Commit graph

468 commits

Author SHA1 Message Date
Tim-Philipp Müller a36ac34d49 docker: fedora: install nasm properly
Add it to install list not debuginfo-install list.

Should make openh264 subproject work in gst-build builds.

Fixes #76

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/301>
2020-06-04 21:55:09 +00:00
Thibault Saunier 3a620ac384 ci: Update fedora image 2020-06-04 12:48:31 -04:00
Thibault Saunier d72551ddc4 gitlab: Update hotdoc to latest release 2020-06-04 12:34:15 -04:00
Thibault Saunier c83578dcd2 Revert "docker: install intel-mediasdk on Fedora image"
This is not available on F30

This reverts commit 9405f9245b.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/299>
2020-06-04 16:01:21 +00:00
Thibault Saunier 3e7e6615b6 gitlab: Ensure devhelp files are installed
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/299>
2020-06-04 16:01:21 +00:00
Thibault Saunier 99830c2bee Revert "docker: fedora: update to Fedora 31"
This is unused yet because it breaks some things

This reverts commit d016fa7453.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/299>
2020-06-04 16:01:21 +00:00
Mathieu Duponchelle 1f07629c3c ci_template: pin hotdoc to pre fundamental documentation 2020-06-04 17:56:00 +02:00
Mathieu Duponchelle 5455320869 ci_template: pin hotdoc to pre gst cache format change 2020-06-04 17:29:49 +02:00
Thibault Saunier 3839f8ab87 gitlab: Enable docs only when building the doc 2020-06-03 22:39:33 -04:00
Jordan Petridis dbe39079ee ci_template: windows: clone the gst-build specified in the manifest
In all our other builds, we are using the clone_manifest_ref script
to fetch the revision of gst-build that we discover in the manifest.
For the windows job this was missed it seems, but didn't cause
any issues till now cause it only affected the gst-build MRs.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/296>
2020-06-02 11:13:09 +03:00
Thibault Saunier 929a611889 ci: Check valgrind for all modules on gst-devtools
Now that we have [`gst-tester`] all module are starting to be tested
with `gst-validate` (core and -base already have validate based unit
tests) so we need to ensure that changes there do not break those
modules.

[gst-tester]: https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/461

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/294>
2020-05-29 22:19:14 +00:00
Tim-Philipp Müller 92b78216d4 ci_template: update cerbero build-deps for gst-validate -> gst-devtools-1.0 rename
https://gitlab.freedesktop.org/gstreamer/gst-devtools/-/merge_requests/199
https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/490

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/295>
2020-05-29 13:05:02 +01:00
Stéphane Cerveau 0928a89125 gst-omx: update vcu-omx-il public release branch
Use release 2020.01 for the xilinx zynq target
from https://github.com/Xilinx/vcu-omx-il

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/293>
2020-05-21 15:40:27 +02:00
Jordan Petridis 822706e473
.gitlab-ci.yml: fix buildah syntax for tagging images 2020-05-14 12:46:54 +03:00
Philippe Normand 61c1680b07 docker/fedora: Remove WPEBackend-FDO
It should now be built as a gst-build subproject. The version in Fedora is not
suitable for CI tests.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/292>
2020-05-14 08:49:47 +00:00
Jordan Petridis 0fb6955c5e
.gitlab-ci.yml: fix typo in buildah images 2020-05-14 11:40:50 +03:00
Matthew Waters 95b35ceaf1 android: update to gradle 5.6.4 without --no-search-upwards
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/291>
2020-05-09 00:04:23 +10:00
Matthew Waters c23bdae050 ci: run CI for the new vulkan android example
HACK: pointing to user repo
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/291>
2020-05-09 00:04:17 +10:00
Jordan Petridis b8fa6db873 docker/windows: add script to build the image with the correct branch and tag
also move the default_branch arg later in the dockerfile to
avoid invalidating previous layers

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/287>
2020-04-27 15:49:32 +00:00
Jordan Petridis 6576f2f7b7 ci_template: try harder to cleanup after build jobs
Let's try extra hard to clean after ourselves, cause gitlab-runner
occasionally decides to reuse existing, polluted, volumes...

so much for clean build environments...

