Upstream changes (NEWS):
* Require GSSDP >= 1.6.5 and reuse its allocated TCP socket for the web
server.
* Do not leak a GError in the ACL handler.
* Fix IPv6 host-header validation.
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
(cherry picked from commit 575122a4a0)
Signed-off-by: Ankur Tyagi <ankur.tyagi85@gmail.com>
Signed-off-by: Anuj Mittal <anuj.mittal@oss.qualcomm.com>
Upstream changes (git log v1.5.4..v1.5.6):
* Group/bonding fixes: check value length in group config storage, fix
deriving string options by a group, fix getting SRTO_RCVBUF/SRTO_SNDBUF
and rejecting options not allowed on a group, and fix a wrong
'connection lost' error when sending to a connection-pending group.
* Fix a stalled connection that should break after rogue NAK/ACK, and a
misleading listening-socket error message.
* Build: Windows-on-Arm64 support, fix Windows installers, fix use of
the OPENSSL_USE_STATIC_LIBS CMake option, deprecation warning for
Windows+PThreads.
Drop 0002-allow-build-with-cmake-4.patch: it backported the
cmake_minimum_required(VERSION 3.5) bump, which is already present in
1.5.6 upstream.
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
(cherry picked from commit 70dfb6465f)
Signed-off-by: Ankur Tyagi <ankur.tyagi85@gmail.com>
Signed-off-by: Anuj Mittal <anuj.mittal@oss.qualcomm.com>
Upstream changes (release notes):
* 1.0.19: security and edge-case correctness fixes - heap-buffer-overflow
read in decode_slice_unit_tiles() from unvalidated PPS tile geometry
(CVE-2026-45382) and heap OOB read in decode_slice_unit_WPP() via an
out-of-bounds CtbAddrRStoTS access (CVE-2026-45383); dec265 SDL fixes
for 4:4:4 streams and mid-stream resolution changes. ABI compatible
with 1.0.18.
Not upgraded to 1.1.x: 1.1.0 reworked the x86 SIMD path to dispatch at
runtime via __builtin_cpu_supports(), which reads the compiler-rt global
__cpu_model. Under this distro's clang/lld toolchain, linking the shared
library then fails with
R_X86_64_PC32 cannot be used against symbol '__cpu_model'; recompile with -fPIC
because clang emits a direct PC-relative access to that exported,
preemptible symbol. Building the objects -fPIC, -Bsymbolic,
--exclude-libs, -fno-direct-access-external-data and a version script
localizing __cpu_model were all tried without success (lld validates the
relocation before applying the localization). 1.0.19 stays on the
compile-time SSE path and builds cleanly, so it is the latest buildable
release here.
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
(cherry picked from commit 99ca84156d)
Signed-off-by: Ankur Tyagi <ankur.tyagi85@gmail.com>
Signed-off-by: Anuj Mittal <anuj.mittal@oss.qualcomm.com>
PipeWire 1.6.8 (2026-07-09)
This is a bugfix release that is API and ABI compatible with the previous
1.6.x releases.
Highlights
- Fix a data race in JACK that could cause lost MIDI events in ardour.
- Fix some unbounded memory allocations.
- Various small fixes.
PipeWire
- Avoid some graph recalcs, which fixes a bug when suspending a node
while it is active.
Modules
- Do Content-Length and allocation check in RAOP to avoid OOM errors.
- Fix a potential memory leak in the error path of client-node. (#5348)
SPA
- Fix filter-graph dynamic graph updates.
- Avoid 100% when unplugging a card.
- Fix filter-graph volumes when the filter is loaded inside a node with
hardware volume. (#5344)
- Add normalize and latency options to the SOFA filter. (#5322)
Bluetooth
- Fix a potential leak when transport fails to start.
Pulse-server
- Avoid stack exhaustion via unbounded alloca.
JACK
- Fix a data race in jack_port_get_buffer() when called from concurrent
threads, like in ardour. (#5324)
GStreamer
- Skip invalid crop metadata.
- Avoid crash because metadata listener was registered twice.
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
(cherry picked from commit 6dbf718454)
Signed-off-by: Ankur Tyagi <ankur.tyagi85@gmail.com>
Signed-off-by: Anuj Mittal <anuj.mittal@oss.qualcomm.com>
PipeWire 1.6.7 (2026-06-18)
This is a bugfix release that is API and ABI compatible with the previous
1.6.x releases.
Highlights
- Fix a race issue where some ports would stay silent after a rate change.
- Fix sync regressions between ALSA cards in some cases.
- Small fixes and improvements.
