ci/mingw: run meson tests #14207
ci/mingw: run meson tests #14207
Conversation
|
Download the artifacts for this pull request: |
|
By getting rid of subprocesses, you could simplify the |
kasper93
force-pushed
the
ci_mingw_test
branch
2 times, most recently
from
70f2bd3
to
178af33
Compare
Removed what I've seen not needed. |
| @@ -141,7 +148,7 @@ _ffmpeg () { | |||
| --pkg-config=pkg-config --target-os=mingw32 | |||
| --enable-cross-compile --cross-prefix=$TARGET- --arch=${TARGET%%-*} | |||
| --cc="$CC" --cxx="$CXX" $commonflags | |||
| --disable-{doc,programs,muxers,encoders} | |||
| --disable-{doc,programs} | |||
There was a problem hiding this comment.
I'd rather only enable one relevant muxer or encoder for the tests (as on the line below)
There was a problem hiding this comment.
What if user wants to use different format? What is the gain to exclude some of them?
There was a problem hiding this comment.
Most important encoders like x264/x265 aren't ffmpeg built-in anyway, so the users likely still won't get what they want. At this point most built-in encoders are unnecessary bloat.
There was a problem hiding this comment.
But what is the gain of excluding what is available?
There was a problem hiding this comment.
I see no significant difference. In cached state it is 4s difference, clean build was also not visibly slower, but I can test that to confirm.
There was a problem hiding this comment.
This excludes more than 200 files so it should definitely make a difference uncached.
There was a problem hiding this comment.
Execution times of Build libraries step:
| master | #14207 | ||
|---|---|---|---|
| i386 | cached | 210 s | 170 s | 80 % |
| i386 | clean | 976 s | 1026 s | 105 % |
| x86_64 | cached | 262 s | 247 s | 94 % |
| x86_64 | clean | 1083 s | 1162 s | 107 % |
It is slower, but the difference in my opinion is not significant enough to making some arbitrary list of enabled/disabled muxers. Note that we almost never do full rebuild, I needed to manually remove cache from my fork to test that.
Tested on CI with the following builds:
https://github.com/mpv-player/mpv/actions/runs/9191373753/job/25277639660
https://github.com/mpv-player/mpv/actions/runs/9183181995/job/25253412400
https://github.com/kasper93/mpv/actions/runs/9192050397/job/25279896478
https://github.com/kasper93/mpv/actions/runs/9192086594/job/25279911198
There was a problem hiding this comment.
Most important encoders like x264/x265 aren't ffmpeg built-in anyway
At least you could save png/jpg screenshot... or something like that.
mktemp is good enough for this test purposes and also tests if libmpv is able to create a file that doesn't exist.
Helps with testing during cross-compilation and avoids external dependency. Output maybe is not that nice, but this output in our tests is not useful anyway. We know if it changes it is beacuse one of the dependency version changed.
If pathcch is not available paths are not normalized on Windows. Fixes tests on ancient mingw.
Allows running complied binaries on build machine.
pkgconfig is deprecated in favor of pkg-config. Define both for compatibility with meson < 1.3.0. https://mesonbuild.com/Release-notes-for-1-3-0.html#machine-files-pkgconfig-field-deprecated-and-replaced-by-pkgconfig
|
No more comments for a week, let's merge this. Can be improved if needed. |
No description provided.