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/288>
2020-04-27 15:39:45 +03:00
Jordan Petridis edac99f441 windows/docker: allow to specify the branch to pull for gst-build
Similar to !232

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/285>
2020-04-27 08:03:54 +00:00
Matthew Waters 38719d0aef docker: remove unused Dockerfiles
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/286>
2020-04-24 14:17:04 +10:00
Jordan Petridis 2f504610b3 ci_tepmlate: disable the msys job again
Its still broken and keeps haning randomly... 😢️

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/284>
2020-04-23 16:59:24 +03:00
Jordan Petridis f546834e6c ci_template: enable the windows x86 and msys native jobs again
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/281>
2020-04-23 07:57:02 +00:00
Jordan Petridis b36d813f5b ci_template: don't use prebuilt binaries for test jobs
Previously we were optimizing for cpu time, so we where building
gst-build once and then exporting that to be used by the test jobs.
However this meant that we where uploading 200mb (previously 600mb)
zipped of artifacts and then re-downloading them for each test job.
This caused big costs in terms of cloud egress since the runners
aren't hosted on the same cloud as the storage/artifacts instance.

Instead we are going to be rebuilding gst-build for each test
job from now, it also doesn't take more time than the network
i/o would of downloading the artifacts, so the impact of rebuilding
shouldn't be noticebly.

We are also using pinned git refs the modules we rebuild from
the manifest, so the binaries should be reproducible for the most
part (minus things like .pyc files).

Close #68

Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/280>
2020-04-22 09:23:18 +03:00
Matthew Waters 0c315875be oss-fuzz: update for autotools removal 2020-04-21 08:47:28 +00:00
Jordan Petridis f3aedd930d build docker image: tag the jobs to use the gstreamer runners
docker builds can be big, and other shared runner have smaller
storage space which we should avoid filling if can be avoided.
The gst tagged runners are a better fit for such builds since
their disk storage is about 1-2 terabytes.
2020-04-20 07:04:16 +00:00
Jordan Petridis dd69902fbf Replace docker-in-docker with buildah
This avoids the need of using privilledged namespaces and dind
as buildah are able to build images unprivilledged.

One thing to note is that buildah inside docker is not a
supported configuration and not tested in upstream podman,
but the possible fallout is still easier to deal with than
dind and requiring privileged runners.
2020-04-20 07:04:16 +00:00
Nicolas Dufresne 95f9b25683 ci: Always allow redirect when using curl
This done by specifying -L parameter. Some of the FDO infrastructure is using
redirect.
2020-04-19 14:30:43 -04:00
Jordan Petridis 815976e4aa
ci_template: avoid exporting the mediafiles into build artifacts
Previously we where accidently exporting the whole repo of
gst-integration-testsuites which includes 350mb of raw media
files and made the artifacts storage explode through the roof
along with the CI bills fd.o had to pay for uploading and
redownloading the artifacts

To deal with this, we clean all the media files from the builddir
and when needed we copy them over from the cache in the docker image,
and then git fetch the repo.

Close #69
2020-04-17 22:46:34 +03:00
Jordan Petridis 7676363645
ci_template: avoid building libnice tests and examples in the subproject
We don't run the libnice testsuite, and when binaries are built
they consume ~45mb of space. This increases the size of the artifacts
we export from the gst-build job for the testsuite and drives up
the storage and bandwith costs when re-downloading the artifacts.

Similary disable the test targets of couple other subprojects as well
2020-04-17 21:59:10 +03:00
Nicolas Dufresne 8cb3a84d97 Don't let marge run post-merge pipeline
This is a small regression from previous change.
2020-04-17 13:48:18 -04:00
Jordan Petridis 26c45d6ca3 clone_manifest_ref: add the ability to fetch instead of cloning the repo 2020-04-17 05:30:01 +00:00
Nicolas Dufresne d692406ab1 ci: Let Marge the merge bot trigger our CI
We have notice that a lot of CI activity is cause by user pushing to their
branch after having created an MR. To reduce our CI foot-print, the CI will
now only be automatically triggered when a reviewer assign the MR to the merge
bot. It will still be possible to run the CI manually but the result of that
CI won't be used by Marge.
2020-04-16 13:04:31 -04:00
Jordan Petridis 2871b293e1
ci_template: reduce the size of exported artifacts in gst-build jobs
This is initial work looking over the deps subprojects and
cleaning up the low hanging fruits. Saves about 90mb compressed.
2020-04-10 21:57:45 +03:00
Jordan Petridis 42830e91f6
windows: udpate the docker image
This is a rebuild of the image to acount for an issue
with a security update during February 2020

https://support.microsoft.com/en-us/help/4542617/you-might-encounter-issues-when-using-windows-server-containers-with-t

This also means we can drop the gstreamer-windows tag
that was added in 32c91139a4
2020-04-07 18:55:56 +03:00
Jordan Petridis ae1b75b51f
windows: switch the image tag to ltsc2019
This should have no functional difference as 1809 is an alias
to ltsc2019 but it makes it more explicit that what branch we
are tracking
2020-04-07 14:45:08 +03:00
Matthew Waters 45d4e8ebc0 Update cerbero and android docker images
Partial revert of 9ed6305e1f

