gst-plugins-base issueshttps://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues2020-08-11T07:35:10Zhttps://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/809GL Wayland/X11/EGL gir files are not pointing at necessary header files2020-08-11T07:35:10ZSebastian DrögeGL Wayland/X11/EGL gir files are not pointing at necessary header filesThey all list
```xml
<c:include name="gst/gl/gl.h"/>
```
Which doesn't make it possible to find any of the types exported by these libraries.
CC @gdesmott @ystreetThey all list
```xml
<c:include name="gst/gl/gl.h"/>
```
Which doesn't make it possible to find any of the types exported by these libraries.
CC @gdesmott @ystreet1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/805audioaggregator: Add support for sample selection API2020-08-07T23:12:16ZSebastian Drögeaudioaggregator: Add support for sample selection APIOtherwise the log will be full of warnings like the following
```
aggregator gstaggregator.c:1323:gst_aggregator_aggregate_func:<audiomixer0> Subclass should call gst_aggregator_selected_samples() from its aggregate implementation.
```
...Otherwise the log will be full of warnings like the following
```
aggregator gstaggregator.c:1323:gst_aggregator_aggregate_func:<audiomixer0> Subclass should call gst_aggregator_selected_samples() from its aggregate implementation.
```
CC @meh1.17.90GStreamer Marge Botgitlab-merge-bot@gstreamer-foundation.orgGStreamer Marge Botgitlab-merge-bot@gstreamer-foundation.orghttps://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/801compile regression: undefined reference to g_hash_table_steal_extended2020-08-03T07:57:33ZU. Artie Eoffcompile regression: undefined reference to g_hash_table_steal_extendedSince commit 2faeb7d394 (!728), on Ubuntu Xenial and Bionic:
```
08:55:25,945 INFO - [722/759] cc -o tests/examples/compositor/signals 'tests/examples/compositor/8f0d26f@@signals@exe/signals.c.o' -L/opt/media/install/lib -Wl,--as-neede...Since commit 2faeb7d394 (!728), on Ubuntu Xenial and Bionic:
```
08:55:25,945 INFO - [722/759] cc -o tests/examples/compositor/signals 'tests/examples/compositor/8f0d26f@@signals@exe/signals.c.o' -L/opt/media/install/lib -Wl,--as-needed -Wl,--no-undefined -Wl,-Bsymbolic-functions -g3 -Wl,--start-group /opt/media/install/lib/libgstreamer-1.0.so /usr/lib/x86_64-linux-gnu/libgobject-2.0.so /usr/lib/x86_64-linux-gnu/libglib-2.0.so /opt/media/install/lib/libgstbase-1.0.so -Wl,--end-group
08:55:25,945 INFO - FAILED: tests/examples/compositor/signals
08:55:25,945 INFO - cc -o tests/examples/compositor/signals 'tests/examples/compositor/8f0d26f@@signals@exe/signals.c.o' -L/opt/media/install/lib -Wl,--as-needed -Wl,--no-undefined -Wl,-Bsymbolic-functions -g3 -Wl,--start-group /opt/media/install/lib/libgstreamer-1.0.so /usr/lib/x86_64-linux-gnu/libgobject-2.0.so /usr/lib/x86_64-linux-gnu/libglib-2.0.so /opt/media/install/lib/libgstbase-1.0.so -Wl,--end-group
08:55:25,945 INFO - tests/examples/compositor/8f0d26f@@signals@exe/signals.c.o: In function `pad_buffer_consumed_cb':
08:55:25,945 INFO - /opt/media/build/gst-plugins-base/__meson_build/../tests/examples/compositor/signals.c:77: undefined reference to `g_hash_table_steal_extended'
08:55:25,945 INFO - tests/examples/compositor/8f0d26f@@signals@exe/signals.c.o: In function `check_aggregated_buffer':
08:55:25,945 INFO - /opt/media/build/gst-plugins-base/__meson_build/../tests/examples/compositor/signals.c:38: undefined reference to `g_hash_table_steal_extended'
08:55:25,945 INFO - collect2: error: ld returned 1 exit status
```
It compiles fine on Ubuntu Focal, Ubuntu Eoan, Fedora 31 and Fedora 32.1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/791oggdemux: regression with libogg 1.3.42020-07-14T15:06:44ZGuillaume Desmottesoggdemux: regression with libogg 1.3.4`./gst-uninstalled.py gst-validate-launcher validate.http.transcode.to_vorbis_and_vp9_in_webm.vorbis_theora_1_ogg`
This test is failing on Fedora 32 and Ubuntu 20.04. Both are using libogg `1.3.4` while our CI (Fedora 30) has `1.3.3`.
