Let's use usual osmocom rate_ctr instead of having one variable +
setter/getter functions, so we can easily add new counters and also
because it makes code more clear (no need to look at what the "update"
function is doing).
Using rate counter also provides info about how recently the MS has been
interacting with the network.
Related: OS#4907
Change-Id: I744507fde4291955c1dbbb9739b18a12a80145b1
There is a big difference between:
if ((a = foo() != 0xffffffff)) { ... }
and
if ((a = foo()) != 0xffffffff) { ... }
In the first case, 'a' is the result of '!=' operation, i.e. either
0 (false) or 1 (true). In the second case, 'a' will hold the value
returned by foo(), and this is exactly what must have been used in
gprs_rlcmac_rcv_rts_block().
The bug was there since SBA allocation feature was added in 2012.
Change-Id: Ifd607ae8a33382e48f9d9e50a28a4bdf4eaf73a2
Fixes: 07e97cf8a5
Related: CID#215835
As we integrate osmo-pcu more and more with libosmocore features, it
becomes really hard to use them since libosmocore relies heavily on C
specific compilation features, which are not available in old C++
compilers (such as designated initializers for complex types in FSMs).
GprsMs is right now a quite simple object since initial design of
osmo-pcu made it optional and most of the logic was placed and stored
duplicated in TBF objects. However, that's changing as we introduce more
features, with the GprsMS class getting more weight. Hence, let's move
it now to be a C struct in order to be able to easily use libosmocore
features there, such as FSMs.
Some helper classes which GprsMs uses are also mostly move to C since
they are mostly structs with methods, so there's no point in having
duplicated APIs for C++ and C for such simple cases.
For some more complex classes, like (ul_,dl_)tbf, C API bindings are
added where needed so that GprsMs can use functionalitites from that
class. Most of those APIs can be kept afterwards and drop the C++ ones
since they provide no benefit in general.
Change-Id: I0b50e3367aaad9dcada76da97b438e452c8b230c
The resulting code is cleaner since it becomes clear the relation
between all those pointers, which are set in one function and used in
another one, passed through the caller of the former two.
Moreover, if more tbf candidates are to be added for other type of
actions, having them in a struct is much easier since only one pointer
is passed.
Change-Id: I55482aa5af7be5a176a7b4879a672ad37e618020
In previous status, if USF for GPRS-only MS was selected, then EGPRS
TBFs were skipped and either a GPRS TBF was selected or a Dummy Block
was sent. That means the behavior was unfair towards EGPRS TBFs, because
sometimes they were skipped in favor of GPRS ones.
This patch imporves the situation in the above mentioned USF scenario, by
first, under specific conditions, allowing selection of an EGPRS TBF and
then forcing it to transmit in EGPRS-GMSK (MCS1-4) so that the
USF-targeted MS can still decode the USF, while at the same time
providing more fairness by allowing the EGPRS TBF to transmit data.
The specific conditions mentioned above are, mainly, related to the fact
that once a DL data block has been sent, and hence a BSN was assigned to
it, it cannot be retransmitted later using another MCS, since lower
MCS1-4 wouldn't be able to contain higher MCS RLC payload.
The set of conditions could be expanded in the future by also selecting
the EGPRS TBF if retransmition is required and the block to be
retransmitted was originally transmitted as MCS1-4.
Related: OS#4544
Change-Id: I9af23e175435fe9ae7b0e4119ad52fcd4707b9ca
There are some restrictions to have both GPRS-only and EGPRS MS attached
to the same MS:
* Any MS needs to be able to successfully decode a DL block at least
every 18 DL blocks (360 ms). That means a Dl block with CS1-4 must be
sent at least once during that time.
* Any MS needs to be able to decode USF targeting it. GPRS-only MS can
successfully decode USF from DL blocks using GMSK: CS1-4 and MCS1-4.
In this patch, if USF of a GPRS-only MS is selected, then all DL EGPRS
TBFs are discarded from data block selection. However, this logic can be
further improved later by still allowing selection of DL EGPRS TBFs and
then forcing construction of a DL EGPRS data block using MCS1-4.