PipeWire
- Fix a scheduler regression where some driver nodes would not run
correctly and cause sync issues. (#5210)
- Fix a race issue with suspend on samplerate changes. It can cause ports
to be silent. (#3547)
Modules
- There are some locking issues in the RT portal, for now reduce the
DBus timeout to something more sane. Also disable portal RT for
pipewire and the pulse server. These are not usually run in a sandbox
and can go directly to RTKit.
- Fix potential incorrect delay in combine-stream.
SPA
- Fix a regression in ALSA period_size calculations. For non-power-of-2
periods, it would in some cases round down a a power-of-2, causing a
mismatch between requested and configured period_size. (#5302)
- Fix a potential segfault when removing a card because of bad ALSA api
usage. (#5255)
- Emit a route param update when card properties change. Otherwise, jack
port updates are not always reflected correctly.
Misc
- Make sure we don't deal with uninitialized spa_dict.
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
(cherry picked from commit ce604315bc)
Signed-off-by: Ankur Tyagi <ankur.tyagi85@gmail.com>
Signed-off-by: Anuj Mittal <anuj.mittal@oss.qualcomm.com>
- ncurses is not a required dependency -> make it a PACKAGECONFIG
PipeWire 1.6.6 (2026-05-26)
This is a bugfix release that is API and ABI compatible with the previous
1.6.x releases.
Highlights
- Fix a bug in the server code that could leave snap clients without
sound. (#5270)
- Relax LADSPA path loading again, absolute paths are only blocked in
unsafe cases.
- Fix a volume restore issue in filter-graph when using custom volume
controls.
- Small fixes and improvements.
PipeWire
- Fix the meta permission check on metadata.
- Make sure we don't accept too many fds.
- Fix potential race with buffer allocation and Suspend. (#3547)
SPA
- Relax LADSPA path loading. Absolute paths are allowed when loading
modules from a config file. They are now only blocked when loading
the pulse ladspa modules and filter-chains in nodes because those
can load ladspa plugins in other processes. (#5222)
- Fix a regression in the dither noise that was silent when no input
was available. (#5260)
- Fix volume initialization in filter-graph. (#5192)
Pulse-server
- Fix a bug in the server code that could leave snap clients without
sound. (#5270)
- Be more careful with the stream suspend messages and only send them
when the stream is monitoring. (#5273)
- Fix monitor mode in pavucontrol.
Tools
- Fix midifile SysEx writing in pw-cat and ensure the header is written
correctly on close.
- Make sure pw-cat does not try to convert Midi to UMP.
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
(cherry picked from commit d712bc991a)
Signed-off-by: Ankur Tyagi <ankur.tyagi85@gmail.com>
Signed-off-by: Anuj Mittal <anuj.mittal@oss.qualcomm.com>
PipeWire 1.6.5 (2026-05-13)
This is a bugfix release that is API and ABI compatible with the previous
1.6.x releases.
Highlights
- Fix muted output in some cases.
- Removed the pipe filter in filter-graph.
- More fixes and improvements.
PipeWire
- Fix an issue in pw-filter where it could end up in a loop where buffers
are stuck on a port and the port becomes silent. (#5249)
Modules
- Improve ROC receiver start/stop, fixes memory leaks. (#5250)
- Remove the pipe filter from filter-graph, it's broken by design and a
security nightmare.
- Fix the midi buffer size in jack-tunnel.
SPA
- Rate limit out-of-buffers errors. (#5249)
- Partially revert the line-out mute patch, it seems to break things and leave
audio muted when plugging-unplugging jacks. (#5246)
- Improve renegotiation in audioconvert when the graph rate changes and the
resampler was disabled. (#4933).
- Fix potential crash in alsa when logging.
Pulse-server
- A whole bunch of extra security checks and hardening fixes.
Older versions:
PipeWire 1.6.4 (2026-04-22)
This is a bugfix release that is API and ABI compatible with the previous
1.6.x releases.
Highlights
- Small improvements and seqfault fixes.
- Try to not emit ports that JACK doesn't understand. Fixes glitches in
ardour and other JACK apps.
PipeWire
- Refuse to load plugins and crash when pw_init() was not called. (!2784)
SPA
- Fix LADSPA plugin loading, support LADSPA_PATH ending with /
- Fix segfault in alsa-seq when removing devices in some cases. (#5221)
- Allow negative gain in mixer. (#5228)
- Improve alsa-seq port names, add : between client and port. (#5229)
- ACP: don’t override user-selected port on availability changes.
Bluetooth
- Backport some important fixes and minor improvements.
JACK
- Ignore non DSP ports to avoid emitting extra callbacks.
GStreamer
- Fix crop metadata.