...`./gst-uninstalled.py gst-validate-launcher validate.http.transcode.to_vorbis_and_vp9_in_webm.vorbis_theora_1_ogg`
This test is failing on Fedora 32 and Ubuntu 20.04. Both are using libogg `1.3.4` while our CI (Fedora 30) has `1.3.3`.
It seems that `oggdemux` is no longer able to parse the file:
```
GST_DEBUG="*ogg*:6" gst-launch-1.0 filesrc location=subprojects/gst-integration-testsuites/medias/defaults/ogg/vorbis_theora.1.ogg ! oggdemux ! fakesink
0:00:00.217097604 912704 0xf9da40 DEBUG oggdemux gstoggdemux.c:5264:gst_ogg_demux_plugin_init: binding text domain gst-plugins-base-1.0 to locale dir /home/cassidy/dev/gst/master/gst-build/install/share/locale
Setting pipeline to PAUSED ...
0:00:00.217662698 912704 0xf9da40 DEBUG oggdemux gstoggdemux.c:2003:gst_ogg_demux_query_duration_push:<oggdemux0> Stream is not seekable, seeking disabled
0:00:00.217733779 912704 0xf9da40 DEBUG oggdemux gstoggdemux.c:5135:gst_ogg_demux_sink_activate:<oggdemux0:sink> activating pull
Pipeline is PREROLLING ...
0:00:00.217979466 912704 0xd93b60 DEBUG oggdemux gstoggdemux.c:4452:gst_ogg_demux_find_chains:<oggdemux0> file length 131
0:00:00.218001762 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2595:gst_ogg_demux_seek:<oggdemux0> seeking to 0
0:00:00.218015984 912704 0xd93b60 LOG oggdemux gstoggdemux.c:4133:gst_ogg_demux_read_chain:<oggdemux0> reading chain at 0
0:00:00.218031111 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2702:gst_ogg_demux_get_next_page:<oggdemux0> get next page, current offset 0, bytes boundary -1
0:00:00.218061928 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2717:gst_ogg_demux_get_next_page:<oggdemux0> pageseek gave 0
0:00:00.218076955 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2729:gst_ogg_demux_get_next_page:<oggdemux0> need more data
0:00:00.218095392 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2613:gst_ogg_demux_get_data:<oggdemux0> get data 0 131 -1
0:00:00.218156949 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2717:gst_ogg_demux_get_next_page:<oggdemux0> pageseek gave -8500
0:00:00.218177896 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2722:gst_ogg_demux_get_next_page:<oggdemux0> skipped -8500 bytes, offset 8500
0:00:00.218195438 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2717:gst_ogg_demux_get_next_page:<oggdemux0> pageseek gave 0
0:00:00.218209790 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2729:gst_ogg_demux_get_next_page:<oggdemux0> need more data
0:00:00.218228529 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2613:gst_ogg_demux_get_data:<oggdemux0> get data 8500 131 -1
0:00:00.218253476 912704 0xd93b60 WARN oggdemux gstoggdemux.c:2666:gst_ogg_demux_get_data:<oggdemux0> got -3 (eos) from pull range
0:00:00.218273742 912704 0xd93b60 LOG oggdemux gstoggdemux.c:2752:gst_ogg_demux_get_next_page:<oggdemux0> returning -3
0:00:00.218289815 912704 0xd93b60 DEBUG oggdemux gstoggdemux.c:4144:gst_ogg_demux_read_chain:<oggdemux0> Reached EOS, done reading end chain
0:00:00.218306157 912704 0xd93b60 DEBUG oggdemux gstoggdemux.c:4184:gst_ogg_demux_read_chain:<oggdemux0> done reading chains
0:00:00.218324120 912704 0xd93b60 WARN oggdemux gstoggdemux.c:4513:gst_ogg_demux_find_chains:<oggdemux0> error: can't get first chain
ERROR: from element /GstPipeline:pipeline0/GstOggDemux:oggdemux0: Could not demultiplex stream.
