This way PCU can now detect whether scheduled UL blocks through USF
were never received. This allows in a follow-up patch to start
increasing N3101 properly.
Related: OS#5033
Change-Id: Ia99c9edad6e5bd837e9baeb4fb2683b227887957
When the scheduler detects it's time to receive a UL block due to a
scheduled poll, if that polling is done on a UL TBF, then use its USF if
available instead of using USF_UNUSED (=7) when sending a DL block on
that same FN.
This is not really needed for correct work, since MS take care
themselves of scheduling a UL block when they receive the poll (RRBP)
some time before, and don't check the USF at the time of transmitting.
In any case, it helps understand better when looking at pcap traces that
indeed it a UL block from that MS was requested, instead of setting USF
to 7.
Related: OS#5033
Change-Id: I2ad9d8ea6afc8f83192033470bd27010a7474430
Simply use the UL TBF pointer all along until the end, instead of setting
both the UL TBF pointer plus the usf var.
This commit is also a preparation for next commit which also selects UL
TBF when a poll is available, to set its USF in the DL message instead
of "USF_UNUSED".
Change-Id: I3aa3886932ef87db18ed7ff6991ea315f481990b
With previous code, a skipped TBF could be returned despite not matching
the conditions, since at the end of the loop the tbf pointer was
returned.
Related: OS#5020
Change-Id: If6dccec86c7a655bf1c62f333cfbc8d2c507c94f
pdch_ul_controller.c: In function ‘pdch_ulc_release_tbf’:
pdch_ul_controller.c:217:7: error: ‘item_tbf’ may be used uninitialized in this function [-Werror=maybe-uninitialized]
217 | if (item_tbf != tbf)
| ^
Change-Id: I42120fdf23753945ebc16bb5469d9fd253c3da37
There's no need for setting the FN in RA.ind since we anyway already
receive a DATA.ind beforehand.
Furthermore, the applied delay of 5 in the call is not really used at
all.
Change-Id: I437f4f95d054aea96bec3b9343e495451020ff3c
Take the time to also do small refactorings to clarify and simplify the
function, by using rts_next_fn() already available in pcu_utils.h and
getting rid of poll_tbf from tbf_candidates, which clearly follows
another objective.
Using PDCH UL Controller has the advantage that we don't need to check
poll_scheduled() on each TBF, but only do the query once.
Related: OS#5020
Change-Id: Ia60bb5249a9837dec1f42180e44d9848334d86d6
TbfTest is updated to submit empty blocks to have somehow meaningful
output (at least as meaningful test results as before, not much). That's
because we must update bts->curr_fn to have polls expire.
Related: OS#5020
Change-Id: I683ca738ce5a133c49c36a1d94439a942d64a831
Right now we handle different types of UL allocations in different
classes like PollAllocator and SBAllocator, and they usually don't take
into account the other one in most cases. Furthermore, those objects are
usually per-BTS object, instead of per PDCH object.
This is a first step towards having a unified per-PDCH controller which
takes care of controlling what is scheduled and hence expected on the
uplink. Each PDCH has a UL Controller which keeps track of all reserved
uplink frame, be it SB, RRBP poll or USF assigned, all under the same
API.
As a first step, only the SBA part is fully implemented and used (being
it the easiest part to replace); TBF poll+usf will come in follow-up
patches later on. As a result, the SBAllocator per-BTS class dissappears
but some of its code is refactored/reused to provide more features to the
gprs_rlcmac_sba object, which is also further integrated into the new UL
Controller.
Related: OS#5020
Change-Id: I84b24beea4a1aa2c1528f41435f77bd16df2b947
Since recently (see Depends below), BTS side submits DATA.ind with len=0
to announce nothing was received on that UL block FN. This will allow
osmo-pcu track time more accurately, and use this information to quickly
find out if a UL block was expected as requested by RRBP or USF poll and
increment counters such as N3101 (finally being able to properly
implement timers such as T3619).
Depends: osmo-bts.git Change-Id I343c7a721dab72411edbca816c8864926bc329fb
Related: OS#5020
Change-Id: Ibc495173119465e74f726ddc36e312334e6dc0fd
Since recently (see Depends below), BTS side submits DATA.ind with len=0
to announce nothing was received on that UL block FN. This will allow
osmo-pcu track time more accurately, and use this information to quickly
find out if a UL block was expected as requested by RRBP or USF poll and
increment counters such as N3101 (finally being able to properly
implement timers such as T3619).
This patch does the same for direct phy feature, where the osmo-pcu
process receives the DATA.ind directly from the DSP.