Tools
- Fix WAVEX saving in pw-cat. (#5233)
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
(cherry picked from commit babcd87414)
Signed-off-by: Ankur Tyagi <ankur.tyagi85@gmail.com>
Signed-off-by: Anuj Mittal <anuj.mittal@oss.qualcomm.com>
WirePlumber 0.5.14
~~~~~~~~~~~~~~~~~~
Additions & Enhancements:
- Added per-device default volume configuration via the
``device.routes.default-{source,sink}-volume`` property, allowing device-specific volume
defaults (e.g. a comfortable default for internal speakers or no attenuation for HDMI) (!772)
- Added Lua 5.5 support; the bundled Lua subproject wrap has also been updated to 5.5.0
(!775, !788)
- Enhanced libcamera monitor to load camera nodes locally within the WirePlumber
process instead of the PipeWire daemon, eliminating race conditions that could occur
during initial enumeration and hotplug events (!790)
- Enhanced Bluetooth loopback nodes to always be created when a device supports both
A2DP and HSP/HFP profiles, simplifying the logic and making the BT profile autoswitch
setting take effect immediately without requiring device reconnection (!782)
- Enhanced Bluetooth loopback nodes to use ``target.object`` property instead of smart
filters, fixing issues that prevented users from setting them as default nodes and
also allowing smart filters to be used with them (#898; !792)
- Enhanced Bluetooth profile autoswitch logic with further robustness improvements,
including better headset profile detection using profile name patterns and resolving
race conditions by running profile switching after ``device/apply-profile`` in a
dedicated event hook (#926, #923; !776, !777, !808)
- Enhanced wpctl ``set-default`` command to accept virtual nodes (e.g.
``Audio/Source/Virtual``) in addition to regular device nodes (#896; !787)
- Improved stream linking to make the full graph rescan optional when linkable items
change, saving CPU on low-end systems and reducing audio startup latency when
connecting multiple streams in quick succession (!800)
- Allowed installation of systemd service units without libsystemd being present,
useful for distributions like Alpine Linux that allow systemd service subpackages
(!793)
- Allowed the ``mincore`` syscall in the WirePlumber systemd sandbox, required for
Mesa/EGL (e.g. for the libcamera GPUISP pipeline)
- Allowed passing ``WIREPLUMBER_CONFIG_DIR`` via the ``wp-uninstalled`` script,
useful for passing additional configuration paths in an uninstalled environment (!801)
Fixes:
- Removed Bluetooth sink loopback node, which was causing issues with KDE and GNOME (!794)
- Fixed default audio source selection to never automatically use ``Audio/Sink`` nodes
as the default source unless explicitly selected by the user (#886; !781)
- Fixed crash in ``state-stream`` when the Format parameter has a Choice for the
number of channels (#903; !795)
- Fixed BAP Bluetooth device set channel properties, where ``audio.position`` was
incorrectly serialized as a pointer address instead of the channel array (!786)
- Fixed memory leaks in ``wp_interest_event_hook_get_matching_event_types`` and in
the Lua ``LocalModule()`` implementation (!784, !810)
- Fixed HFP HF stream media class being incorrectly assigned due to
``api.bluez5.internal=true`` being set on HFP HF streams (!809)
- Fixed Lua 5.4 compatibility in ``state-stream`` script
- Updated translations: Bulgarian, Georgian, Kazakh, Swedish
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
- remove backport patch
PipeWire 1.6.3 (2026-04-09)
Highlights
- Fix some RAOP compatibility regressions.
- Fix segfault in the mixer in some cases.
- Most nodes now produce and consume MIDI1 again and avoid
conversions to and from UMP.
- Various small fixes and improvements.
PipeWire
- Fix regression with sample rate changes. (#5207)
- Fix a potential integer overflow in the memory mapping.
Modules
- Align RTP timestamps to make RAOP work on more devices. (#5167)
- Avoid crashes in RTP streams because of concurrent event
emmission.
- Avoid invalid fd usage in native-protocol with special crafted
messages.
- Fix properties and params enumeration in filter-chain (#5202).
SPA
- Fix compilation with -Werror=discarded-qualifiers
- Avoid OOB read in mix matrix. (#5176)
- Avoid loading plugins from absolute paths that are not in the
search path.
- Avoid MIDI conversions to and from UMP. (#5183)
Bluetooth
- Backport some fixes and avoid some crashes.
JACK
- Make sure timebase callback is never called with 0 frames.
- Increase the notify queue to avoid losing notifications.
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Set ac_cv_prog_cc_c23=no to prevent autoconf from detecting C23
compiler support, avoiding potential build failures as the package
is not yet fully ported to support C23 standard.
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
systemd_system_unitdir was accidentally used as the destination for
the user systemd unit, which means it overwrites the system unit.
Correct it to systemd_user_unitdir to fix starting with the system
unit.
Signed-off-by: Scott Murray <scott.murray@konsulko.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
Add a recommendation for a sans serif font to print text on images. But
fswebcam works fine without a font to save only plain images.