Additional debug info:
../subprojects/gst-plugins-base/ext/ogg/gstoggdemux.c(4513): gst_ogg_demux_find_chains (): /GstPipeline:pipeline0/GstOggDemux:oggdemux0:
can't get first chain
ERROR: pipeline doesn't want to preroll.
0:00:00.218531554 912704 0xd93b60 LOG oggdemux gstoggdemux.c:4961:gst_ogg_demux_loop:<oggdemux0> pausing task, reason error
Setting pipeline to NULL ...
0:00:00.218551956 912704 0xd93b60 WARN oggdemux gstoggdemux.c:4993:gst_ogg_demux_loop:<oggdemux0> error: Internal data stream error.
0:00:00.218564424 912704 0xd93b60 WARN oggdemux gstoggdemux.c:4993:gst_ogg_demux_loop:<oggdemux0> error: streaming stopped, reason error (-5)
ERROR: from element /GstPipeline:pipeline0/GstOggDemux:oggdemux0: Internal data stream error.
Additional debug info:
../subprojects/gst-plugins-base/ext/ogg/gstoggdemux.c(4993): gst_ogg_demux_loop (): /GstPipeline:pipeline0/GstOggDemux:oggdemux0:
streaming stopped, reason error (-5)
ERROR: pipeline doesn't want to preroll.
0:00:00.218638926 912704 0xd93b60 WARN oggdemux gstoggdemux.c:5011:gst_ogg_demux_loop:<oggdemux0> error: EOS before finding a chain
ERROR: from element /GstPipeline:pipeline0/GstOggDemux:oggdemux0: Could not demultiplex stream.
Additional debug info:
../subprojects/gst-plugins-base/ext/ogg/gstoggdemux.c(5011): gst_ogg_demux_loop (): /GstPipeline:pipeline0/GstOggDemux:oggdemux0:
EOS before finding a chain
ERROR: pipeline doesn't want to preroll.
Freeing pipeline ...
```1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/789gl: provide platform specific pkg-config files2020-08-06T12:32:21ZTim-Philipp Müllertim@centricular.comgl: provide platform specific pkg-config filesFollow-up from #735.
> I think, [we want to provide] at least one [`gstreamer-gl-*` pkg-config file] for each of the extra platform-specific headers. At the moment that is x11, wayland, egl and gl-prototypes. Others may be useful instea...Follow-up from #735.
> I think, [we want to provide] at least one [`gstreamer-gl-*` pkg-config file] for each of the extra platform-specific headers. At the moment that is x11, wayland, egl and gl-prototypes. Others may be useful instead of the pc variables in `gstreamer-gl-1.0` (or compile-checking `GST_GL_HAVE_*` from gstglconfig.h) but I don't think adding one for each option is strictly required. This also would conveniently match nicely with the extra .gir files needed for platform-specific functions.
Related:
- [gl: Various issues with generated gir file](https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/651)
- https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/661
- https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/6081.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/785gst-device-monitor: v4l2 provider is not queried if v4l2-gudev is disabled2020-07-13T07:48:32ZVíctor Manuel Jáquez Lealgst-device-monitor: v4l2 provider is not queried if v4l2-gudev is disabledhttps://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/427 introduced a regression when gstv4l2 is compiled without gudev support: gst-device-monitor doesn't report back any device.
I'm not sure if the proper fix sho...https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/427 introduced a regression when gstv4l2 is compiled without gudev support: gst-device-monitor doesn't report back any device.
