simple web page. This commit modifies the scheduler to regularly send Ack/Nack requests on idle DL TBFs. It does so by extending the priority based scheduling algorithm to have 5 priority levels (highest priority first): - Control block is pending - High age (100%) threshold reached (-> request Ack/Nack) - Data is waiting or there are pending Nacks - Low age (200ms) threshold reached (-> request Ack/Nack) - Pending Nacks that have been resent already - None of the above (-> send DL dummy control block) The 'age' refers to the time since since the last control block has been sent on the TBF. This high age threshold is set to dl-tbf-idle-time or to 50% of T3190 (whichever is smaller), aiming for at least a poll (and TBF shutdown) after the TBF has expired and to safely prevent expiry of T3190. So if dl-tbf-idle-time > 200ms, there will be a poll every 200ms and a final poll after dl-tbf-idle-time. On high load, the interval between polls can get higher, but the 'high age' poll should be in place. This commit implements the scheduling with respect to GSM 44.060, 9.3.1a ("Delayed release of downlink TBF"). Ticket: #556 Sponsored-by: On-Waves ehf
2015-03-20 13:26:05 +00:00
enum {
TIMER_T3190_MSEC = 5000,
};
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
BTS();
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
~BTS();
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
static BTS* main_bts();
struct gprs_rlcmac_bts *bts_data();
sba: Create a SBAController that will manage the sbas for a BTS The PollController is a friend of the SBAController and is allowed to access the internal list. The list is hidden from everyone else. This is done because the calculation of timeout should belong into the PollController and not into the SBAController.
2013-10-19 18:47:12 +00:00
SBAController *sba();
ta: Create TimingAdvance class and make it belong to the BTS This allows us to easily flush the state in case a PCU/BTS connection is lost.
2013-10-19 18:04:57 +00:00
TimingAdvance *timing_advance();
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
bts: Move the frame_number into the BTS sructure The current_frame is an attribute of the BTS. Move it from the pcu_l1_if.cpp into the BTS. As the next step we can trigger actions depending on the change of the frame.
2013-10-17 17:59:56 +00:00
/** TODO: change the number to unsigned */
void set_current_frame_number(int frame_number);
int current_frame_number() const;
bts/pdch: Move the adding of paging to the BTS/PDCH objects Only the gprs_rlcmac_pdch will manipulate the paging list now. There can be various more refactorings of the code but they can be done later. E.g. on memory allocation failure we can continue instead of leaving the code, we should also set any_tbf only after things have been paged.
2013-10-19 16:28:59 +00:00
/** add paging to paging queue(s) */
int add_paging(uint8_t chan_needed, uint8_t *identity_lv);
bts: Return the special type for {ul,dl}_tbf_by_* functions Start returning the special type instead of the base gprs_rlcmac_tbf when retrieving a TBF. Ticket: SYS#389 Sponsored-by: On-Waves ehf
2014-07-10 15:44:06 +00:00
gprs_rlcmac_dl_tbf *dl_tbf_by_tlli(uint32_t tlli);
gprs_rlcmac_ul_tbf *ul_tbf_by_tlli(uint32_t tlli);
gprs_rlcmac_dl_tbf *dl_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts);
gprs_rlcmac_ul_tbf *ul_tbf_by_poll_fn(uint32_t fn, uint8_t trx, uint8_t ts);
gprs_rlcmac_dl_tbf *dl_tbf_by_tfi(uint8_t tfi, uint8_t trx);
gprs_rlcmac_ul_tbf *ul_tbf_by_tfi(uint8_t tfi, uint8_t trx);
bts/tbf: Move the lists into the BTS and do the look-up from the BTS The list belongs to the BTS. This makes cleaning this up more easy and establishes a hierachy of resources that start from the BTS. The debug_diagram code is now broken.