Signed-off-by: Jörg Sommer <joerg.sommer@navimatix.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
This commit adds a PACKAGECONFIG to build the av1 gdk-pixbuf-loader/thumbnailer
and enables it by default.
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
- add missing dependencies
- add PACKAGECONFIG for aom,svt-av1
1.4.1 - 2026-03-20
Changed since 1.4.0
Fix build with CMake 3.22
Update aom.cmd/LocalAom.cmake: v3.13.2
Update libxml2.cmd/LocalLibXml2.cmake: v2.15.2
Update libyuv.cmd/LocalLibyuv.cmake: 6067afde5 (1922)
Support long path names in Windows
Fix cicp management and memory leaks in avifgainmaputil #3102.
Removed since 1.4.0
Remove experimental status for the following options of avifenc: --progressive, --layered and --scaling-mode, and the extraLayerCount option of avifEncoder.
1.4.0 - 2026-03-04
Added since 1.3.0
Allow avifenc to read png or jpeg files through stdin using --stdin-format.
Support some Sample Transform schemes as defined in the version 1.2 of the AVIF specification.
Add an optional argument to the --depth flag of avifenc used to enable a bit depth extension scheme in the encoded file.
Add support for converting jpeg files with Apple style gain maps.
Add support for PNG cICP chunk when decoding PNG files. If a PNG file contains a cICP chunk and other color information chunks, such as iCCP (ICC profile), the other chunks are ignored as per the PNG Specification Third Edition Section 4.3.
Support reading Sample-Transform-based 16-bit AVIF files when avifDecoder::imageContentToDecode & AVIF_IMAGE_CONTENT_SAMPLE_TRANSFORMS is not zero.
Support Sample Transform derived image items with grid input image items.
Add --sato flag to avifdec to enable Sample Transforms support at decoding.
Add --grid option to avifgainmaputil.
Apply clean aperture crop, rotation and mirror when decoding to PNG or JPEG. Remove orientation information from Exif if present.
Add avif::RGBImageCleanup to the C++ API.
Changed since 1.3.0
Set avifDecoder::image->depth to the same value after avifDecoderParse() as after avifDecoderNextImage() when the file to decode contains a 'sato' derived image item.
avifdec only enables Sample Transform decoding when --depth is set to 16.
Update dav1d.cmd/dav1d_android.sh/LocalDav1d.cmake: 1.5.3
Update googletest.cmd/LocalGTest.cmake: v1.17.0
Update libgav1.cmd: v0.20.0
Update libjpeg.cmd/LocalJpeg.cmake: 3.1.3
Update libyuv.cmd/LocalLibyuv.cmake: deeb764bb (1922)
Update libsharpyuv.cmd/LocalLibsharpyuv.cmake: v1.6.0
Update libxml2.cmd/LocalLibXml2.cmake: v2.15.1
Update aom.cmd/LocalAom.cmake: v3.13.1
Update LocalAvm.cmake: research-v13.0.0
Update rav1e.cmd/LocalRav1e.cmake: cargo-c v0.10.20, corrosion v0.6.1, rav1e v0.8.1
Update svt.cmd/svt.sh/LocalSvt.cmake: v4.0.1
Update zlibpng.cmd/LocalZlibpng.cmake: libpng 1.6.55, zlib 1.3.2
Fix grayscale conversion when changing the bit depth.
Bump cmake_minimum_required from 3.13 to 3.22
Associate transformative properties with alpha auxiliary image items.
Always forward the CICP color primaries, transfer characteristics, and matrix coefficients to the AV1 encoder, which writes them in the Sequence Header OBU, for compatibility with libraries that wrongly ignore the colr box.
Use a "quality to quantizer (QP)" mapping formula designed for AOM_TUNE_IQ.
Set tuning before applying the user-provided specific aom codec options.
Use AOM_TUNE_PSNR by default when encoding alpha with libaom because AOM_TUNE_SSIM causes ringing for alpha.
Use AOM_TUNE_IQ by default when encoding still non-RGB color samples with libaom v3.13.0 or later.
Converting an image containing a gain map using avifenc with the --grid flag now also splits the gain map into a grid.
In avifenc, set Exif orientation to 1 (no transformation) when converting JPEGs to AVIF.
Use all-intra encoding for a layered image if the total number of layers is 2 and the quality of the first layer is very low (q <= 10).
Signed-off-by: Markus Volk <f_l_k@t-online.de>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>
While in this case `RRECOMMENDS:class-target +=` wouldn't result in any
unwanted override, there is no guarantee there won't be a change, which
would be hidden by this override. To avoid any surprises in the future
let's use `:append:class-target =` syntax here.
Signed-off-by: Michal Sieron <michalwsieron@gmail.com>
Signed-off-by: Khem Raj <khem.raj@oss.qualcomm.com>