Vladimír Vondruš
90f614aa84
package/ci: unify comments.
4 years ago
Vladimír Vondruš
9e56e6e5f4
package/ci: no reason to not use GCC 4.8 also for Vulkan.
...
There isn't any STL used on the interface to libulkan.so, so the ABI
mismatch shouldn't be a problem.
4 years ago
Vladimír Vondruš
33385a7670
package/ci: use Ubuntu 18.04 on CircleCI.
4 years ago
Vladimír Vondruš
4b6248cfb2
package/ci: retry Android tests 10x more after a two minute timeout.
...
Should finally get rid of random CI hangups. No, three time was NOT
enough, and some GL tests/benchmarks need more than 15 seconds.
4 years ago
Vladimír Vondruš
d087dbb349
package/ci: parametrize CMake download.
4 years ago
Vladimír Vondruš
65e09afa75
package/ci: xcode 11.4 image on CircleCI is useless now.
...
Waiting for 2+ hours for a VM assignment and then failing on a timeout.
4 years ago
Vladimír Vondruš
8d604a4803
package/ci: Emscripten 2.0.17 to 2.0.24 is utterly broken, skip.
...
UGH.
4 years ago
Pablo Escobar
4350f1358a
package/ci: Run emscripten-webgl2 on Emscripten 2.0.17
...
Required for Buffer::data()/subData() used by GLES3 tests. Otherwise
all the tests would have to check the Emscripten version and skip if
it's too old.
4 years ago
Vladimír Vondruš
e6b0a746cc
CMake: fix and harden Git version detection on Windows.
...
Same change as in Corrade.
4 years ago
Vladimír Vondruš
0a8a3f0f6d
Don't use git://.
4 years ago
Vladimír Vondruš
f44ea76ada
package/archlinux: run tests faster.
...
I'm tired of waiting for 5 Vukan tests all getting stuck on device
enumeration for 15 seconds for NO REASON.
4 years ago
Vladimír Vondruš
5a11fc3df7
package/ci: I know why is it like this now.
...
Because the executor is a VM, not a Docker thingamajig.
4 years ago
Vladimír Vondruš
822fa0d644
Updated copyright year.
4 years ago
Vladimír Vondruš
89e3c52dfa
Bye, Travis! You won't be missed.
...
The only leftover is the platform-specific info, which might still be
useful to derive info from for the non-shit CI providers.
4 years ago
Vladimír Vondruš
d3a411ca72
package/ci: add an iOS GLES3 build.
...
I don't think GLES2 is worth bothering with on this platform, given that
GL as a whole is deprecated by that company.
4 years ago
Vladimír Vondruš
a7b94a4b09
package/ci: add an ARM64 build.
4 years ago
Vladimír Vondruš
ba7267ba0c
package/ci: enable image importers for Vulkan tests.
...
The VkMeshVkTest was mostly skipped otherwise.
4 years ago
Vladimír Vondruš
fce8fc3bfd
package/ci: give up and don't run GLES3 tests on Android.
...
GOOGLE!!! You owe me a WEEK of my time. A WEEK WASTED!
4 years ago
Vladimír Vondruš
eaa95dd332
package/ci: add Android x86 GLES2, GLES3 and Vulkan builds on CircleCI.
...
This apparently can test even the GPU! Let's C.
4 years ago
Vladimír Vondruš
6804dfe8e3
package/ci: add MSVC 2022 builds and MSVC 2019 /permissive- build.
...
Equivalently to what's done in Corrade, so a dynamic, static, clang-cl
and /permissive- build; plus dedicated ES2/ES3 builds.
4 years ago
Vladimír Vondruš
b3f5d87640
package/ci: drop MSVC 2017 ES builds and RT ES2 build.
...
Making room for MSVC 2022. GLES will be tested more than enough on 2019
and 2022. The RT port is rarely (if ever) used and obnoxiously slow, so
ES3 has to suffice.
4 years ago
Vladimír Vondruš
8c65eeb27e
package/ci: SceneTools aren't needed in the Vulkan-specific builds.
4 years ago
Vladimír Vondruš
18b6369c11
package/ci: we don't need SceneTools nor ShaderTools for ES builds.
...
Just the stuff that touches GL functionality in some way. Which
unfortunately means all of DebugTools and SceneGraph for debug
renderers, which then pull in also all of Primitives and all of
MeshTools.
4 years ago
Vladimír Vondruš
ae7524dcbb
package/archlinux: skip benchmarks in Clang PKGBUILD checks.
...
Their output isn't shown anyway and they just take ages to run.
4 years ago
Vladimír Vondruš
01660a7538
package/ci: ah yes, Xcode images rotting away right under my very hands.
...
TBF I have to thank CircleCI for sending me no less than three warning
e-mails about Xcode image deprecation. I just didn't bother
updating until it started actually failing.
