ffmpeg's CNG expects a fixed frame size, or at least a maximum frame
size, as it allocates internal buffers on construction based on that
frame size. Calling the decoder with a larger frame size therefore leads
to an internal buffer overflow. Calling it with a smaller frame size is
fine.
Retain the simple one-shot CNG for the most common case (frame size
smaller than the max). For larger frames, call the CNG in a loop and
manually construct a return frame.
Change-Id: I0ebe0b7a9ad8a08bbabd0d8dea264b08cb44feee
(cherry picked from commit dfcb9a8322)
(cherry picked from commit be9d19786b)
Make sure data is 64-bit aligned. Fixes alignment issues on certain
32-bit archs (armhf).
Change-Id: I66c1fc20e97c62d3d5e266e874812aed74d295ed
(cherry picked from commit 31375ca349)
(cherry picked from commit f0346ab0cd)
Already used by `--no-fallback`, so the rest just probably
inherited the same letter when doing copy-paste.
That is ambiguous and may cause option parsing conflicts.
Keep `F` only for the historical(most important) option
and set the rest to 0.
Change-Id: I6e4841f6e6614bfdfcf53d159130d6d676073d47
(cherry picked from commit b659b2371f)
(cherry picked from commit 03ac74520a)
Run the included backports script for distro-specific packaging.
Add distro suffix to the version so that all build artefacts can be
included in the release without collisions.
Include G.729. This is available in both Sid and Trixie.
Change-Id: I7efcf6a115158c5154936494c5c85628e37b21ab
(cherry picked from commit 68c6bf60a8)
(cherry picked from commit 8d24fff788)
1) Add a release job
Automatically create a GitHub release with attached Debian packages,
whenever a tag starting with 'mr' is pushed.
Generate release notes grouped by author from git commits between the
previous mr* tag and the current tag, similar to the format used in
debian/changelog.
2) Automatically generate Debian packages for PRs
This is useful for contributors as well as developers, to test and
verify proposed changes.
3) Consolidate Debian package workflows
Merge debpkg.yml into package-build.yml using a matrix strategy for
building on multiple Debian distributions (trixie, sid), using
jtdor/build-deb-action instead of custom Docker actions.
Now our release job merges artifacts from all distros.
4) Update from bookworm to trixie
Bump Debian package builds from bookworm to trixie, while at it, given
that Debian trixie/v13 is the current stable Debian release.
5) Bump workflow actions from in package build workflow
Update from v4 to current version v6, see
https://github.com/actions/checkoutCloses#2063
Change-Id: Ic1bfb3a7f9fa7e263f8e8d3b115baa01633db1a1
(cherry picked from commit e9fc1b1057)
(cherry picked from commit ac1864f865)
Using the stream object after dropping the reference is unsafe.
Change-Id: I58df474272cc7e168d6d65e5f6ace3a42fe22ef1
(cherry picked from commit 9a3f33a703)
(cherry picked from commit 4e8b3c8b9b)
Basically a revert of Ie62fa4ed
Maybe this is correct now?
Change-Id: Ia2b1b5d35c2a209430a23110f1b929e2b1ac80fb
(cherry picked from commit 8a3d8c911d)
(cherry picked from commit 4afce7e0fa)
RFC says that annex B is the default in the absence of the annexb=no
format string. Consider the alias "G729a" as annex A.
Closes#1951
Change-Id: I9a8483ee5520bf4688601123f1ebec4f1f480642
(cherry picked from commit 3df26cf2d0)
(cherry picked from commit 754837ea9f)
(cherry picked from commit 0119014d46)
Fixes regression from Ib4285e7aae
RTCP multiplexing requires the RTCP sender to maybe lock the same output
stream, maybe lock some other one. Allow for both.
Change-Id: I6fcef32e656f8f0de46ad777f11a19c259ce35c7
(cherry picked from commit cec0afa25e)
(cherry picked from commit d1d4110f0f)
(cherry picked from commit ea26245c4d)
(cherry picked from commit 9f5bc1be2e)
With selected_sfd being protected by in_lock, we pretty much have to
hold at least in_lock everywhere, and end up requiring both locks in
many places. The distinction has become pointless.
Change-Id: Ic0ad976c2d68d9639b9434da7f0e6e9c0d84c185
(cherry picked from commit e03f814855)
(cherry picked from commit 222fcaa4b0)
(cherry picked from commit abf1ad73d2)
(cherry picked from commit 209a27bbdb)
Reverses the side effect of I225b43dff
Restores consistency and fixes tests and very slow systems
Change-Id: Ib4285e7aae95cd58eccd1d3c4f84e5786b4b0810
(cherry picked from commit ffacde5748)
(cherry picked from commit b36b9abf13)
Protect selected_sfd with in_lock.
Protect RTCP sending with in_lock and out_lock as appropriate.
Has the odd side effect of RTCP reports expected in tests to be sent one
packet later than before.
Closes#1966
Probably fixes#1927
Change-Id: I225b43dff8e8fbb938d3be6aad50249997615d77
(cherry picked from commit ffc539c0d8)
(cherry picked from commit 28576a6efc)
Using `l->next` as a test for whether only one element is left in the
list is not reliable. Use the actual list length instead.
Update one affected test.
Closes#1961
Change-Id: I5773715700220cd762e61090bac941ddd33afd9b
(cherry picked from commit 02897ccb46)
(cherry picked from commit 5b97378063)
In some cases, in particular when multiple input codecs are transcoding
to the same output codec (e.g. audio + DTMF -> audio), it's possible
that one of the input decoders produces frames in a format different
from the one expected by the encoder. Add an encoder-side resampler to
compensate for that.
Change-Id: I74d55edf47ac2fa65f950e68f4a5975f1ab947bb
(cherry picked from commit 461f663e58)
(cherry picked from commit 6028d064cc)
Hopefully this doesn't break other RHEL builds
Change-Id: I50e32bdafd2fef8a4791e745ccb90d80ff091fe1
(cherry picked from commit e8af201778)
(cherry picked from commit f236cf2f06)
Sipwise Jenkins Builder
Richard Fuchs
Donat Zenichev
Orgad Shaneh
Michael Prokop