Closed pykpkg47 closed 1 year ago
I am a bot, here are the test results for this PR: https://ci-tests.linuxserver.io/lspipepr/tvheadend/52c3ed3e-pkg-31e93f3d-pr-209/index.html https://ci-tests.linuxserver.io/lspipepr/tvheadend/52c3ed3e-pkg-31e93f3d-pr-209/shellcheck-result.xml
I am a bot, here are the test results for this PR: https://ci-tests.linuxserver.io/lspipepr/tvheadend/52c3ed3e-pkg-31e93f3d-pr-209/index.html https://ci-tests.linuxserver.io/lspipepr/tvheadend/52c3ed3e-pkg-31e93f3d-pr-209/shellcheck-result.xml
I am a bot, here are the test results for this PR: https://ci-tests.linuxserver.io/lspipepr/tvheadend/52c3ed3e-pkg-31e93f3d-pr-209/index.html https://ci-tests.linuxserver.io/lspipepr/tvheadend/52c3ed3e-pkg-31e93f3d-pr-209/shellcheck-result.xml
This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.
Description:
there exist a optimized libdvbcsa which speeds up encrypted streams a lot (especially on arm32v7 and aarch64)
Benefits of this PR and context:
on x86_64 i used the sse2 option which is mandatory on x64 ( there are ssse3 and avx2 options too ) on arm32v7 and aarch64 the neon extension which is mandatory on aarch64 and on arm32v7 not but most arm32v7 have it except some nvidie cpus(tegra 2) , if this is a problem the option can be disabled on arm32v7 and it still will be faster than orginal libdvbcsa
Benefits: less cpu usage on encrpted streams
How Has This Been Tested?
tested on allwiner h3 cpu before 10 - 12 load per stream after 4-5 % on https://github.com/pykpkg47/docker-tvheadend_image_mod/pkgs/container/tvheadend you can find docker images with the pr to test it yourself
Source / References:
https://tvheadend.org/boards/13/topics/35411 https://github.com/glenvt18/libdvbcsa