Depends: osmo-bts.git Change-Id I343c7a721dab72411edbca816c8864926bc329fb
Related: OS#5033
Change-Id: I9a835e16ef0e5a68c003a93d1a33233aa43464ae
Since recently (see Depends below), BTS side submits DATA.ind with len=0
to announce nothing was received on that UL block FN. This will allow
osmo-pcu track time more accurately, and use this information to quickly
find out if a UL block was expected as requested by RRBP or USF poll and
increment counters such as N3101 (finally being able to properly
implement timers such as T3619).
Depends: osmo-bts.git Change-Id I343c7a721dab72411edbca816c8864926bc329fb
Related: OS#5020
Change-Id: I17c28abf63b153448b533971ac5cf2e48daadea8
Use primitives instead of parsing the message a second time.
Set bctx->is_sgsn to false to allow the BSSGP layer to send back a
RESET_ACK with cell information.
Related: OS#3879
Depends: Ibcbaffa94cbdc4296a8a7c372304ac11d50d9559 (libosmocore)
Change-Id: I3afaf826798e362270ffa622c24bfd124ef25cd1
If Idle TBF timer (X2031) is set to 0, it means the TBF release is
immediately started once all queued data has been scheduled. In that
case, we must set FBI=1 (by setting cv=0) and move to FINISH state.
This used to work over the usual path where X2031 != 0, because release
start will alays happen at a later sched poll time where a dummy LLC
frame is sent and FBI set accordingly.
Change-Id: Ib20602936ae084c413f6bfe14eea33b602020be0
Old commit getting rid of LLC UI dummy and updating create_new_bsn()
function introduced a bug by not moving update of value m_last_dl_drained_fn
prior to a new break introduced.
As a result, the value is not updated in the case LLC queue becomes
drained but last few bytes are drained at exactly that moment.
Furthermore, then the IDLE tbf timer (X2031, keep_open())) returns always
true since according to it the drain never happened.
The impact of the bug is basically delaying a bit more than expected the
time the TBF stays in IDLE state with the TBF release process yet
to be started.
Related: OS#4849
Fixes: 7d0f9a0ec3
Change-Id: I7420aeffda3500bcdc990291e4a56511af433ff9
This is specially useful to detect for instance if a cell is handling
too many users, ending up in TFI or USF exhaustions. This information
can be later in the future used to tune TBF allocation algorithm behavior
(either manually/statially through config file, or
automatically/dynamically in code based on some thresholds).
Related: OS#5042
Change-Id: I5402e937ff8d800684655e500ef8e5c867141dc3
Since a while ago, the data architecture was changed so that TBF is
guaranteed to always have a MS object associated. Hence, it makes no
sense to pass the MS object as a separate param as we can take it from
tbf object and makes code less confusing.
Change-Id: Idc0c76cf6f007afa4236480cdad0d8e99dabec5f
Before this patch, allocate_usf() was implemented to only allocate 1 USF
per TBF, regardless of the available ul_slot mask.
As a result, only 1 slot at max was allocated to any TBF. That's a pity
because usual multislot classes like 12 support up to 2 UL slots per TBF
(in common TS with DL).
This patch reworks allocate_usf() to allocate as many UL multislots as
possible (given mslot class, current USF availability, TFI availability,
related DL TBF slots for the same MS, etc.).
As a result, it can be seen that AllocTest results change substantially
and maximum concurrent TBF allocation drops under some conditions.
That happens due to more USFs being reserved (because each TBF has now
more UL slots reserved). Hence now USF exhaustion becomes the usual
limitation factor as per the number of concurrent TBFs than can be
handled per TRX (as opposed to TFIs previously).
Some of the biggest limitations in test appear though because really
high end multislot classes are used, which can consume high volumes of
UL slots (USFs), and which are probably not the most extended devices in
the field.
Moreover, in general the curren timeslot allocator for a given
multislot class will in general try to optimize the DL side gathering
most of the possible timeslots there. That means, for instance on ms
class 12 (4 Tx, 4Rx, 5 Sum), 4 DL slots and 1 UL slot will still be
selected. But in the case where only 3 PDCHs are available, then with
this new multi-slot UL support a TBF will reserve 3 DL slots and 2 UL
slots, while before this patch it would only taken 1 UL slot instead of
2.
This USF exhaustion situation can be improved in the future by
parametrizing (VTY command?) the maximum amount of UL slots that a TBF
can reserve, making for instance a default value of 2, meaning usual
classes can gather up 2 UL timelosts at a time while forbidding high-end
hungry classes to gather up to 8 UL timeslots.
Another approach would be to dynamically limit the amount of allowed
reservable UL timeslots based on current USF reservation load.
