The output and input paths are totally independent, they share no code
or structures holding state. They can be operated totally independently.
Hence, let's split the code into separate file since when someone looks
at the osmux code one is really looking specifically at Tx or Rx side,
but not at both sides.
Moreover, the "input" and "output" concepts have always been a bit
difficult to understand (would have been better calling them "src" and
"sink" instead), which adds more confusion when trying to find the
relevant part of code.
Change-Id: Ia72995092a36ca50147611e617cb88c4dcf231d5
It was not being used anywhere, yet older applications used to set it
(always to 0, which was the default value applied internally).
Let's make use of it and apply it as first seqnum to be used on a
circuit.
This value is applied upon call to osmux_xfrm_input_open_circuit(),
hence it can be set independently for every new circuit.
Change-Id: Ia26fcba5d7364a5744b2d64d0542a2b3880eee34
There's no real use in having it split, it just makes it more difficult
to extend since extra prams must be passed over 2 functions.
In fact, the duplicity of "struct osmux_in_handle" and "struct
osmux_batch" as its ->internal_data should go away in the future.
Change-Id: Ie4c6b55827ac27bcdfbe1b14fb238f5873243000
The wrap around case was not properly considered in the condition
setting the Marker bit. Let's fix it.
Related: SYS#5987
Change-Id: I6e01f29a6239f930c9be2bcb2efe447e5de8fedf
There's no real use in having a global seqnum. The seqnum, as specified
by the osmux protocol specification, relates to that of the RTP
seq+timestamp, hence it is per circuit.
Having it per circuit allows detecting gaps and lost batches on the
receiver side, applying the Marker bit on the re-assembled RTP packets.
As a resut of the fix, tests/osmux/osmux_test part validating Marker bit
started failing. It failed because it was wrongly written to start with,
since it used only one osmux_out_handle for the 4 CIDs in use, which was
wrong, since each CID must have its own osmux_out_handle as state is
kept there per circuit.
Related: SYS#5987
Change-Id: I562de6a871d8a35475c314ca107c2a12b55d6683
The value of the first RTP packet in the batch was being encoded,
instead of the last one (as documented in the Osmux protocol
specification).
Related: SYS#5987
Change-Id: I06f3d07a05181d938c22bbd0da7172b5dad6659a
The AMR FT (and hence payload content and size) may well change over the
lifetime of an RTP stream. Since all AMR payloads in an osmux batch
share the same FT (its joined payload is calculated based on
osmuxh->ctr*amr_bytes((osmuxh->amr_ft)), if a new RTP/AMR with a
different FT arrives we cannot simply append it to the current batch.
Instead, the current batch must be considered done and a new batch with
anew osmuxh header must be prepared for the resulting osmux frame to
send over UDP.
Before this patch, when a packet with a different AMR FT was submitted
for batching, it would be added in the same batch and decoding would
fail since the sizes of the batches would be wrong.
Related: SYS#5987
Change-Id: I25eb6ee54c898f575cc71ccfd6d190fe51d56dbe
If there's an error while replaying lost packets, return the error to
the caller.
If the batch is found full, early return indicating so, there's no need
to continue flow calling osmux_batch_enqueue() in osmux_batch_add()
again.
Change-Id: I62f494d2b2e7903a6683f6dea00781bb721a3d41
The osmux header goes before in the packet, so let's move the line
checking is size before the content.
Change-Id: I33feac834700d22ed06472d8971abd0567ce623b
The RTP msg will be dropped, so it makes no sense to signal the caller
to deliver the batchbeing built, since it may still have space for next
non-duplicated message.
The exception is the case where the new packet has the M marker bit set,
since the sequence numbers can be reset or jump in those scenarios.
Change-Id: Idc457bc3b26bed68796d714bc3f37a2d016ba5c3
This way the caller can log or make statistics based on the return code.
All known implementations simply check the return code to be >0, so we
are fine here.
Change-Id: I981cc7e560cd9c792a8a2a219b3612f9834296ce
This is useful for users of the API which need to keep forwarding the
msgb to lower layers which may need prepending a new header to the msgb,
like osmo-bts with l1sap.
Related: SYS#5987
Change-Id: I632654221826340423e1e97b0f8ed9a2baf6c6c3
Until now, the osmux_out_handle was allocated by the client, and passed
to the API to initialize it. This makes it really hard to improve the
implementation without breaking the ABI.
Let's break the ABI now one last time (hopefully) by allocating the
struct through an API. With only this change, the already built users
(osmo-mgw, openbsc) can still work fine, since there's no change on the
struct osmux_out_handle. However, they will somehow break next time the
struct is changed until they are ported to the same API (easy to do).
Related: OS#5987
Change-Id: Ie8df581f375c9a183a7af60b431561bda82f6e34
Those APIs where deprecated 4 years ago (end of Aug 2018), and they have
not been used since around that time. Hence it can be considered safe to
drop them, since they only make the whole code more complex to
understand.
API osmux_xfrm_output_init() is left since openbsc.git is still using
it.
Related: OS#5987
Change-Id: Icbdd364a8161a8113dbf1406716946f684d0a853
It seems different compiler versions (jenkins and my workstation) are
generating different line number for that message, and it makes osmux_test
fail when improving logging on next commit.
