gst-plugins-base issueshttps://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues2021-09-24T13:26:07Zhttps://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/820video-convert: Add fast paths from/to NV122021-09-24T13:26:07ZSebastian Drögevideo-convert: Add fast paths from/to NV12It's a quite common format nowadays, on the level of I420, and we should probably add a few more fast paths for it. Like one between I420 and NV12, but maybe also all the others we already have for I420.It's a quite common format nowadays, on the level of I420, and we should probably add a few more fast paths for it. Like one between I420 and NV12, but maybe also all the others we already have for I420.https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/765audioresample: add OpenPOWER ppc64 support2021-09-24T13:25:53ZDaniel Pocockaudioresample: add OpenPOWER ppc64 supportaudioresample includes some custom assembly
There is growing interest in the OpenPOWER architecture.
IBM is currently offering bounties to developers who will port free software projects to OpenPOWER
Providing an implementation of the...audioresample includes some custom assembly
There is growing interest in the OpenPOWER architecture.
IBM is currently offering bounties to developers who will port free software projects to OpenPOWER
Providing an implementation of the code for OpenPOWER ISA appears to be a good idea for a bounty
Alternatively, could the custom assembly be managed as part of liborc?
A bounty was offered for similar work in ffmpeg libswscale and a developer has provided a solution for that
https://www.bountysource.com/issues/34315232-power8-vsx-vectorization-libswscale-input-c
https://patchwork.ffmpeg.org/project/ffmpeg/patch/1585056463-7934-1-git-send-email-pestov.vyach@yandex.ru/https://gitlab.freedesktop.org/gstreamer/gst-plugins-base/-/issues/742videoencoder: Traversing queued force key unit events can be optimized2021-09-24T13:25:43ZSeungha Yangseungha@centricular.comvideoencoder: Traversing queued force key unit events can be optimized`videoencoder` baseclass stores upstream/downstream force key unit events in `GList` and whenever new event is arrived, append it at the end of list (i.e., `g_list_append()`).
Then, encoder traverses the list per incoming frame. It's ob...`videoencoder` baseclass stores upstream/downstream force key unit events in `GList` and whenever new event is arrived, append it at the end of list (i.e., `g_list_append()`).
Then, encoder traverses the list per incoming frame. It's obviously suboptimal