The window elements are not being correctly cleaned up when clear() is called
or when overwritten by newer SN. Furthermore, the window count member is not being
correctly updated when the insertion of a newer SN overwrites the previous one.
I used the circular_map container to avoid this sort of bugs
previously we were only counting retx if we retx the start of a segment.
this could lead to unwanted behaviour, i.e., not counting retx
correctly and thus not triggering the maxretx attempt, if the receive
always sends NACKs with a SO_start.
The RLC spec is not clear on how this should be handled correctly but
IMHO using an integer number of retx is reasonable, even for segments
that might be retransmitted more often.
The alternative of using a fractional retx counter that may be increamented
proportional to the segment size that is retx is another alternative
but considered too complex to implement (and test correctly).
Main changes:
- addition of pool utilities
- The node size/alignment and batch allocation threshold are now runtime arguments
- object pool and memory pool are not anymore based on the same class.
The object pool cannot use intrusive free list because it would overwrite the object
memory
- use of inheritance to simplify testing
- removal of global network manager
- pass of custon socket manager to s1ap and gtpu ctors
- overhauled the registration of socket fd,callback in socket manager
- Implement a common event "log_rrc" for all RRC events and discriminate by procedure using an enum.
- Log events for connection, reestablishment, reconfig, reject and release.
- Log the corresponding ASN1 message used by each procedure.
- Redefine the JSON object for this event to match the new structure.
Fixed a compilation error detected by the static analyzer in gcc9.3 where bounded_vector::data() was using taking the address of the internal buffer which confused it, prefer to use the data method of std::array.
* refactor some common methods to mac_common.cc
* add common mux_base class
* move UL PDU generation to mux class
* add logical channel registration to MAC, MUX, BSR
* add initial proc BSR
* add basic MAC NR test
* rework MAC interfaces
This was done so it would work when circular buffer holds other things
that are not unique_pointers. Queue and pop_func had to be made public
to be able to call the pop_func when an SDU is discarded.
* Added ability to discard to dyn_block_queue
* Change way of keeping track of SDUs
* Check nullptr in poping callback. Starting to check for nullptr in RLC read_pdu.
* Adding RLC discard tests
* Clearing PDCP info when RLC discard happens
* Read SDUs until they are no longer nullptr
* Changed discard_if to use template argument
we've seen a heap-buffer overflow in fmt because printf wasn't using
the right formtter for size_t, which should be %zu
this patch fixes it for the PDCP LTE entity but we might have it elsewhere too
[1m[31m==7595==ERROR: AddressSanitizer: heap-buffer-overflow on address 0x629000e6f1fc at pc 0x562273a45289 bp 0x7f35567641f0 sp 0x7f35567641e0
[1m[0m[1m[34mREAD of size 4 at 0x629000e6f1fc thread T12 (STACK)[1m[0m
0 0x562273a45288 in fmt::v7::basic_format_arg<fmt::v7::basic_printf_context<std::back_insert_iterator<fmt::v7::detail::buffer<char> >, char> > fmt::v7::detail::make_arg<fmt::v7::basic_printf_context<std::back_insert_iterator<fmt::v7::detail::buffer<char> >, char>, unsigned int>(unsigned int const&) (/osmo-gsm-tester-srsue/srslte/bin/srsue+0x9dc288)
1 0x562273a3aa86 in void fmt::v7::dynamic_format_arg_store<fmt::v7::basic_printf_context<std::back_insert_iterator<fmt::v7::detail::buffer<char> >, char> >::emplace_arg<unsigned int>(unsigned int const&) (/osmo-gsm-tester-srsue/srslte/bin/srsue+0x9d1a86)
2 0x562273a308e7 in void fmt::v7::dynamic_format_arg_store<fmt::v7::basic_printf_context<std::back_insert_iterator<fmt::v7::detail::buffer<char> >, char> >::push_back<unsigned int>(unsigned int const&) /mnt/data/jenkins/workspace/srslte_ogt_trial_builder_x86-ubuntu1804-asan/srsLTE/lib/include/srslte/srslog/bundled/fmt/core.h:1548
3 0x562274361541 in void srslog::log_channel::operator()<unsigned int&, unsigned int&, unsigned long>(char const*, unsigned int&, unsigned int&, unsigned long&&) /mnt/data/jenkins/workspace/srslte_ogt_trial_builder_x86-ubuntu1804-asan/srsLTE/lib/include/srslte/srslog/log_channel.h:101
4 0x56227430d9e7 in srslte::pdcp_entity_lte::update_rx_counts_queue(unsigned int) /mnt/data/jenkins/workspace/srslte_ogt_trial_builder_x86-ubuntu1804-asan/srsLTE/lib/src/upper/pdcp_entity_lte.cc:451
the patch refactor the logging when a new PDCP SDU is enqueued for
transmission at RLC.
If the SN is already present, only a warning is logged. From the RLC
perspective operation continues and the SDU will be transmitted.
The patch also changes the order of logs. When the SN cannot be inserted
inside the queue of undelivered SDUs, only one message is logged.
since the tests are random we sometimes hit the maxRetx threshold
with 8 retx allowed that made the unit test fail.
increase threshold to lower the likelihood for this to happen.
undeliverd_sdus_queue. Added a queue to store information about
rx_counts received.
Added unit test for when the SNs wrap-around in status report genaration
this patch replaces the std::vector type used in the interface between
PDCP and RLC to signal delivery status between both layers. The new
data type is a configurable but fixed-size vector.
The RLC AM doesn't need to dynamically allocate the vector for every SN but
uses the tx_window for storage.