2013-10-19 19:10:38 +00:00
tbf/bts: Move the tfi_find_free into the bts
2013-10-26 17:17:58 +00:00
int tfi_find_free(enum gprs_rlcmac_tbf_direction dir, uint8_t *_trx, int8_t use_trx);
bts: Move rcv_imm_ass_cnf into the bts code
2013-10-26 17:49:16 +00:00
int rcv_imm_ass_cnf(const uint8_t *data, uint32_t fn);
bts: Move gprs_rlcmac_rcv_rach into the BTS class
2013-10-26 18:56:20 +00:00
int rcv_rach(uint8_t ra, uint32_t Fn, int16_t qta);
bts: Move rcv_imm_ass_cnf into the bts code
2013-10-26 17:49:16 +00:00
bts: Change parameter in BTS::trigger_dl_ass() to DL TBF This method is always called with a DL TBF as argument so make it clear. Ticket: SYS#389 Sponsored by: On-Waves ehf
2014-08-07 10:39:26 +00:00
void trigger_dl_ass(gprs_rlcmac_dl_tbf *tbf, gprs_rlcmac_tbf *old_tbf, const char *imsi);
tbf: Move gprs_rlcmac_poll_timeout into the tbf Move the gprs_rlcmac_poll_timeout method into the tbf class and gprs_rlcmac_downlink_assignment into the BTS.
2013-10-26 18:12:59 +00:00
void snd_dl_ass(gprs_rlcmac_tbf *tbf, uint8_t poll, const char *imsi);
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
/*
* Statistics
*/
void tbf_dl_created();
void tbf_dl_freed();
void tbf_ul_created();
void tbf_ul_freed();
tbf: Count how often we re-use a TBF that was already being deactivated
2013-11-13 14:02:50 +00:00
void tbf_reused();
rlc: Count the sent and resent RLC blocks
2013-11-13 18:36:57 +00:00
void rlc_sent();
void rlc_resent();
tbf: Count how often we re-start a BSN in the send routine There appears to be a scheduling issue. Even without any NACKs we are re-transmitting a lot of frames. It might be because of this.
2013-11-23 23:01:50 +00:00
void rlc_restarted();
rlc: Count the window stalls on the RLC level
2013-11-19 16:09:37 +00:00
void rlc_stalled();
rlc: Count nacked frames in the statistics too
2013-11-23 00:01:19 +00:00
void rlc_nacked();
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
void decode_error();
sba: Count SBA allocation, frees and timeouts Add a warning about the receive message poking in the internal of the sba. This will be cleaned up in a follow up commit
2013-10-27 09:00:47 +00:00
void sba_allocated();
void sba_freed();
void sba_timedout();
bts: Count the number of llc frames that were "scheduled" to be sent This does not mean that they have been successfully transferred to the SGSN/MS but at least that they have reached a certain point in the message flow.
2013-11-13 15:43:26 +00:00
void llc_timedout_frame();
void llc_dropped_frame();
void llc_frame_sched();
bts: Count the rach frames we receive
2013-10-27 09:50:35 +00:00
void rach_frame();
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
/*
* Below for C interface for the VTY
*/
struct rate_ctr_group *rate_counters() const;
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
private:
bts: Move the frame_number into the BTS sructure The current_frame is an attribute of the BTS. Move it from the pcu_l1_if.cpp into the BTS. As the next step we can trigger actions depending on the change of the frame.
2013-10-17 17:59:56 +00:00
int m_cur_fn;
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
struct gprs_rlcmac_bts m_bts;
bts: Introduce a PollController that has the responsibility to poll For each frame indication received by the BTS the poll controller is asked to expire timedout entries.
2013-10-17 18:12:37 +00:00
PollController m_pollController;
sba: Create a SBAController that will manage the sbas for a BTS The PollController is a friend of the SBAController and is allowed to access the internal list. The list is hidden from everyone else. This is done because the calculation of timeout should belong into the PollController and not into the SBAController.
2013-10-19 18:47:12 +00:00
SBAController m_sba;
ta: Create TimingAdvance class and make it belong to the BTS This allows us to easily flush the state in case a PCU/BTS connection is lost.