Related: OS#2282
Change-Id: Id97cc6e3b769511b591b1694549e0dac55227c43
The old VTY command is marked as deprecated and still overrides the use
in case it's used.
Related: SYS#5358
Depends: libosmocore.git Change-Id I74fb0a3afc1ac4aadbfc609b882d929401f790eb
Depends: osmo-bsc.git Change-Id I8b97ea11bad5fe05f2f634945b5703ee9abde81d
Change-Id: I46f2a955b157a409055fca7fb917dc4f75482426
The gprs_ns2 now requires to specify every bind which should be used by the NSE for IP-SNS
Related: SYS#5354
Depends: I9ab8092bf286e7d90e92f5702a5404425e959c84 (libosmocore)
Change-Id: I35c987224ce098f7ee9f189ce0fce9e68ad3feac
The DL NACC related message (PKT Cell Neighbor Data/Change Continue)
are filled with the TFI of the target TBF. Hence, only select the tbf
for NACC transmission if the related TBF already has a TFI assigned.
Otherwise, "OSMO_ASSERT(tbf_is_tfi_assigned(tbf));" in nacc_fsm.c when
generating messages may be hit.
Related: SYS#4909
Change-Id: I72b2dff28aacdb04909c098c94834ff79f55b31d
Add support for multiple SNS endpoints.
Move the NSE allocation to the top in preparation of IP-SNS binds.
The future gprs_ns2 library will require to manual add every bind to the NSE for IP-SNS.
Rename the function to match more it's purpose.
Related: SYS#5354
Change-Id: I69cf48ab168a6dca4f649157bf6556d7cd27d4fb
Similar to what's done in the previous commit, but this time when we are
further forward in the resolution process.
This can be triggered for instance because we are taking too much time
to resolve and MS has timer to retransmit the Pkt cell Change Notify in
case no response was received in time.
This commit fixes osmo-pcu exiting due to ASSERT(0) since the event was
already accepted but not being handled in the state function.
Related: SYS#4909
Change-Id: I0c29e5979fec6eebe9dfb151907a4cd2f5e4a737
If the message is a duplicate (same tgt cell), simply ignore it.
If the message contains a different tgt cell, restart the resolution:
* Avoid re-creating the socket in that case
* Avoid potentially picking a CTRL response for an older request
Related: SYS#4909
Change-Id: Ia2ed2580bbbdd6d3464833257b0dcb8ec6f8d699
We don't care about those messages anymore if we already transitioned
further than NACC_ST_WAIT_REQUEST_SI. Furthermore, RIM code dispatching
the event to the FSM is only doing it in the mentioned state above.
Related: SYS#4909
Change-Id: If420b49e437ff02073669522408763e5e84fe477
Same filler will be needed in different places since that message can
arrive at different points of time (different states).
It also helps supporting newer key types in the future.
Change-Id: Idfd4db8408f767b1847b04c88047a1c4996e543e
Now that we have the required System Information in osmo-pcu to craft
the Packet Neigbour Cell Change packet for cells it controls, let's
avoid starting a RIM procedure to gather the SI info, since the SGSN
would end up routing the RIM request back at us and we'd answer back to
ourselves.
This same optimization cannot be done on the first step (CTRL Neighbor
Resolution against BSC), because the PCU cannot know if the target
ARFCN+BSIC requested by the MS is actually managed by a cell under the
PCU or it's another cell managed by another external PCU, because
ARFCN+BSIC keys can be resued among non-neighbor cells. Hence, it shall
always ask the BSC since only it holds the information about neighboring
cells.
Related: OS#4909
Change-Id: I928875b6b66dff55fc12f51b6b1ad919fef7d03b
Use the fact that the MS must answer the RRBP of the Pkt Cell Change
Continue with a CTRL ACK to find out whether the message was received
successfuly or a retransmission is potentially required.
3GPP TS 44.060:
"""
When the mobile station receives the PACKET CELL CHANGE ORDER or
the PACKET CELL CHANGE CONTINUE message the mobile station shall
transmit a PACKET CONTROL ACKNOWLEDGMENT message in the specified
uplink radio block if a valid RRBP field is received as part of the
message; the mobile station may then switch to a new cell.
"""
Related: SYS#4909
Change-Id: I7cc28922e71699598da0ef6eb90136a47d3c002f
Answer an incoming RAN INFORMATION REQUEST RIM PDU with RAN INFORMATION
PDU that contains system information type 1, 3 and 13
Depends: osmo-bts I5138ab183793e7eee4dc494318d984e9f1f56932
Change-Id: Id72118120c14984d2fb1b918b41fac4868150d41
Related: SYS#5103
Those fields were ment to be 0 (non-valid), just the comments were
copied over from somewhere else, and they are misleading.