Sources:
* 3GPP TS 03.64 version 8.12.0 "6.6.4.1.1.2 Multiplexing of GPRS and EGPRS MSs"
* 3GPP TS 05.08 version 8.23.0 "10.2.2 BTS output power"
Related: OS#4544
Change-Id: Ib4991c864eda6864533363443f76ae5d999532ae
Return an interface to the window base class so that the tbf base class
can access the common window methods, such as set_ws(). It will be used
in next commit to get rid of duplicated function enable_egprs in both
dl_tbf and ul_tbf subclasses.
The user of the function can then decide to access more specific
functionaltiites of the window class by static casting it to the
specific direction (which is known by the caller since it operates on a
ul_tbf or a dl_tbf).
Change-Id: Ia2e1decf91be1184668e28297c2126affb9c7ae4
It's super annoying seeing lots of functions being called everywhere
only to find out they are only incrementing a counter. Let's drop all
those functions and increment the counter so people looking at code
doesn't see dozens of code paths evyerwhere.
Most of the commit was generated by following sh snippet:
"""
#!/bin/bash
define_pattern="^CREATE_COUNT_ADD_INLINE"
generic_func="do_rate_ctr_add"
grep -r -l "${define_pattern}" . | xargs cat | grep "${define_pattern}("| tr -d ",;" | tr "()" " " | awk '{ print $2 " " $3 }' >/tmp/hello
while read -r func_name ctr_name
do
#echo "$func_name -> $ctr_name";
files="$(grep -r -l "${func_name}(" .)"
for f in $files; do
echo "$f: $func_name -> $ctr_name";
sed -i "s#${func_name}(#${generic_func}(${ctr_name}, #g" $f
done;
done < /tmp/hello
grep -r -l "void ${generic_func}" | xargs sed -i "/void ${generic_func}(CTR/d"
grep -r -l "$define_pattern" | xargs sed -i "/$define_pattern/d"
"""
Change-Id: I966221d6f9fb9bb4f6068bf45ca2978008a0efed
It's super annoying seeing lots of functions being called everywhere
only to find out they are only incrementing a counter. Let's drop all
those functions and increment the counter so people looking at code
doesn't see dozens of code paths evyerwhere.
Most of the commit was generated by following sh snippet:
"""
#!/bin/bash
grep -r -l ^CREATE_COUNT_INLINE . | xargs cat | grep "^CREATE_COUNT_INLINE("| tr -d ",;" | tr "()" " " | awk '{ print $2 " " $3 }' >/tmp/hello
while read -r func_name ctr_name
do
#echo "$func_name -> $ctr_name"
files="$(grep -r -l "${func_name}()" .)"
for f in $files; do
echo "$f: $func_name -> $ctr_name";
sed -i "s#${func_name}()#do_rate_ctr_inc(${ctr_name})#g" $f
done;
done < /tmp/hello
grep -r -l "void do_rate_ctr_inc" | xargs sed -i "/void do_rate_ctr_inc(CTR/d"
grep -r -l "CREATE_COUNT_INLINE" | xargs sed -i "/^CREATE_COUNT_INLINE/d"
"""
Change-Id: I360e322a30edf639aefb3c0f0e4354d98c9035a3
Receive an Application Information Request from the BTS via PCU
interface. Construct a Packet Application Information message from it
(3GPP TS 44.060 11.2.47) and send it to all MS with active TBF.
The TTCN-3 test infrastructure to test this feature is not quite ready
yet, so I've added C unit tests instead.
Related: OS#4048
Change-Id: Ie35959f833f46bde5f2126314b6f96763f863b36
The PDCH class and corresponding functions are rather self-contained and
independent from BTS implementation. Let's move them into separate file
to make bts.cpp more manageable. As additional benefit it allow us to
somewhat untangle all the different cross-dependent includes.