Change-Id: Ie2bb0ecf4cc165b9a1080e2558b33ba37014278c
Previously payload_type was always hardcoded to 98 for generated rtp
packets from incoming osmux frame.
Change-Id: I5cbeb494a8932953d9fd2dc24dacf8cd97fd84e4
This message is expected as all code filling batches call
osmux_batch_enqueue() and checks for error to know if it must tell the
user of the lib to call osmux_xfrm_input_deliver.
Change-Id: I3d8227f2281f6ca92fd2502d3e328765dc7ecfe9
In Change-Id: I2efed6d726a1b8e77e686c7a5fe1940d3f4901a7 we're adding a
new member to 'struct osmux_out_handle' which is not initialized.
Rather than initializing this single new member, let's do a memset()
over the entire osmux_out_handle at the beginnign of
osmux_xfrm_output_init().
Change-Id: I751e9414c6de2413a9f977e5ae5655ebfd114f45
Closes: OS#3219
Until this patch, we didn't notify in any way to the RTP reader when an
Osmux frame was lost. Instead, we updated the seq×tamp as if there
was no lost, and as a result the RTP reader would only see a steady
increase of delay every time an osmux frame was lost.
As the batch_factor for the lost packet is unknown, we cannot assume any
number of amr payloads lost, and thus we cannot simply increment seq and
timestamp for a specific amount. Instead, the only viable solution seems
to set the M marker bit in the first rtp packet generated after a
non-consecutive osmux frame is received.
The implementation may act differently with the first generated RTP
packet based on the first osmux seq number used for the stream. In case
0 it's used as first osmux seq number, M will be set depending on
request from original RTP packet having the M bit set. If it's not 0,
the first RTP packer will unconditionally have the M bit. That's not an
issue because it's anyway expect for receiver to sync on the first
packet.
Related: OS#3185
Change-Id: I2efed6d726a1b8e77e686c7a5fe1940d3f4901a7
With old implementation, in conditions with jitter we could end up
scheduling RTP generated packets from two consecutive osmux frames in an
interleaved way (from seq field point of view).
This new implementation should make it easier for any RTP
reader/playback to have better results in those conditions.
Old APIs osmux_xfm_output and osmux_tx_sched are marked as deprecated in
favour of the new one, which has a better control of generated RTP
packets. However, they are still usable despite the implementation changes
done to support the new API.
Related: OS#3180
Change-Id: I4e05ff141eb4041128ae77812bbcfe84ed4c02de
Previous implementation handled all types as if they were Osmux AMR
frames. For Dummy frames, we account the padding but we don't care about
the padding content. For Signalling ones, as they are not in the
specification yet, it is better avoid using unespecified fields and
return an error because it's still not known how extra data will be
handled in the input msgb.
Change-Id: I48565472b47c2a0e5db50881fbb005537af8c70d
The current code still expects to parse only AMR osmux frames, but that
will be fixed in following patches.
Change-Id: Ic2f4d1d3cc88af912bb43c8ecd90eacc6ff7190f
This patch inconditionally initializes the buffer we get to
nul-terminate it, whenever possible. It's a very simple solution to
catch three overly corner cases:
1) snprintf() returns -1, very much unlikely in practise.
2) msg->len == 0: In such case, I would expect this function is never
called with an empty message, but let's be safe in this case too.
3) If your buffer is empty, it doesn't nul-terminate the buffer.
Change-Id: I97e517f2d98e83894ea707c63489559302ff6bd2
SNPRINTF_BUFFER_SIZE() looks too complex, previous version maintains two
different variables to account for the remaining space in the buffer,
one of them is always decremented based on what snprintf() returns,
which may result in underflow. These variables are swapped - not used
consistently - all over this code.
Replace this macro by a simplified version, with one single parameter to
account for remaining space. This macro also deals with two corner
cases:
1) snprintf() fails, actually never happens in practise, but
documentation indicates it may return -1, so let's catch this case
from here to stick to specs.
2) There is not enough space in the buffer, in that case, keep
increasing offset, so we know how much would have been printed, just
like snprintf() does.
Thanks to Pau Espin for reporting, and Holger for clues on this.
I have run osmux_test and, at quick glance, it looks good.
Change-Id: I5b5d6ec57a02f57c23b1ae86dbd894bad28ea797
The buffer for osmux_test is increased as the former doesn't seem to be
able to cope with the whole output.
Change-Id: Ic838dd9d7ad89b4510ccfa58c0390c69a075b616
According to RFC4867 (RTP payload format for AMR):
"The RTP header marker bit (M) SHALL be set to 1 if the first frameblock
carried in the packet contains a speech frame which is the first in a
talkspurt. For all other packets the marker bit SHALL be set to zero (M=0)."
This information bit provides a way for the receiver to better
synchronize the delay with ther sender.
This is specially useful if AMR DTX features are supported and
enabled on the sender.
Change-Id: I0315658159429603f1d80a168718b026015060e9
This commit should fix a bug present if for instance batch_factor < 8
and osmux_batch_enqueue is called from osmux_replay_lost_packets and
enough packets were lost from last received packet.
Change-Id: I5d643810949aeca4762f0cad05eed534d35087f7
Pau Espin
Harald Welte
Pablo Neira Ayuso