I'm not sure if the proper fix should be here or in v4l2deviceprovider.1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/784decodebin3: Avoid selection recalculation2020-07-07T13:09:01ZEdward Herveydecodebin3: Avoid selection recalculationCurrently the way decodebin3 figures out whether it should modify the current selection is not reliable.
What it should do, after posting a new collection is to check whether there already was a selection requested for **that** collecti...Currently the way decodebin3 figures out whether it should modify the current selection is not reliable.
What it should do, after posting a new collection is to check whether there already was a selection requested for **that** collection.
The order of events is:
* Updated collection is stored as `dbin->collection`
* Collection is posted on the bus
* `update_requested_selection` is called
Three things should happen:
* When the dbin->collection is modified, dbin->select_streams_seqnum should be resetted
* *UNCHANGED* : If a select_streams event comes in, we store it along with the seqnum
* Within `update_requested_selection`, and with the SELECTION lock taken, Only calculate a new selection **IFF** the seqnum is not set, otherwise just use the selection.1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/783GST_VIDEO_TRANSFER_BT2020_10 alias causes negotiation issues2020-07-03T09:56:21ZSebastian DrögeGST_VIDEO_TRANSFER_BT2020_10 alias causes negotiation issuesThis was added in 9d2f9d16. For caps negotiation this is a different value but in practice for conversion it's the same as `BT709`.
At least `videoconvert` needs to get some support to understand that it's the same and not do any conver...This was added in 9d2f9d16. For caps negotiation this is a different value but in practice for conversion it's the same as `BT709`.
At least `videoconvert` needs to get some support to understand that it's the same and not do any conversion.1.17.90Vivia NikolaidouVivia Nikolaidouhttps://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/780gst-device-monitor-1.0 no device on macOS2020-07-06T19:31:08ZRoman Shpuntovgst-device-monitor-1.0 no device on macOSmacOS 10.15.5, gstreamer 1.17.1
app output:
```
gst-device-monitor-1.0
Probing devices...
```macOS 10.15.5, gstreamer 1.17.1
app output:
```
gst-device-monitor-1.0
Probing devices...
```1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/758(uri)decodebin3: Remove custom force-sw-decoders property2022-02-15T11:16:13ZEdward Hervey(uri)decodebin3: Remove custom force-sw-decoders propertySadly I wasn't quite paying attention when https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/485 went in.
I'd *very* much like to have that force-sw-decoders property removed from (uri)decodebin3 and instead hav...Sadly I wasn't quite paying attention when https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/merge_requests/485 went in.
I'd *very* much like to have that force-sw-decoders property removed from (uri)decodebin3 and instead have a generic way of influencing what elements should be plugged/tried.
## Proposal
Add a new signal for selecting a decoder **if and only if** there is more than
one candidate decoder for a given stream.
``` c
/**
* decoder-select
*
* Return : NULL to let (uri)decodebin3 do its own selection,
* or the specific elementfactory from the candidates.
*/
GstElementFactory* decoder-select (GstElement *decodebin3,
GstStream *stream,
GValueArray *candidates)
```
The default implementation would return NULL to let decodebin3 do its
(pre-force-sw-decoder-patch) standard behaviour. And playbin3 can still strip
out candidates based on hardware flags, phase of the moon, day of the month or
whatever you chose.
## Use-cases
* Match ideal decoder based on configured sink (ex: match optimal decoder to
configured sink).