2013-10-19 18:04:57 +00:00
TimingAdvance m_ta;
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
struct rate_ctr_group *m_ratectrs;
bts, tbf: Separate functions for dl/ul tbf_by_tlli lookup In the future we want to separate ul and dl tbf into different classes that inherit from a common base. Ticket: SYS#389 Sponsored-by: On-Waves ehf
2014-07-02 15:58:14 +00:00
gprs_rlcmac_tbf *tbf_by_tlli(uint32_t tlli, enum gprs_rlcmac_tbf_direction dir);
bts: Separate functions for dl/ul tbf_by_tfi lookup Ticket: SYS#389 Sponsored-by: On-Waves ehf
2014-07-02 15:58:15 +00:00
gprs_rlcmac_tbf *tbf_by_tfi(uint8_t tfi, uint8_t trx, enum gprs_rlcmac_tbf_direction dir);
bts: Introduce a PollController that has the responsibility to poll For each frame indication received by the BTS the poll controller is asked to expire timedout entries.
2013-10-17 18:12:37 +00:00
private:
/* disable copying to avoid slicing */
BTS(const BTS&);
BTS& operator=(const BTS&);
bts: Move struct gprs_rlcmac_bts and other structs into a bts.h Begin to make the BTS a real C++ object with real responsibilities. The biggest issue will be the pcu_vty.c that might not like C++ at all.
2013-10-17 15:01:54 +00:00
};
bts: Move the frame_number into the BTS sructure The current_frame is an attribute of the BTS. Move it from the pcu_l1_if.cpp into the BTS. As the next step we can trigger actions depending on the change of the frame.
2013-10-17 17:59:56 +00:00
inline int BTS::current_frame_number() const
{
return m_cur_fn;
}
ta: Create TimingAdvance class and make it belong to the BTS This allows us to easily flush the state in case a PCU/BTS connection is lost.
2013-10-19 18:04:57 +00:00
inline TimingAdvance *BTS::timing_advance()
{
return &m_ta;
}
sba: Create a SBAController that will manage the sbas for a BTS The PollController is a friend of the SBAController and is allowed to access the internal list. The list is hidden from everyone else. This is done because the calculation of timeout should belong into the PollController and not into the SBAController.
2013-10-19 18:47:12 +00:00
inline SBAController *BTS::sba()
{
return &m_sba;
}
pdch: Remove the the bts parameter from rcv_data_block_acknowledged
2013-10-26 16:39:36 +00:00
inline BTS *gprs_rlcmac_pdch::bts() const
{
return trx->bts;
}
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
inline struct rate_ctr_group *BTS::rate_counters() const
{
return m_ratectrs;
}
sba: Count SBA allocation, frees and timeouts Add a warning about the receive message poking in the internal of the sba. This will be cleaned up in a follow up commit
2013-10-27 09:00:47 +00:00
#define CREATE_COUNT_INLINE(func_name, ctr_name) \
inline void BTS::func_name() {\
rate_ctr_inc(&m_ratectrs->ctr[ctr_name]); \
}
bts: Provide the first set of counters
2013-10-27 08:50:15 +00:00
sba: Count SBA allocation, frees and timeouts Add a warning about the receive message poking in the internal of the sba. This will be cleaned up in a follow up commit
2013-10-27 09:00:47 +00:00
CREATE_COUNT_INLINE(tbf_dl_created, CTR_TBF_DL_ALLOCATED)
CREATE_COUNT_INLINE(tbf_dl_freed, CTR_TBF_DL_FREED)
CREATE_COUNT_INLINE(tbf_ul_created, CTR_TBF_UL_ALLOCATED)
CREATE_COUNT_INLINE(tbf_ul_freed, CTR_TBF_UL_FREED)
tbf: Count how often we re-use a TBF that was already being deactivated
2013-11-13 14:02:50 +00:00
CREATE_COUNT_INLINE(tbf_reused, CTR_TBF_REUSED)
rlc: Count the sent and resent RLC blocks
2013-11-13 18:36:57 +00:00
CREATE_COUNT_INLINE(rlc_sent, CTR_RLC_SENT)
CREATE_COUNT_INLINE(rlc_resent, CTR_RLC_RESENT)
tbf: Count how often we re-start a BSN in the send routine There appears to be a scheduling issue. Even without any NACKs we are re-transmitting a lot of frames. It might be because of this.