Change-Id: Ic95853e115f60c65f7f11187d49d6e870d08c7bb
If fore some reason we fail to fetch SI of target cell, we move
directly to NACC_ST_TX_CELL_CHG_CONTINUE in order to submit a Cell
Change Continue against the MS without providing any Packet Neighbor
Cell Data beforehand, as per spec that's probably the best we can do
in this scenario (TS 44.060):
"""
1) The network responds with a PACKET CELL CHANGE CONTINUE message.
If a mobile station as response to a PACKET CELL CHANGE NOTIFICATION
message receives a PACKET CELL CHANGE CONTINUE message without receiving
any neighbour cell system information, the mobile station shall stop timer
T3208, stop timer T3210 if still running, leave CCN mode and continue cell
reselection in NC0/NC1 mode.
"""
This commit also fixes a use-after-free triggered by TTCN3 test
TC_nacc_outbound_rac_ci-resolve_fail_parse_response, where the "cmd"
pointer passed to nacc_fsm_ctrl_reply_cb() was freed during FSM
termination (its talloc ctx was under ctx->neigh_ctrl_conn) and the
libosmocore code calling that callback was later on accessing
cmd->defer.
Since due to this change the FSM is no longer syncrhonously freed, the
issue is gone.
Related: SYS#4909
Change-Id: Ie3f12a08ad611b1086d3f4ab7c3d34af43c07961
Upon timeout, we move directly to NACC_ST_TX_CELL_CHG_CONTINUE in order
to submit a Cell Change Continue against the MS without providing any
Packet Neighbor Cell Data beforehand, as per spec that's probably the
best we can do in this scenario (TS 44.060):
"""
1) The network responds with a PACKET CELL CHANGE CONTINUE message.
If a mobile station as response to a PACKET CELL CHANGE NOTIFICATION
message receives a PACKET CELL CHANGE CONTINUE message without receiving
any neighbour cell system information, the mobile station shall stop timer
T3208, stop timer T3210 if still running, leave CCN mode and continue cell
reselection in NC0/NC1 mode.
"""
Related: SYS#4909
Change-Id: Ia9932ab082ec095294e85dc4d532046970e17986
This way, we don't open a socket and do the IPA handshake in the event
the request is already cached.
Related: SYS#4909
Change-Id: Ib1ea85e1196c8b9dc40c8837ab5d4a54f2a1f2d4
Older versions of osmo_ctrl_conn_alloc() may not properly initialize
write_queue.bfd.fd to -1, which means if osmo_sock_init2_ofd() failed
during nacc_fsm_alloc(), the destructor would wrongly enter the conditon
where the whole structure is set and unregister the unregistered fd.
Related: libosmocore Change-Id I98f744d2880fbb883719cdf1d3eb31f2b22a13b6
Related: SYS#4909
Change-Id: I253bd9087b1f7ab039aa1127e9dc586f5106905a
A new nacc_fsm is introduced per MS object, with its partner priv
structure struct nacc_fsm_ctx, which exists and is available in the MS
object only during the duration of the NACC procedure.
The NACC context is created on an MS whenever a Pkt Cell Change
Notification is received on Uplink RLCMAC, which asks for neighbor
information of a given ARFCN+BSIC.
First, the target ARFCN+BSIC needs to be translated into a CGI-PS
(RAC+CI) address. That's done by asking the BSC through the Neighbour
Resolution Service available in osmo-bsc using the CTRL interface.
Once the CGI-PS of the target cell is known, PCU starts a RIM RAN-INFO
request against the SGSN (which will route the request as needed), and
wait for a response containing the SI bits from the target cell.
After the SI are received, the scheduler is instructed to eventually
poll a TBF for the MS originating the CCN, so that we can send the SI
encapsulated into multiple Packet Neighbor Cell Data messages on the
downlink.
One all the SI bits are sent, the scheduler is instructed to send a
Packet Cell Change Continue message.
Once the message above has been sent, the FSM autodestroys itself.
Caches are also introduced in this patch which allows for re-using
recently known translations ARFCN+BSIC -> CGI-PS and CGI-PS -> SI_INFO
respectively.
Change-Id: Id35f40d05f3e081f32fddbf1fa34cb338db452ca
This also changes the vty configuration. If only timeout has been
configured for ns the new configuration is compatible.