4 years ago
Vladimír Vondruš
b7876ed1cf
package/ci: reduce parallelism in the linux static build.
...
Uh oh, do I have my files too large now?
4 years ago
Vladimír Vondruš
6d6676e1cd
package/ci: drop Gitter webhook notifications.
...
It never really worked, was either too verbose or too quiet, and now it
404s on AppVeyor.
4 years ago
Vladimír Vondruš
8088ed8cda
package/ci: GitHub says I can't use git:// anymore.
...
I kinda get the point, to avoid rogue third parties tainting my
ŕepositories with nasty backdoors, yes, but unless *everything* on the
way to the server including DNS queries is encrypted, then it's still
just a half of the solution and any silly corporate firewall can still
prevent me from cloning stuff from github dot com.
4 years ago
Vladimír Vondruš
dae4a35fd9
package/archlinux: let the coverage build go through even with failures.
4 years ago
Vladimír Vondruš
23f2eae702
package/ci: it might be good to *actually use* the job limiters, heh.
...
And here I was wondering why even -j8 isn't low enough. F.
4 years ago
Vladimír Vondruš
031d1ceec3
package/ci: add a GLES3 macOS build.
...
There's a lot of places in the code that assume macOS is never GLES, and
with the increasingly rotten native GL support on Apple platforms people
are switching to ANGLE. So we should make sure this use case works.
4 years ago
Vladimír Vondruš
c07e35e31c
package/ci: adopt MinGW lcov changes from Corrade.
5 years ago
Vladimír Vondruš
754e010794
package/homebrew: make this work on ARM Macs as well.
...
Thank you, Homebrew! FFS!
5 years ago
Vladimír Vondruš
971d24b5fa
package/homebrew: it's time to start documenting Homebrew insanities.
5 years ago
Vladimír Vondruš
1beb2edb64
package/ci: I'm tired of Emscripten builds randomly failing.
...
Let's reduce the parallelism even more. SIGH.
5 years ago
Vladimír Vondruš
7346dc2d01
package/ci: move a comment to where the macOS image is defined.
5 years ago
Vladimír Vondruš
202dff63ae
package/ci: MSYS, the ancient curses seem to have returned.
5 years ago
Vladimír Vondruš
37fc638d12
package/ci: had to upgrade to 10.15 because of MoltenVK.
...
Sigh, that's like those projects that require GCC 13 to build.
5 years ago
Vladimír Vondruš
ef9da0ec96
Shaders: add UBO support to all shaders.
5 years ago
Vladimír Vondruš
73ba2b9d64
Shaders: benchmark all shader variants.
...
Took me a while (several years?) to figure out a way to benchmark this
without basically duplicating the testing effort and without new
variants being too hard to add.
5 years ago
Vladimír Vondruš
666fe17603
package/ci: don't even attempt to run benchmarks on SwiftShader.
...
The damn thing has more bugs than features and doesn't support timer
queries anyway, so there's no point in trying anything.
5 years ago
Vladimír Vondruš
18dcfae51e
package/archlinux: don't run benchmarks in check().
...
Takes too long and is useless, as we don't see the output anyway. Also
deduplicate the env var setters and export them just once at the start.
5 years ago
Vladimír Vondruš
bc0bb075ab
Update Android installation layout to better comply with NDK r19+.
...
And take CMake 3.20 into account.
5 years ago
Vladimír Vondruš
1e7fafdd75
Vk: VK_KHR_create_renderpass depends on KHR_maintenance2 and multiview.
...
The validation layer complained. So enable these two implicitly as well
if we're on 1.0.
And since those are now disabled in certain test runs to test related
code paths, DeviceVkTest needs to be updated to skip affected tests if
this happens.
5 years ago
Vladimír Vondruš
baaa083af8
Vk: implicitly enable VK_KHR_maintenance1.
5 years ago
Vladimír Vondruš
475b502a1d
package/ci: cap Emscripten jobs.
...
On every CI run, one of these gets aborted due to OOM reasons. Didn't
happen before.
5 years ago
Vladimír Vondruš
df6da797ab
package: toggle more ES extensions for higher test coverage.
5 years ago
Vladimír Vondruš
d4ab9f4cb0
Vk: implemented a Mesh wrapper.
...
Together with:
* CommandBuffer::draw()
* Support for indexed and non-indexed meshes
* Support for setting primitive and stride dynamically
I took one shortcut and vkCmdBindVertexBuffers() is currently called
once for each binding. The interface is ready for this, but I'm not yet
100% sure how to test that it actually does batch the buffers, so it's
left at the lazy implementation for now.
5 years ago
Vladimír Vondruš
5e74ec7be3
package: test with VK_KHR_copy_commands2 disabled as well.
...
For a broader test coverage.
5 years ago
Vladimír Vondruš
3d136503d8
Updated copyright year.
5 years ago