2013-11-23 23:01:50 +00:00
CREATE_COUNT_INLINE(rlc_restarted, CTR_RLC_RESTARTED)
rlc: Count the window stalls on the RLC level
2013-11-19 16:09:37 +00:00
CREATE_COUNT_INLINE(rlc_stalled, CTR_RLC_STALLED)
rlc: Count nacked frames in the statistics too
2013-11-23 00:01:19 +00:00
CREATE_COUNT_INLINE(rlc_nacked, CTR_RLC_NACKED)
sba: Count SBA allocation, frees and timeouts Add a warning about the receive message poking in the internal of the sba. This will be cleaned up in a follow up commit
2013-10-27 09:00:47 +00:00
CREATE_COUNT_INLINE(decode_error, CTR_DECODE_ERRORS)
CREATE_COUNT_INLINE(sba_allocated, CTR_SBA_ALLOCATED)
CREATE_COUNT_INLINE(sba_freed, CTR_SBA_FREED)
CREATE_COUNT_INLINE(sba_timedout, CTR_SBA_TIMEDOUT)
bts: Count the number of llc frames that were "scheduled" to be sent This does not mean that they have been successfully transferred to the SGSN/MS but at least that they have reached a certain point in the message flow.
2013-11-13 15:43:26 +00:00
CREATE_COUNT_INLINE(llc_timedout_frame, CTR_LLC_FRAME_TIMEDOUT);
CREATE_COUNT_INLINE(llc_dropped_frame, CTR_LLC_FRAME_DROPPED);
CREATE_COUNT_INLINE(llc_frame_sched, CTR_LLC_FRAME_SCHED);
bts: Count the rach frames we receive
2013-10-27 09:50:35 +00:00
CREATE_COUNT_INLINE(rach_frame, CTR_RACH_REQUESTS);
bts: Provide the first set of counters
2013-10-27 08:50:15 +00:00
sba: Count SBA allocation, frees and timeouts Add a warning about the receive message poking in the internal of the sba. This will be cleaned up in a follow up commit
2013-10-27 09:00:47 +00:00
#undef CREATE_COUNT_INLINE
bts: Provide the first set of counters
2013-10-27 08:50:15 +00:00
pdch: Remove the the bts parameter from rcv_data_block_acknowledged
2013-10-26 16:39:36 +00:00
inline gprs_rlcmac_bts *gprs_rlcmac_pdch::bts_data() const
{
return trx->bts->bts_data();
}
pdch: Remove the trx_no/ts_no parameter and use/caclulate it on demand Simplify the depedencies and use the inline functions when we need to figure out the numbers.
2013-10-26 16:53:16 +00:00
inline uint8_t gprs_rlcmac_pdch::trx_no() const
{
return trx->trx_no;
}
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
#endif
#ifdef __cplusplus
extern "C" {
#endif
struct gprs_rlcmac_bts *bts_main_data();
bts: Start creating statistics inside the BTS code
2013-10-27 08:02:31 +00:00
struct rate_ctr_group *bts_main_data_stats();
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
#ifdef __cplusplus
}
pdch: Move enable/disable into the PDCH code When a PDCH is disabled all resources should be freed. This is currently not possible as the PDCH does not know where it belongs to. On top of that the list (and other resources) should be properly initialized on construction so that disable() is idempotent and does not check if it was disabled. During the re-factoring I noticed that during a sysmobts re-start some resources are not freed. I left a warning in the code to resolve this issue later.
2013-10-19 15:37:48 +00:00
inline bool gprs_rlcmac_pdch::is_enabled() const
{
return m_is_enabled;
}
bts: Introduce a singleton for the BTS and use it in the code Compared to the previous code there will be a branch to get the global pointer so the code will be slightly slower than the previous version but it allows us to start creating objects but still use the code from C. It is best approach I have found so far. One downside of C++ is that by default talloc will not be used (unless we override the new operator to use talloc. Right now we need to memset the C data structure by hand. The benefit of enforcing a better structure should is more important though.
2013-10-17 17:41:11 +00:00
#endif