Change-Id: Ie05e25361e6741a81b024679f9675c98d4923683
Related: OS#1539
Many files include unnecessary headers and don't include headers which
are actually used. Because of that combined with the fact that OsmoPCU
is a mixture of C and C++, it makes it hard to modularize code. Fix
this (using iwyu [1] tool):
* add missing headers
* remove unused headers
[1] https://include-what-you-use.org/
Related: OS#1539
Change-Id: I8c9f488a43b099c72b2d30d3245e7ba50872fc00
* add functions/macros for setting TBF's poll state
* add function for checking TBF's poll state
Change-Id: I6db1c4e7bd0a49aeb5e391afe371c36b96c6a702
Related: OS#1539
* add functions/macros for setting TBF's UL ack state
* add functions for checking TBF's UL ack state
N. B: this should not be confused with TBF-UL state.
Change-Id: I144483447d4b0b93e775da0e926ee45eb8ab39f3
Related: OS#1539
* add functions/macros for setting TBF's UL/DL state
* add functions for checking TBF's UL/DL state
* move pre-free check into separate function
N. B: this should not be confused with TBF-UL or TBF-DL state.
Change-Id: Idcbf5775d17b1247f2ed01788f9b0788ce66e871
Related: OS#1539
We do not use this style (#warning as an issue tracker replacement) in
any other Osmocom project. Also those warnings clutter compiler output
making it harder to spot warnings for the actual code.
Change-Id: I72070e2a027e60e8b80c12ccfa23ff075434689f
When looking at GSMTAP output so far, one is easily overwhelmed by way
too much information being presented. A lot of is consists of DUMMY
frames, which are probably of lowest interest, ever.
A concept similar to the "gsmtap-sapi" of OsmoBTS is introduced, by
which the user can configure which particular categories (uplink or
downlink control or data, gprs or egprs, ...) he actually wants to
see in his logs.
Change-Id: I297183690e98a7234dfc1608c18847d8981306e4
Commit b78a4a6dfe tried to fix a NULL dereference
error, but apparently was overly eager to return, because it looked like all
code paths would dereference the tbf.
In fact the code path further above, for msg != NULL, has "always" dereferenced
the tbf, but the lower code path, the one effecting the paging, has only
started to dereference tbf since shortly before the overly eager fix: in
da7250ad2c, to "update the dl ctrl msg counter
for ms". It seems that this tbf dereference in the paging path is bogus and the
cause for the segfault that made me write the early exit fix.
Fix that fix:
Do not exit early if tbf == NULL, stay in there to be able to reach the paging
path below.
In case of a message to be sent, assume that tbf is present, and verify: print
an error message and abort if there is a msg but no tbf, so that we will see
the error if I'm wrong there. If a tbf is missing, free the msg.
In case of no message, go on to send pending pagings, but do not attempt to
count ctrl messages for a tbf -- IIUC there will never be a tbf if we're
paging.
This should avoid segfaults while keeping PACCH paging intact.
Tweak a comment for and add a blank line above the paging section.
Related: OS#2176 CID#158969
Change-Id: Ib79f4a945e211a13ac7d1e511cc37b0940ac6202
When Packet resource request is received, PCU will generate the
packet access reject if no resources are present. The encoding is done
based on section 7.1.3.2.1 and 8.1.2.5 of 44.060 version 7.27.0 Release 7.
This patch also includes the test case to validate the generated
packet access reject message.
This patch is integration tested on Osmo-trx setup with Ettus B210 board
and LG F70 MS with some simulation code changes in Osmo-pcu.
Change-Id: I05ff25124b58905586caa0c0c37023d69724f121
Adds spb counters at BTS level(show bts statistics).
Adds RLC/MAC downlink control msg at ms level(show ms imsi <imsi_val>).
Adds the number of coding schemes counter for UL at TBF level.
Change-Id: Icbe4ba95e34bea89ee36f532d099db68204b7c38
When PDAN/EPDAN with channel description is received, PCU will generate the
packet access reject if no resources are present. The encoding is done
based on section 7.1.3.2.1 and 8.1.2.5 of 44.060 version 7.27.0 Release 7.
This patch also includes the test case to validate the generated
packet access reject message.
This patch is integration tested on Osmo-trx setup with Ettus B210 board
and LG F70 MS with some simulation code changes in Osmo-pcu.
Change-Id: I096a3bb44a65533b9e9b091925dd5f70a8696d6
ARFCN is already part of TRX struct so there's no need to supply it
explicitly in a separate parameter. I've tested and those are the same
anyway.
Change-Id: I8e975c52cbc819427880093b1e5371fe1f8ce460
rlc_sent_dummy count the amount of dummy package which are
sent in case no data packet is in the queue.