For post-merge of android ndk-21 as part of:
https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/266
https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/463
2020-04-06 21:52:59 +10:00
Matthew Waters 3b273b01e8 Partial revert of 9ed6305e1f
For post-merge of android ndk-21 as part of:
https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/266
https://gitlab.freedesktop.org/gstreamer/cerbero/-/merge_requests/463
2020-04-06 18:38:41 +10:00
Matthew Waters 9ed6305e1f docker/cerbero: checkout proposed ndk-21 cerbero branch 2020-04-06 11:50:21 +10:00
Matthew Waters afab7cc48b docker: update android image container for ndk-21 2020-04-06 11:50:21 +10:00
Nicolas Dufresne e137093def ci: Update to use new external storage for cerbero deps
The goal of this change is to reduce egress on the FDO servers.
2020-04-03 16:31:45 -04:00
Matthew Waters b00c68e69e gitlab: don't upload gst-build artifacts we never use 2020-04-02 23:42:41 +11:00
Matthew Waters 18f285f15c gitlab: compress deps using xz on master
Results in a better compression for artifacts that are going to be used
multiple times.
2020-03-31 13:30:22 +11:00
Nirbheek Chauhan 6c379762c2 gst-build: Disabling uploading of artifacts for now
It seems to be timing out with high frequency only on Windows runners.

```
Version:      12.8.0
00:47
Git revision: 1b659122
Git branch:   12-8-stable
GO version:   go1.13.7
Built:        2020-02-22T03:03:07+0000
OS/Arch:      windows/amd64
Uploading artifacts...
gst-build/build/meson-logs/: found 2 matching files
WARNING: Failed to load system CertPool: crypto/x509: system root pool is not available on Windows
ERROR: Job failed (system failure): aborted: <nil>
```

See: https://gitlab.freedesktop.org/gstreamer/gst-ci/-/merge_requests/261
2020-03-28 16:34:03 +05:30
Seungha Yang e0544b256f ci_template: Increase timeout of -good valgrind
splitmux unit test includes encoding task which might require more time
to finish test
2020-03-28 18:15:47 +09:00
Nirbheek Chauhan a5686ae478 gst-build: Disable vs2017 x86 job too
This might be related to the same issue described in the previous
commit: Till we can update the container image to the Feb 11 security
update, x86 executables and in general the container image will behave
badly because of:

https://support.microsoft.com/en-us/help/4542617/you-might-encounter-issues-when-using-windows-server-containers-with-t

vs2017 x86 has been failing with a runner system failure while
uploading artifacts / submitting job status:

```
Uploading artifacts...
gst-build/build/meson-logs/: found 2 matching files
WARNING: Failed to load system CertPool: crypto/x509: system root pool is not available on Windows
ERROR: Job failed (system failure): aborted: <nil>
```

https://gitlab.freedesktop.org/slomo/gst-plugins-good/-/jobs/2084184

Disable it for now.
2020-03-27 20:54:28 +05:30
Nirbheek Chauhan 32c91139a4 gst-build: Only use gstreamer runners for windows jobs
Till we can update the container image to the Feb 11 security update,
all x86 executables will crash because of:

https://support.microsoft.com/en-us/help/4542617/you-might-encounter-issues-when-using-windows-server-containers-with-t

So restrict to the gstreamer runners that are running an older Windows
version. For example:

```
subprojects\gstreamer\gst\parse\meson.build:10:0: ERROR: Command
  "c:\python38\python.exe
  C:\builds\thaytan\gst-plugins-good\gst-build\subprojects\gstreamer\gst\parse\get_flex_version.py
  C:\builds\thaytan\gst-plugins-good\gst-build\subprojects\win-flex-bison-binaries\win_flex.exe"
  failed with status 1.
```

https://gitlab.freedesktop.org/thaytan/gst-plugins-good/-/jobs/2066909

```
meson.build:1:0: ERROR: Executables created by c compiler cl are not runnable.
```

https://gitlab.freedesktop.org/nirbheek/gst-ci/-/jobs/2070394
2020-03-26 19:57:18 +05:30
Nirbheek Chauhan 3d06b08d49 gst-build: Always include meson-logs in the artifacts
Can't figure out build errors without this.
2020-03-26 18:56:38 +05:30
Nicolas Dufresne 8ba527416c ci: Re-enable example builds for MR
As we should still build the examples to make sure they are not broken.

Reported-by: Matthew Waters <matthew@centricular.com>
2020-03-25 17:22:09 -04:00