For further information see:
https://osmocom.org/projects/libosmocore/wiki/Network_service_(NS)
Depends-on: I8c3f2afecc74b78f7f914f7dce166cbcb63444eb (libosmocore)
Change-Id: I14af821a8d1fda670643c3d5f81299a3abf3c583
All gprs_ns2 enums have now GPRS_NS2 as prefix.
Depends-on: I548ff12f7277cbb7e1a630a3dc02b738ce89be72 (libosmocore)
Change-Id: Ifdc7956318c07d680feab33c22bc2c6f20927bf9
Sine we don't yet implement properly all resegmentation of blocks from
same MCS family type, when requiring a GMSK DL block (due to GPRS+EGPRS
multiplexing limitations) we need to skip retransmitions, otherwise we'd
be incorrectly picking a DL block which was already built with a
potentially higher MCS value.
The "DL_PRIO_NEW_DATA" prio serves two purposes:
* There's new data to send
* There's some nacked data to be retransmitted
The 2nd purpose has, later on, more priority over the 1st one when the tbf
is selected (see gprs_rlcmac_dl_tbf::take_next_bsn()).
Until now we were handling correctly the case where the tbf was skipped
in case the prio was to resend unacked data (DL_PRIO_SENT_DATA), but
was incorrectly selected when it'd send nacked data. Let's fix it by
specifically checking w->resend_needed() < 0.
Change-Id: I253de8e1a190a9adb56160f38892c9e43e2c0272
Under some circumstances, it could happen that a DL TBF is created as a
GPRS TBF due to not yet having enough information of the MS, and only
after the TBF is created the PCU gains that information and upgrades the
MS mode to "EGPRS". Hence, there's the possibility to run into a
situation where a GPRS TBF is attached to a EGPRS MS.
It may also happen sometimes that despite the TBF and the MS be EGPRS,
there's need to further limit the DL MCS to use, eg. MCS1-4 (GMSK).
As a result, when asking for the current DL (M)CS to use, we must tell
the MS which kind of limitations we want to apply. The later reasoning
was already implemented when GPRS+EGPRS multiplexing was added, but the
former was not being checked. Hence, by further spreading through the
call stack the "req_kind_mode" we match both cases.
Related: OS#4973
Change-Id: Ic0276ce045660713129f0c72f1158a3321c5977f
The TBF can sometimes be detached from an MS, for eg. when switching
from one MS object to another due to them being merged after we found
duplicate objects upon receiving new information from it, but that
change is instantaneous so it shouldn't be a problem. The only other way
where an MS can be detached from an MS is during the end of its (or the
MS) life, where it is not sending data anymore.
Hence, it is safe to drop those checks for MS not being null. Those
being trigger, it should be considered a bug.
Change-Id: If292a53a09a64664031e756bff4735b9c6ee8651
It could happen as of current implementation that a TBF was created as
GPRS due to the MS being non-egprs, and later on the MS was upgraded to
EGPRS due to newly received information from the MS.
Hence, in order to infer if the data block is EGPRS or GPRS, let's
better check for the TBF info, which is the one really mandating the
kind of dl block to generate.
Change-Id: I49720fb3a69ca972cd1973de937ac8ee77615431
MS is always assigned to a BTS, since it's set during MS constructor.
Hence, the check removed in this patch would never hold true (and if it
did, it'd be a bug).
Change-Id: I86a71c64623f7bec031226938a54306148370ffb
it was noticed that gprs_pcu_get_bts_by_cgi_ps() sometimes failed to
return the BTS even if the CGI-PS fields matched, probably due to memcmp
checking too padding bytes which may not be zero-initialized in one of
the two memory regions being checked. Let's be on the safe side and use
libosmocore APIs to check them.
Depends: libosmocore.git Change-Id I00e329bc5be8674b30267dec238e7656ddfc21db
Change-Id: I7c8ee2c447634e45b367bb8f84adf0140ae48591
This code doesn't do anything yet app-related with the received RIM
messages, but already provides the initial infrastructure to handle them
in the future, and does first checkings.
Related: SYS#5103
Change-Id: Ia0ade0e97ea781ec655439c008b6cefaf3e90dec
This patch doesn't really tests whether osmo-pcu can work on a multi-bts
environment, but it prepares the data structures to be able to do so at
any later point in time.
Change-Id: I6b10913f46c19d438c4e250a436a7446694b725a
There's no real point in using C++ there, and using C++ makes the
compiler fail to use llist_head in multi-bts patches added later due to:
"""
'offsetof' within non-standard-layout type is conditionally-supported
"""
Change-Id: I8965b5cc5a713e64788b5b6aa183d3035341ddbb
Pau Espin
Alexander Couzens
Harald Welte
Vadim Yanitskiy
Philipp Maier