Change-Id: Ia60eab853d9145980f30d63e4ce4b520b8c51381
Reviewed-on: https://gerrit.osmocom.org/85
Tested-by: Jenkins Builder
Reviewed-by: Holger Freyther <holger@freyther.de>
The counter rlc_sent has nothing to do with the TBF.
The RLC packet got sent in the gprs_rlcmac_sched().
Change-Id: I5d2b910ea7cc250f17530406eda3be9b29b051fd
Reviewed-on: https://gerrit.osmocom.org/84
Tested-by: Jenkins Builder
Reviewed-by: Holger Freyther <holger@freyther.de>
Currently the selection of a pending control message is done round
robin. It can possibly happen, that a DL assigment is sent on a DL
TBF while an UL assignment is pending. The MS will replace the old
DL TBF in that case, so that it can no longer be used for the PACCH
so that the UL assignment is lost.
Give assigment requests for the same direction a lower priority.
Sponsored-by: On-Waves ehf
Currently there are some places where tbf->control_ts != ts is
evaluated to check, whether ts is a control slot.
Replace these expressions by tbf->is_control_ts(ts) which does
the same whitout exposing internal fields.
Sponsored-by: On-Waves ehf
Currently the checks in that function are based on the different
internal TS values of a TBF. It is assumed that they match the TS
that the current RTS refers to.
This commit adds a TS parameter to create_ul_ass, create_dl_ass,
and create_ul_ack to make this more explicit.
Sponsored-by: On-Waves ehf
LListHead does basically the same like llist_pods, but more C++ish
and with type safety.
This commit turns the former list field of gprs_rlcmac_tbf into a
private field, provides accessors, moves the related code from
pcu_vty.c to pcu_vty_functions.cpp, and removes the llist_pods
type and related code.
Sponsored-by: On-Waves ehf
Currently the msg data is accessed and index 0 assuming the msg is
not empty. While this should be always the case, the msg can be
empty if there are software errors in the message creation functions.
This commit adds an assertion to catch this kind of bugs.
Sponsored-by: On-Waves ehf
Currently the control block scheduler does not seem to be fair in all
cases. At least it happend while testing the EGPRS code, that a
Uplink Ack/Nack message got never be sent, because a Downlink
assignment took over all the times.
This commit changes the round robin code to always increment the
priority offset by 1 instead of (i + 1). The former definitely
ensures that every message type gets the highest priority after some
steps. The latter might be more fair in some situations, but that and
its correctness are not proven.
Sponsored-by: On-Waves ehf
The leak rate sent to the SGSN does not reflect the current CS level,
lost frames, and control message overhead. So the SGSN cannot do
proper queue control under non-optimal conditions.
This commit computes the leak rate for the last flow control interval
by computing the maximum theoretical leak rate and basically
substracting control blocks, nacked blocks, and reduced block sizes
due to CS downgrade. By using this approach, the value will by more
stable on low load, where the value will tend to be near the value
derived from the configuration. On full load the transmitted value is
completely derived from the measurements.
Note that the MS default values are no adapted to the adapted BVC
leak rate, since a single MS which has a lower link quality would
otherwise be reducing the rate of another MS with good radio
conditions, which would not make much sense if they did not share any
PDCH.
Sponsored-by: On-Waves ehf
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
Currently the DL blocks are scheduled round robin to each TBF that is
either in state FLOW or FINISHED and not waiting for an IMM.ASS
confirmation. This way, if single blocks has been NACK'ed by the MS
and the PCU has already resent the missing packets, the PCU starts
retransmitting them until it has received an ACK/NACK even if other
TBF have RLC blocks that need to be transmitted.
This commit changes sched_select_downlink to select the next TBF with
the highest priority, where blocks that are going to be resent again
have a lower priority unless the window is stalling. If there is only
one TBF the old behaviour is kept, since there is no other TBF that
can have a higher priority.
If there is much packet loss on a single phone, this modification can
lead to a higher latency for that MS.
Sponsored-by: On-Waves ehf
Pau Espin
Vadim Yanitskiy
Pau Espin Pedrol
Oliver Smith
Harald Welte
Neels Hofmeyr