* Avoid software decoders
* **FIXME** What is the rationale for this use-case ?1.17.90Edward HerveyEdward Herveyhttps://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/750gir fails to parse GST_GL_COLOR_CONVERT_FORMATS2020-07-10T11:32:20ZVivia Nikolaidougir fails to parse GST_GL_COLOR_CONVERT_FORMATS```
#if G_BYTE_ORDER == G_LITTLE_ENDIAN
#define COLOR_CONVERT_EXT_FORMATS \
", BGR10A2_LE, RGB10A2_LE, P010_10LE, P012_LE, P016_LE, Y212_LE, Y412_LE"
#else
#define COLOR_CONVERT_EXT_FORMATS \
", P010_10BE, P012_BE, P016_BE, Y212_...```
#if G_BYTE_ORDER == G_LITTLE_ENDIAN
#define COLOR_CONVERT_EXT_FORMATS \
", BGR10A2_LE, RGB10A2_LE, P010_10LE, P012_LE, P016_LE, Y212_LE, Y412_LE"
#else
#define COLOR_CONVERT_EXT_FORMATS \
", P010_10BE, P012_BE, P016_BE, Y212_BE, Y412_BE"
#endif
#define GST_GL_COLOR_CONVERT_FORMATS "{ RGBA, RGB, RGBx, BGR, BGRx, BGRA, xRGB, " \
"xBGR, ARGB, ABGR, Y444, I420, YV12, Y42B, " \
"Y41B, NV12, NV21, NV16, NV61, YUY2, UYVY, Y210, AYUV, " \
"VUYA, Y410, GRAY8, GRAY16_LE, GRAY16_BE, " \
"RGB16, BGR16, ARGB64 " COLOR_CONVERT_EXT_FORMATS "}"
```
```
Constant value mismatch for GST_GL_COLOR_CONVERT_FORMATS
Rust: "{ RGBA, RGB, RGBx, BGR, BGRx, BGRA, xRGB, xBGR, ARGB, ABGR, Y444, I420, YV12, Y42B, Y41B, NV12, NV21, NV16, NV61, YUY2, UYVY, Y210, AYUV, VUYA, Y410, GRAY8, GRAY16_LE, GRAY16_BE, RGB16, BGR16, ARGB64 "
C: "{ RGBA, RGB, RGBx, BGR, BGRx, BGRA, xRGB, xBGR, ARGB, ABGR, Y444, I420, YV12, Y42B, Y41B, NV12, NV21, NV16, NV61, YUY2, UYVY, Y210, AYUV, VUYA, Y410, GRAY8, GRAY16_LE, GRAY16_BE, RGB16, BGR16, ARGB64 , BGR10A2_LE, RGB10A2_LE, P010_10LE, P012_LE, P016_LE, Y212_LE, Y412_LE}"
```
See: https://gitlab.freedesktop.org/vivia/gstreamer-rs-sys/-/jobs/25177691.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/735gl: WARNING: The variable(s) GL_CFLAGS in the input file gstreamer-gl.pc.in a...2020-07-08T13:41:58ZDominique Leuenbergergl: WARNING: The variable(s) GL_CFLAGS in the input file gstreamer-gl.pc.in are not present in the given configuration data.## .pc files differ between autoconf and meson build system
I tried building gst-plugins-base using meson instead of autoconf. Generally, this 'works', but the generated .pc file is missing relevant CFLAG entries, which makes other, dep...## .pc files differ between autoconf and meson build system
I tried building gst-plugins-base using meson instead of autoconf. Generally, this 'works', but the generated .pc file is missing relevant CFLAG entries, which makes other, dependent packages possibly fail.
Diff of autoconf vs meson generated .pc file (gstreamer-gl.pc - did not check all the others yet)
```
> diff -ur /usr/lib64/pkgconfig/gstreamer-gl-1.0.pc gstreamer-gl-1.0.pc
--- /usr/lib64/pkgconfig/gstreamer-gl-1.0.pc 2020-02-07 20:44:37.000000000 +0100
+++ gstreamer-gl-1.0.pc 2020-03-05 18:26:19.000000000 +0100
@@ -1,11 +1,11 @@
prefix=/usr
-exec_prefix=/usr
-libdir=/usr/lib64
-includedir=/usr/include/gstreamer-1.0
-pluginsdir=/usr/lib64/gstreamer-1.0
-gl_platforms=glx egl
-gl_winsys=gbm wayland x11
-gl_apis=gles2 gl
+exec_prefix=${prefix}
+libdir=${prefix}/lib64
+includedir=${prefix}/include/gstreamer-1.0
+pluginsdir=${prefix}/lib64/gstreamer-1.0
+gl_platforms=egl glx
+gl_winsys=wayland x11 gbm
+gl_apis=gles2 gl
Name: GStreamer OpenGL Plugins Libraries
Description: Streaming media framework, OpenGL plugins libraries
@@ -13,4 +13,4 @@
Requires: gstreamer-video-1.0 gstreamer-base-1.0 gstreamer-1.0
Libs: -L${libdir} -lgstgl-1.0
-Cflags: -I${includedir} -I${libdir}/gstreamer-1.0/include -I/usr/include/libdrm -I/usr/include/wayland
+Cflags: -I${includedir} -I${libdir}/gstreamer-1.0/include
```
The Cflags is the critical one, as paths to other headers we relay on are not correctly added to the .pc file1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/720[REGRESSION] device-monitor: no longer show devices from provider that can't ...2020-07-06T19:31:08ZNicolas Dufresne[REGRESSION] device-monitor: no longer show devices from provider that can't monitorThis regression was introduced by gst-plugins-base!427 I believe. gstreamer#473 could be one way to fix it, but we need something prior to 1.18 as the gst-device-monitor-1.0 tool no longer show some devices.This regression was introduced by gst-plugins-base!427 I believe. gstreamer#473 could be one way to fix it, but we need something prior to 1.18 as the gst-device-monitor-1.0 tool no longer show some devices.1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/684video: Lots of new formats not added to the scalers fast-path2020-07-10T14:15:23ZSebastian Drögevideo: Lots of new formats not added to the scalers fast-pathSee https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/blob/30672ba7d134553e59935ddc875104adba26f25c/gst-libs/gst/video/video-converter.c#L6591
E.g. `GST_VIDEO_FORMAT_GBRA` (1.12) is missing and all the 10/12/16 bit formats. Most...See https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/blob/30672ba7d134553e59935ddc875104adba26f25c/gst-libs/gst/video/video-converter.c#L6591
E.g. `GST_VIDEO_FORMAT_GBRA` (1.12) is missing and all the 10/12/16 bit formats. Most of them should also be possible to go directly through `convert_scale_planes`1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/651gl: Various issues with generated gir file2020-08-06T04:33:49ZSebastian Drögegl: Various issues with generated gir fileThe meson build includes lots of EGLImage related types that the autotools build does not include, is this intentional? If it is
* `gst_egl_get_error_string()`'s return value has to be fixed so it is a type that gir actually knows about...The meson build includes lots of EGLImage related types that the autotools build does not include, is this intentional? If it is
* `gst_egl_get_error_string()`'s return value has to be fixed so it is a type that gir actually knows about. `EGLint` it is not. Something like the following
```diff
@@ -8634,7 +8634,7 @@ you are writing to OpenGL. Conversely, combining #GST_MAP_GL with
<parameters>
<parameter name="err" transfer-ownership="none">
<doc xml:space="preserve">an EGL error code</doc>
- <type c:type="EGLint"/>
+ <type name="gint" c:type="EGLint"/>
</parameter>
</parameters>
</function>
```
* `GstGLMemoryEGLAllocator` references `GLMemoryEGLAllocatorClass` as class struct, but the class struct is not actually in the gir file
CC @ystreet @nirbheek
Marking as blocker because this would cause regressions once meson is the only remaining build system :)1.17.90https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/640egl: Might leak displays2022-09-16T06:20:43ZSebastian Drögeegl: Might leak displaysThe following discussion from !314 should be addressed:
- [ ] @ystreet started a [discussion](https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/merge_requests/314#note_193724): (+1 comment)
> This is not always true. An E...The following discussion from !314 should be addressed:
- [ ] @ystreet started a [discussion](https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/merge_requests/314#note_193724): (+1 comment)
> This is not always true. An EGLDisplay can be non-foreign from a foreign native display and in such cases, `eglTerminate` will not be called and thus can cause leaks.1.17.90